advertisement
w WM8958
Multi-Channel Audio Hub CODEC for Smartphones
DESCRIPTION FEATURES
The WM8958 [1] is a highly integrated ultra-low power hi-fi
CODEC designed for smartphones and other portable devices rich in multimedia features.
An integrated stereo class D/AB speaker driver and class W headphone driver minimize power consumption during audio playback.
The device requires only two voltage supplies, with all other internal supply rails generated from integrated LDOs.
Stereo full duplex asynchronous sample rate conversion and multi-channel digital mixing combined with powerful analogue mixing allow the device to support a huge range of different architectures and use cases.
A multiband compressor and programmable parametric EQ provide volume maximisation and speaker compensation in the digital playback paths. The dynamic range controller can be used in record or playback paths for maintaining a constant signal level, maximizing loudness and protecting speakers against overloading and clipping.
A smart digital microphone interface provides power regulation, a low jitter clock output and decimation filters for up to four digital microphones. Microphone activity detection with interrupt is available. Impedance sensing and measurement is provided for external accessory / push-button detection.
24-bit 4-channel hi-fi DAC and 2-channel hi-fi ADC
100dB SNR during DAC playback (‘A’ weighted)
Smart MIC interface
- Power, clocking and data input for up to four digital MICs
- High performance analogue MIC interface
- MIC activity detect & interrupt allows processor to sleep
- Impedance sensing for accessory / push-button detection
2W stereo (2 x 2W) class D/AB speaker driver
Capless Class W headphone drivers
- Integrated charge pump
- 5.3mW total power for DAC playback to headphones
4 Line outputs (single-ended or differential)
BTL Earpiece driver
Digital audio interfaces for multi-processor architecture
- Asynchronous stereo duplex sample rate conversion
- Powerful mixing and digital loopback functions
ReTune TM Mobile 5-band, 6-channel parametric EQ
Multiband compressor and dynamic range controller
Dual FLL provides all necessary clocks
- Self-clocking modes allow processor to sleep
- All standard sample rates from 8kHz to 96kHz
Active noise reduction circuits
- DC offset correction removes pops and clicks
- Ground loop noise cancellation
Integrated LDO regulators
72-ball W-CSP package (4.516 x 4.258 x 0.698mm)
APPLICATIONS Fully differential internal architecture and on-chip RF noise filters ensure a very high degree of noise immunity. Active ground loop noise rejection and DC offset correction help prevent pop noise and suppress ground noise on the headphone outputs.
Smartphones and music phones navigation
Tablets
eBooks
Portable Media Players
WOLFSON MICROELECTRONICS plc
[1] This product is protected by Patents US 7,622,984, US 7,626,445,US 7,765,019 and GB 2,432,765
Pre-Production, August 2012, Rev 3.4
WM8958
Pre-Production
TABLE OF CONTENTS
DESCRIPTION ....................................................................................................... 1
FEATURES ............................................................................................................ 1
APPLICATIONS ..................................................................................................... 1
TABLE OF CONTENTS ......................................................................................... 2
BLOCK DIAGRAM ................................................................................................ 7
PIN CONFIGURATION .......................................................................................... 8
ORDERING INFORMATION .................................................................................. 8
PIN DESCRIPTION ................................................................................................ 9
ABSOLUTE MAXIMUM RATINGS ...................................................................... 12
RECOMMENDED OPERATING CONDITIONS ................................................... 13
THERMAL PERFORMANCE ............................................................................... 14
ELECTRICAL CHARACTERISTICS ................................................................... 15
INPUT SIGNAL LEVEL .................................................................................................. 15
INPUT PIN RESISTANCE .............................................................................................. 16
PROGRAMMABLE GAINS ............................................................................................. 18
OUTPUT DRIVER CHARACTERISTICS ....................................................................... 19
ADC INPUT PATH PERFORMANCE ............................................................................. 20
DAC OUTPUT PATH PERFORMANCE ......................................................................... 21
BYPASS PATH PERFORMANCE .................................................................................. 24
MULTI-PATH CROSSTALK ........................................................................................... 26
DIGITAL INPUT / OUTPUT ............................................................................................ 28
DIGITAL FILTER CHARACTERISTICS ......................................................................... 28
MICROPHONE BIAS CHARACTERISTICS ................................................................... 29
MISCELLANEOUS CHARACTERISTICS ...................................................................... 30
TERMINOLOGY ............................................................................................................. 31
TYPICAL PERFORMANCE ................................................................................. 32
TYPICAL POWER CONSUMPTION .............................................................................. 32
TYPICAL SIGNAL LATENCY ......................................................................................... 33
SPEAKER DRIVER PERFORMANCE ........................................................................... 34
SIGNAL TIMING REQUIREMENTS .................................................................... 35
SYSTEM CLOCKS & FREQUENCY LOCKED LOOP (FLL) .......................................... 35
AUDIO INTERFACE TIMING ......................................................................................... 36
DIGITAL MICROPHONE (DMIC) INTERFACE TIMING .................................................................................................. 36
DIGITAL AUDIO INTERFACE - MASTER MODE ........................................................................................................... 37
DIGITAL AUDIO INTERFACE - SLAVE MODE ............................................................................................................... 38
DIGITAL AUDIO INTERFACE - TDM MODE .................................................................................................................. 39
CONTROL INTERFACE TIMING ................................................................................... 40
DEVICE DESCRIPTION ...................................................................................... 41
INTRODUCTION ............................................................................................................ 41
ANALOGUE INPUT SIGNAL PATH ............................................................................... 43
MICROPHONE INPUTS .................................................................................................................................................. 44
MICROPHONE BIAS CONTROL .................................................................................................................................... 44
MICROPHONE ACCESSORY DETECT ......................................................................................................................... 46
LINE AND VOICE CODEC INPUTS ................................................................................................................................ 46
INPUT PGA ENABLE ...................................................................................................................................................... 47
INPUT PGA CONFIGURATION ...................................................................................................................................... 48
INPUT PGA VOLUME CONTROL ................................................................................................................................... 49
INPUT MIXER ENABLE ................................................................................................................................................... 52
INPUT MIXER CONFIGURATION AND VOLUME CONTROL ....................................................................................... 52
DIGITAL MICROPHONE INTERFACE .......................................................................... 56
DIGITAL PULL-UP AND PULL-DOWN ............................................................................................................................ 59
PP, August 2012, Rev 3.4
w
2
Pre-Production
WM8958
ANALOGUE TO DIGITAL CONVERTER (ADC) ............................................................ 59
ADC CLOCKING CONTROL ........................................................................................................................................... 60
DIGITAL CORE ARCHITECTURE ................................................................................. 61
DIGITAL MIXING ............................................................................................................ 63
AUDIO INTERFACE 1 (AIF1) OUTPUT MIXING ............................................................................................................. 64
DIGITAL SIDETONE MIXING .......................................................................................................................................... 65
DIGITAL SIDETONE VOLUME AND FILTER CONTROL ............................................................................................... 65
DAC OUTPUT DIGITAL MIXING ..................................................................................................................................... 67
AUDIO INTERFACE 2 (AIF2) DIGITAL MIXING ............................................................................................................. 68
ULTRASONIC (4FS) AIF OUTPUT MODE ...................................................................................................................... 69
MULTIBAND COMPRESSOR (MBC) ............................................................................ 70
RMS LIMITER .................................................................................................................................................................. 71
MBC CLOCKING CONTROL ........................................................................................................................................... 71
MBC CONTROL SEQUENCES ....................................................................................................................................... 72
DYNAMIC RANGE CONTROL (DRC) ........................................................................... 75
DRC COMPRESSION / EXPANSION / LIMITING ........................................................................................................... 76
GAIN LIMITS .................................................................................................................................................................... 78
DYNAMIC CHARACTERISTICS ..................................................................................................................................... 79
ANTI-CLIP CONTROL ..................................................................................................................................................... 79
QUICK RELEASE CONTROL ......................................................................................................................................... 79
SIGNAL ACTIVITY DETECT ........................................................................................................................................... 79
DRC REGISTER CONTROLS ......................................................................................................................................... 80
RETUNE TM MOBILE PARAMETRIC EQUALIZER (EQ) ................................................ 89
DEFAULT MODE (5-BAND PARAMETRIC EQ) ............................................................................................................. 90
RETUNE TM MOBILE MODE............................................................................................................................................. 92
EQ FILTER CHARACTERISTICS ................................................................................................................................... 93
3D STEREO EXPANSION ............................................................................................. 94
DIGITAL VOLUME AND FILTER CONTROL ................................................................. 95
AIF1 - OUTPUT PATH VOLUME CONTROL .................................................................................................................. 95
AIF1 - OUTPUT PATH HIGH PASS FILTER ................................................................................................................... 97
AIF1 - INPUT PATH VOLUME CONTROL ...................................................................................................................... 99
AIF1 - INPUT PATH SOFT MUTE CONTROL .............................................................................................................. 102
AIF1 - INPUT PATH NOISE GATE CONTROL ............................................................................................................. 103
AIF1 - INPUT PATH MONO MIX CONTROL ................................................................................................................. 104
AIF2 - OUTPUT PATH VOLUME CONTROL ................................................................................................................ 104
AIF2 - OUTPUT PATH HIGH PASS FILTER ................................................................................................................. 105
AIF2 - INPUT PATH VOLUME CONTROL .................................................................................................................... 106
AIF2 - INPUT PATH SOFT MUTE CONTROL .............................................................................................................. 106
AIF2 - INPUT PATH NOISE GATE CONTROL ............................................................................................................. 107
AIF2 - INPUT PATH MONO MIX CONTROL ................................................................................................................. 108
DIGITAL TO ANALOGUE CONVERTER (DAC) .......................................................... 109
DAC CLOCKING CONTROL ......................................................................................................................................... 109
DAC DIGITAL VOLUME ................................................................................................................................................ 111
DAC SOFT MUTE AND SOFT UN-MUTE ..................................................................................................................... 114
ANALOGUE OUTPUT SIGNAL PATH ......................................................................... 116
OUTPUT SIGNAL PATHS ENABLE .............................................................................................................................. 117
HEADPHONE SIGNAL PATHS ENABLE ...................................................................................................................... 119
OUTPUT MIXER CONTROL ......................................................................................................................................... 121
SPEAKER MIXER CONTROL ....................................................................................................................................... 125
OUTPUT SIGNAL PATH VOLUME CONTROL ............................................................................................................. 128
SPEAKER BOOST MIXER ............................................................................................................................................ 133
EARPIECE DRIVER MIXER .......................................................................................................................................... 133
LINE OUTPUT MIXERS ................................................................................................................................................ 134 w PP, August 2012, Rev 3.4
3
WM8958
Pre-Production
CHARGE PUMP ........................................................................................................... 138
DC SERVO ................................................................................................................... 140
DC SERVO ENABLE AND START-UP ......................................................................................................................... 140
DC SERVO ACTIVE MODES ........................................................................................................................................ 142
GPIO / INTERRUPT OUTPUTS FROM DC SERVO ..................................................................................................... 143
ANALOGUE OUTPUTS ............................................................................................... 144
SPEAKER OUTPUT CONFIGURATIONS ..................................................................................................................... 144
HEADPHONE OUTPUT CONFIGURATIONS ............................................................................................................... 147
EARPIECE DRIVER OUTPUT CONFIGURATIONS ..................................................................................................... 148
LINE OUTPUT CONFIGURATIONS .............................................................................................................................. 148
EXTERNAL ACCESSORY DETECTION ..................................................................... 151
ACCESSORY DETECTION WITH LOW FREQUENCY SYSCLK ................................................................................. 155
GENERAL PURPOSE INPUT/OUTPUT ...................................................................... 156
GPIO CONTROL ........................................................................................................................................................... 156
GPIO FUNCTION SELECT ........................................................................................................................................... 157
BUTTON DETECT (GPIO INPUT) ................................................................................................................................. 159
LOGIC ‘1’ AND LOGIC ‘0’ OUTPUT (GPIO OUTPUT) .................................................................................................. 159
INTERRUPT (IRQ) STATUS OUTPUT .......................................................................................................................... 159
OVER-TEMPERATURE DETECTION ........................................................................................................................... 159
MICROPHONE ACCESSORY STATUS DETECTION .................................................................................................. 160
FREQUENCY LOCKED LOOP (FLL) LOCK STATUS OUTPUT .................................................................................. 160
SAMPLE RATE CONVERTER (SRC) LOCK STATUS OUTPUT ................................................................................. 160
DYNAMIC RANGE CONTROL (DRC) SIGNAL ACTIVITY DETECTION ...................................................................... 161
CONTROL WRITE SEQUENCER STATUS DETECTION ............................................................................................ 163
DIGITAL CORE FIFO ERROR STATUS DETECTION.................................................................................................. 163
OPCLK CLOCK OUTPUT .............................................................................................................................................. 163
FLL CLOCK OUTPUT .................................................................................................................................................... 164
INTERRUPTS .............................................................................................................. 165
DIGITAL AUDIO INTERFACE ...................................................................................... 170
MASTER AND SLAVE MODE OPERATION ................................................................................................................. 171
OPERATION WITH TDM ............................................................................................................................................... 171
AUDIO DATA FORMATS (NORMAL MODE) ................................................................................................................ 172
AUDIO DATA FORMATS (TDM MODE) ....................................................................................................................... 175
DIGITAL AUDIO INTERFACE CONTROL ................................................................... 178
AIF1 - MASTER / SLAVE AND TRI-STATE CONTROL ................................................................................................ 178
AIF1 - SIGNAL PATH ENABLE ..................................................................................................................................... 179
AIF1 - BCLK AND LRCLK CONTROL ........................................................................................................................... 179
AIF1 - DIGITAL AUDIO DATA CONTROL ..................................................................................................................... 182
AIF1 - MONO MODE ..................................................................................................................................................... 183
AIF1 - COMPANDING ................................................................................................................................................... 184
AIF1 - LOOPBACK ........................................................................................................................................................ 186
AIF1 - DIGITAL PULL-UP AND PULL-DOWN ............................................................................................................... 186
AIF2 - MASTER / SLAVE AND TRI-STATE CONTROL ................................................................................................ 187
AIF2 - SIGNAL PATH ENABLE ..................................................................................................................................... 188
AIF2 - BCLK AND LRCLK CONTROL ........................................................................................................................... 188
AIF2 - DIGITAL AUDIO DATA CONTROL ..................................................................................................................... 191
AIF2 - MONO MODE ..................................................................................................................................................... 193
AIF2 - COMPANDING ................................................................................................................................................... 193
AIF2 - LOOPBACK ........................................................................................................................................................ 193
AIF2 - DIGITAL PULL-UP AND PULL-DOWN ............................................................................................................... 194
AIF3 - SIGNAL PATH CONFIGURATION AND TRI-STATE CONTROL ....................................................................... 195
AIF3 - BCLK AND LRCLK CONTROL ........................................................................................................................... 197
AIF3 - DIGITAL AUDIO DATA CONTROL ..................................................................................................................... 198
AIF3 - COMPANDING ................................................................................................................................................... 199
AIF3 - LOOPBACK ........................................................................................................................................................ 199 w PP, August 2012, Rev 3.4
4
Pre-Production
WM8958
CLOCKING AND SAMPLE RATES .............................................................................. 200
AIF1CLK ENABLE ......................................................................................................................................................... 201
AIF1 CLOCKING CONFIGURATION ............................................................................................................................ 202
AIF2CLK ENABLE ......................................................................................................................................................... 204
AIF2 CLOCKING CONFIGURATION ............................................................................................................................ 204
MISCELLANEOUS CLOCK CONTROLS ...................................................................................................................... 206
BCLK AND LRCLK CONTROL ...................................................................................................................................... 209
CONTROL INTERFACE CLOCKING ............................................................................................................................ 211
FREQUENCY LOCKED LOOP (FLL) ............................................................................................................................ 211
FREE-RUNNING FLL CLOCK ....................................................................................................................................... 217
GPIO OUTPUTS FROM FLL ......................................................................................................................................... 218
EXAMPLE FLL CALCULATION ..................................................................................................................................... 219
EXAMPLE FLL SETTINGS ............................................................................................................................................ 220
SAMPLE RATE CONVERSION ................................................................................... 221
SAMPLE RATE CONVERTER 1 (SRC1) ...................................................................................................................... 221
SAMPLE RATE CONVERTER 2 (SRC2) ...................................................................................................................... 221
SAMPLE RATE CONVERTER RESTRICTIONS .......................................................................................................... 221
SAMPLE RATE CONVERTER CONFIGURATION ERROR INDICATION ................................................................... 222
CONTROL INTERFACE ............................................................................................... 224
CONTROL WRITE SEQUENCER ................................................................................ 227
INITIATING A SEQUENCE ............................................................................................................................................ 227
PROGRAMMING A SEQUENCE .................................................................................................................................. 228
DEFAULT SEQUENCES ............................................................................................................................................... 230
POP SUPPRESSION CONTROL ................................................................................ 237
DISABLED LINE OUTPUT CONTROL .......................................................................................................................... 237
LINE OUTPUT DISCHARGE CONTROL ...................................................................................................................... 238
VMID REFERENCE DISCHARGE CONTROL .............................................................................................................. 238
INPUT VMID CLAMPS .................................................................................................................................................. 238
LDO REGULATORS .................................................................................................... 239
REFERENCE VOLTAGES AND MASTER BIAS ......................................................... 241
POWER MANAGEMENT ............................................................................................. 243
THERMAL SHUTDOWN .............................................................................................. 248
POWER ON RESET..................................................................................................... 249
QUICK START-UP AND SHUTDOWN ........................................................................ 251
SOFTWARE RESET AND DEVICE ID ......................................................................... 252
REGISTER MAP ................................................................................................ 253
REGISTER BITS BY ADDRESS .................................................................................. 265
APPLICATIONS INFORMATION ...................................................................... 360
RECOMMENDED EXTERNAL COMPONENTS .......................................................... 360
AUDIO INPUT PATHS ................................................................................................................................................... 360
HEADPHONE OUTPUT PATH ...................................................................................................................................... 361
EARPIECE DRIVER OUTPUT PATH ............................................................................................................................ 362
LINE OUTPUT PATHS .................................................................................................................................................. 362
POWER SUPPLY DECOUPLING ................................................................................................................................. 363
CHARGE PUMP COMPONENTS ................................................................................................................................. 364
MICROPHONE BIAS CIRCUIT ..................................................................................................................................... 364
EXTERNAL ACCESSORY DETECTION COMPONENTS ............................................................................................ 366
CLASS D SPEAKER CONNECTIONS .......................................................................................................................... 367
RECOMMENDED EXTERNAL COMPONENTS DIAGRAM ......................................................................................... 368
DIGITAL AUDIO INTERFACE CLOCKING CONFIGURATIONS ................................. 370
PCB LAYOUT CONSIDERATIONS ............................................................................. 373
CLASS D LOUDSPEAKER CONNECTION .................................................................................................................. 373 w PP, August 2012, Rev 3.4
5
WM8958
Pre-Production
PACKAGE DIMENSIONS .................................................................................. 374
IMPORTANT NOTICE ....................................................................................... 375
ADDRESS: ................................................................................................................... 375
REVISION HISTORY ......................................................................................... 376
w PP, August 2012, Rev 3.4
6
Pre-Production
BLOCK DIAGRAM w
DMICDAT2
DMICDAT1
WM8958
V
VS, NG
VS, NG
V
CPVDD
CPGND
LINEOUTFB
HPOUT1FB
SPKMODE
ADDR
SDA
SCLK
BCLK2
BCLK1
LRCLK2
LRCLK1
GPIO11/BCLK3
GPIO10/LRCLK3
GPIO8/DACDAT3
GPIO9/ADCDAT3
ADCDAT2
GPIO6/ADCLRCLK2
DACDAT2
LRCLK2
BCLK2
GPIO1/ADCLRCLK1
ADCDAT1
DACDAT1
LRCLK1
BCLK1
MCLK1
MCLK2
AVDD1
VMIDC
AGND
REFGND
LDO2ENA
LDO1ENA
VREFC
PP, August 2012, Rev 3.4
7
WM8958
PIN CONFIGURATION
Pre-Production
ORDERING INFORMATION
ORDER CODE
WM8958ECS/R
TEMPERATURE RANGE PACKAGE MOISTURE
SENSITIVITY LEVEL
-40 C to +85C 72-ball
(Pb-free, Tape and reel)
PEAK SOLDERING
TEMPERATURE
MSL1 260 C
Note:
Reel quantity = 5000 w PP, August 2012, Rev 3.4
8
Pre-Production
WM8958
PIN DESCRIPTION
A description of each pin on the WM8958 is provided below.
Note that a table detailing the associated power domain for every input and output pin is provided on the following page.
Note that, where multiple pins share a common name, these pins should be tied together on the PCB.
PIN NO NAME TYPE DESCRIPTION
Audio interface 1 ADC digital audio data
Audio interface 2 ADC digital audio data
2-wire (I2C) address select
D7, E6 AGND Supply Analogue ground (Return path for AVDD1, AVDD2 and LDO1VDD)
Analogue core supply / LDO1 Output
Output
Bandgap reference, analogue class D and FLL supply
F2 BCLK1 Audio interface 1 bit clock
G3 BCLK2 Audio interface 2 bit clock
Charge pump fly-back capacitor pin
Charge pump fly-back capacitor pin
Charge pump ground (Return path for CPVDD)
Charge pump supply
Charge pump negative supply decoupling pin (HPOUT1L, HPOUT1R)
Charge pump positive supply decoupling pin (HPOUT1L, HPOUT1R)
Audio interface 1 DAC digital audio data
Audio interface 2 DAC digital audio data
Digital buffer (I/O) supply (core functions and Audio Interface 1)
Digital buffer (I/O) supply (for Audio Interface 2)
Digital buffer (I/O) supply (for Audio Interface 3)
Digital core supply / LDO2 output
Output
Digital ground (Return path for DCVDD, DBVDD1, DBVDD2, DBVDD3)
Digital MIC clock output
H1 GPIO1/ General Purpose pin GPIO 1 /
ADCLRCLK1 Audio interface 1 ADC left / right clock
F5 GPIO10/ General Purpose pin GPIO 10 /
LRCLK3 Audio interface 3 left / right clock
E5 GPIO11/ General Purpose pin GPIO 11 /
BCLK3 Audio interface 3 bit clock
H3 GPIO6/ General Purpose pin GPIO 6 /
ADCLRCLK2 Audio interface 2 ADC left / right clock
G4 GPIO8/ General Purpose pin GPIO 8 /
DACDAT3 Audio interface 3 DAC digital audio data
H4 GPIO9/ General Purpose pin GPIO 9 /
ADCDAT3 Audio interface 3 ADC digital audio data
HPOUT1L and HPOUT1R ground loop noise rejection feedback
Left headphone output
Right headphone output
Earpiece speaker inverted output
Earpiece speaker non-inverted output
Left channel single-ended MIC input /
Left channel negative differential MIC input
Left channel line input /
Left channel positive differential MIC input w PP, August 2012, Rev 3.4
9
WM8958
Pre-Production
PIN NO NAME
DMICDAT1
DMICDAT2
TYPE
Digital Input
Digital Input
DESCRIPTION
Right channel single-ended MIC input /
Right channel negative differential MIC input
Right channel line input /
Right channel positive differential MIC input
Left channel line input /
Left channel negative differential MIC input /
Digital MIC data input 1
Left channel line input /
Left channel positive differential MIC input /
Mono differential negative input (RXVOICE -)
Right channel line input /
Right channel negative differential MIC input /
Digital MIC data input 2
Left channel line input /
Left channel positive differential MIC input /
Mono differential positive input (RXVOICE +)
Enable pin for LDO1
Supply for LDO1
Enable pin for LDO2
Negative mono line output / Positive left or right line output
Positive mono line output / Positive left line output
Negative mono line output / Positive left or right line output
Positive mono line output / Positive left line output
Line output ground loop noise rejection feedback
D4 LRCLK1 Audio interface 1 left / right clock
H2 LRCLK2 Audio interface 2 left / right clock
Master clock 1
Master clock 2
Microphone bias 1
Microphone bias 2
Microphone & accessory sense input
Analogue ground
Control interface clock input
G2 SDA Control interface data input and output / acknowledge output
Ground for speaker driver (Return path for SPKVDD1)
Ground for speaker driver (Return path for SPKVDD2)
Mono / Stereo speaker mode select
Left speaker negative output
Left speaker positive output
Right speaker negative output
Right speaker positive output
Supply for speaker driver 1 (Left channel)
Supply for speaker driver 2 (Right channel)
Midrail voltage decoupling capacitor
Bandgap reference decoupling capacitor w PP, August 2012, Rev 3.4
10
Pre-Production
The following table identifies the power domain and ground reference associated with each of the input / output pins.
PIN NO NAME POWER DOMAIN GROUND DOMAIN
WM8958
CPGND
CPGND
MICBIAS1 (DMICDAT1)
MICBIAS1 (DMICDAT2)
AGND
AGND (IN2RN) or
DGND (DMICDAT2) w PP, August 2012, Rev 3.4
11
WM8958
Pre-Production
ABSOLUTE MAXIMUM RATINGS
Absolute Maximum Ratings are stress ratings only. Permanent damage to the device may be caused by continuously operating at or beyond these limits. Device functional operating limits and guaranteed performance specifications are given under Electrical Characteristics at the test conditions specified.
ESD Sensitive Device. This device is manufactured on a CMOS process. It is therefore generically susceptible to damage from excessive static voltages. Proper ESD precautions must be taken during handling and storage of this device.
Wolfson tests its package types according to IPC/JEDEC J-STD-020B for Moisture Sensitivity to determine acceptable storage conditions prior to surface mount assembly. These levels are:
MSL1 = unlimited floor life at <30 C / 85% Relative Humidity. Not normally stored in moisture barrier bag.
MSL2 = out of bag storage for 1 year at <30 C / 60% Relative Humidity. Supplied in moisture barrier bag.
MSL3 = out of bag storage for 168 hours at <30 C / 60% Relative Humidity. Supplied in moisture barrier bag.
The Moisture Sensitivity Level for each package type is specified in Ordering Information.
Supply voltages (AVDD1, DBVDD2, DBVDD3)
Supply voltages (AVDD2, DCVDD, DBVDD1)
Supply voltages (CPVDD)
Supply voltages (SPKVDD1, SPKVDD2, LDO1VDD)
Voltage range digital inputs (DBVDD1 domain)
Voltage range digital inputs (DBVDD2 domain)
Voltage range digital inputs (DBVDD3 domain)
Voltage range digital inputs (DMICDATn)
Voltage range analogue inputs (AVDD1 domain)
Voltage range analogue inputs (MICDET, LINEOUTFB)
Voltage range analogue inputs (HPOUT1FB)
Ground (DGND, CPGND, SPKGND1, SPKGND2, REFGND, HP2GND)
Operating temperature range, T
A
Junction temperature, T
JMAX
Storage temperature after soldering
-0.3V +4.5V
-0.3V +2.5V
-0.3V +2.2V
-0.3V +7.0V
AGND - 0.3V
AGND - 0.3V
DBVDD1 + 0.3V
DBVDD2 + 0.3V
AGND - 0.3V
AGND - 0.3V
AGND - 0.3V
AGND - 0.3V
AGND - 0.3V
DBVDD3 + 0.3V
AVDD1 + 0.3V
AVDD1 + 0.3V
AVDD1 + 0.3V
AGND + 0.3V
AGND - 0.3V AGND + 0.3V
-40ºC +85ºC
-40ºC +150ºC
-65ºC +150ºC w PP, August 2012, Rev 3.4
12
Pre-Production
RECOMMENDED OPERATING CONDITIONS
WM8958
Digital supply range (Core)
See notes 7, 8
Digital supply range (I/O)
Digital supply range (I/O)
Analogue supply 1 range
See notes 3, 4, 5, 6
Analogue supply 2 range
Charge Pump supply range
Speaker supply range
LDO1 supply range
Ground
DBVDD2, DBVDD3
AVDD1
1.62 1.8 3.6 V
2.4 3.0 3.3 V
LDO1VDD 2.7 5.0 5.5 V
DGND, AGND, CPGND,
SPKGND1, SPKGND2,
REFGND, HP2GND
0 V
All supplies 1 s Power supply rise time
See notes 9, 10, 11
Operating temperature range T
A
-40
Notes:
1. Analogue, digital and speaker grounds must always be within 0.3V of AGND.
2. There is no power sequencing requirement; the supplies may be enabled in any order.
3. AVDD1 must be less than or equal to SPKVDD1 and SPKVDD2.
4. An internal LDO (powered by LDO1VDD) can be used to provide the AVDD1 supply.
5. When AVDD1 is supplied externally (not from LDO1), the LDO1VDD voltage must be greater than or equal to AVDD1.
6. The WM8958 can operate with AVDD1 tied to 0V; power consumption may be reduced, but the analogue audio functions will not be supported.
7. An internal LDO (powered by DBVDD1) can be used to provide the DCVDD supply.
8. When DCVDD is supplied externally (not from LDO2), the DBVDD1 voltage must be greater than or equal to DCVDD.
9. DCVDD and AVDD1 minimum rise times do not apply when these domains are powered using the internal LDOs.
10. The specified minimum power supply rise times assume a minimum decoupling capacitance of 100nF per pin.
However, Wolfson strongly advises that the recommended decoupling capacitors are present on the PCB and that appropriate layout guidelines are observed (see “Applications Information” section).
11. The specified minimum power supply rise times also assume a maximum PCB inductance of 10nH between decoupling capacitor and pin. w PP, August 2012, Rev 3.4
13
WM8958
Pre-Production
THERMAL PERFORMANCE
Thermal analysis should be performed in the intended application to prevent the WM8958 from exceeding maximum junction temperature. Several contributing factors affect thermal performance most notably the physical properties of the mechanical enclosure, location of the device on the PCB in relation to surrounding components and the number of PCB layers. Connecting the GND balls through thermal vias and into a large ground plane will aid heat extraction.
Three main heat transfer paths exist to surrounding air as illustrated below in Figure 1:
- Package top to air (radiation).
- Package bottom to PCB (radiation).
- Package balls to PCB (conduction).
Figure 1 Heat Transfer Paths
The temperature rise T
R
is given by T
R
= P
D
* Ө
JA
- P
D
is the power dissipated in the device.
- Ө
JA
is the thermal resistance from the junction of the die to the ambient temperature and is therefore a measure of heat transfer from the die to surrounding air. Ө
JA
is determined with reference to JEDEC standard JESD51-9.
The junction temperature T
J
is given by T
J
= T
A
+T
R
, where T
A
is the ambient temperature.
Operating temperature range
Operating junction temperature
Thermal Resistance
T
A
-40
T
J
-40
Ө
JA
Note:
Junction temperature is a function of ambient temperature and of the device operating conditions. The ambient temperature limits and junction temperature limits must both be observed. w PP, August 2012, Rev 3.4
14
Pre-Production
WM8958
ELECTRICAL CHARACTERISTICS
INPUT SIGNAL LEVEL
Test Conditions
AVDD1 = 3.0V.
With the exception of the condition(s) noted above, the following electrical characteristics are valid across the full range of recommended operating conditions.
A1 Full-Scale PGA Input
Signal Level
See notes 1, 2, 3 and 4
Single-ended PGA input 1.0 Vrms
0 dBV
Differential PGA input 1.0 Vrms
0 dBV
A2 Full-Scale Line Input
Signal Level
See notes 1, 2, 3 and 4
Single-ended Line input to
MIXINL/R, SPKMIXL/R or
MIXOUTL/R mixers
1.0
0
Vrms dBV
Differential mono line input on VRXP/VRXN to
RXVOICE or Direct Voice paths to speaker outputs or earpiece output
1.0
0
Notes:
1. The full-scale input signal level changes in proportion with AVDD1. It is calculated as AVDD1/3.0.
2. When mixing line inputs, input PGA outputs and DAC outputs the total signal must not exceed 1.0Vrms (0dBV).
3. A 1.0Vrms differential signal equates to 0.5Vrms/-6dBV per input.
4. A sinusoidal input signal is assumed.
Vrms dBV w PP, August 2012, Rev 3.4
15
WM8958
Pre-Production
INPUT PIN RESISTANCE
Test Conditions
T
A
= +25 o C.
With the exception of the condition(s) noted above, the following electrical characteristics are valid across the full range of recommended operating conditions.
B1
B2
B3
PGA Input Resistance
Differential Mode
See note 5
See “Applications
Information” for details of
Input resistance at all
PGA Gain settings.
PGA Input Resistance
Single-Ended Mode
See note 5
See “Applications
Information” for details of
Input resistance at all
PGA Gain settings.
Line Input Resistance
See note 5
Gain = -16.5dB
(INnx_VOL=00h)
Gain = 0dB
(INnx_VOL=0Bh)
Gain = +30dB
(INnx_VOL=1Fh)
Gain = -16.5dB
(INnx_VOL=00h)
Gain = 0dB
(INnx_VOL=0Bh)
Gain = +30dB
(INnx_VOL=1Fh)
IN1LP to MIXINL, or
IN1RP to MIXINR
Gain = -12dB
(IN1xP_MIXINx_VOL=001)
IN1LP to MIXINL, or
IN1RP to MIXINR
Gain = 0dB
(IN1xP_MIXINx_VOL=101)
IN1LP to MIXINL, or
IN1RP to MIXINR
Gain = +6dB
(IN1xP_MIXINx_VOL=111)
IN1LP to MIXINL, or
IN1RP to MIXINR
Gain = +15dB
(IN1xP_MIXINx_VOL=111,
IN1xP_MIXINx_BOOST=1)
IN1LP to SPKMIXL, or
IN1RP to SPKMIXR
(SPKATTN = -12dB)
IN1LP to SPKMIXL, or
IN1RP to SPKMIXR
(SPKATTN = 0dB)
IN2LN, IN2RN, IN2LP or
IN2RP to MIXOUTL or
MIXOUTR
Gain = -9dB
(*MIXOUTx_VOL=011)
IN2LN, IN2RN, IN2LP or
IN2RP to MIXOUTL or
MIXOUTR
Gain = 0dB
(*MIXOUTx_VOL=000) w
53 k
25 k
1.3 k
58 k
36 k
2.5 k
56 k
17 k
9.8 k
3.7 k
89 k
27 k
43 k
18 k
PP, August 2012, Rev 3.4
16
Pre-Production
WM8958
Test Conditions
T
A
= +25 o C.
With the exception of the condition(s) noted above, the following electrical characteristics are valid across the full range of recommended operating conditions.
RXVOICE to MIXINL or
MIXINR
Gain = -12dB
(IN2LRP_MIXINx_VOL=001)
RXVOICE to MIXINL or
MIXINR
Gain = 0dB
(IN2LRP_MIXINx_VOL=101)
RXVOICE to MIXINL or
MIXINR
Gain = +6dB
(IN2LRP_MIXINx_VOL=111)
Direct Voice to Earpiece
Gain = -6dB
(HPOUT2_VOL=1)
Direct Voice to Earpiece
Gain = 0dB
(HPOUT2_VOL=0)
Direct Voice to Speaker
Gain = 0dB
(SPKOUTx_BOOST=000)
Direct Voice to Speaker
Gain = +6dB
(SPKOUTx_BOOST=100)
Direct Voice to Speaker
Gain = +9dB
(SPKOUTx_BOOST=110)
Direct Voice to Speaker
Gain = +12dB
(SPKOUTx_BOOST=111)
48 k
12 k
6.0 k
20 kΩ
10 kΩ
170 kΩ
85 kΩ
60 kΩ
43 kΩ
Note 5: Input resistance will be seen in parallel with the resistance of other enabled input paths from the same pins
w PP, August 2012, Rev 3.4
17
WM8958
Pre-Production
PROGRAMMABLE GAINS
Test Conditions
The following electrical characteristics are valid across the full range of recommended operating conditions.
Input PGAs (IN1L, IN2L, IN1R and IN2R)
C1 Minimum Programmable Gain
C2
C3
Maximum Programmable Gain
Programmable Gain Step Size
Input Mixers (MIXINL and MIXINR)
C6 Minimum Programmable Gain
C7 Maximum Programmable Gain
C8
C9
Programmable Gain Step Size
Minimum Programmable Gain
C10 Maximum Programmable Gain
C11 Programmable Gain Step Size
Minimum Programmable Gain
Maximum Programmable Gain
Guaranteed monotonic
Input PGA signal paths
Direct IN1xP input signal paths
(Note the available gain settings are
-12, -9, -6, -3, 0, +3, +6, +15dB)
MIXOUTx Record signal paths
-16.5
+30 dB
1.5 dB
0 dB dB
+30 dB
30 dB
-12 dB
+15 dB
3 dB
-12 dB
+6 dB
C12
Programmable Gain Step Size
Minimum Programmable Gain
C13 Maximum Programmable Gain
C14 Programmable Gain Step Size
Output Mixers (MIXOUTL and MIXOUTR)
C17 Minimum Programmable Gain
RXVOICE (VRXP-VRXN) signal paths
3 dB
-12 dB
+6 dB
3 dB
C18 Maximum Programmable Gain
C19 Programmable Gain Step Size
Speaker Mixers (SPKMIXL and SPKMIXR)
C21 Minimum Programmable Gain
C22 Maximum Programmable Gain
0 dB
3 dB
C23 Programmable Gain Step Size
C27 Programmable Gain Step Size
Line Output Drivers (LINEOUT1NMIX, LINEOUT1PMIX, LINEOUT2NMIX and LINEOUT2PMIX)
0 dB
3 dB
Output PGAs (HPOUT1LVOL, HPOUT1RVOL, MIXOUTLVOL, MIXOUTRVOL, SPKLVOL and SPKRVOL)
C25 Minimum Programmable Gain
C26 Maximum Programmable Gain
Guaranteed monotonic -57 dB
+6 dB
1 dB
C29 Minimum Programmable Gain
C30 Maximum Programmable Gain
C31 Programmable Gain Step Size
0 dB
6 dB
Earpiece Driver (HPOUT2MIX)
C33 Minimum Programmable Gain
C34 Maximum Programmable Gain 0 dB
6 dB C35 Programmable Gain Step Size
Speaker Output Drivers (SPKOUTLBOOST and SPKOUTRBOOST)
C38 Minimum Programmable Gain
C39 Maximum Programmable Gain
(Note the available gain settings are
0, +1.5, +3, +4.5, +6, +7.5, +9, +12dB)
C40 Programmable Gain Step Size
0 dB
+12 dB
1.5 dB w PP, August 2012, Rev 3.4
18
Pre-Production
OUTPUT DRIVER CHARACTERISTICS
Test Conditions
The following electrical characteristics are valid across the full range of recommended operating conditions.
WM8958
Line Output Driver (LINEOUT1P, LINEOUT1N, LINEOUT2P, LINEOUT2N)
Output discharge resistance
Direct connection
Connection via 1kΩ series resistor
LINEOUTn_DISCH=1, VROI=0
LINEOUTn_DISCH=1, VROI=1,
LINEOUTn_ENA=0
Headphone Output Driver (HPOUT1L, HPOUT1R)
Normal operation
DC offset across load
Earpiece Output Driver (HPOUT2L, HPOUT2R)
Device survival with load applied indefinitely
(see note 6)
DC Servo complete
500 Ω
15
100
8
100
2000 pF kΩ
Ω mΩ
2 nF
TBD mV
Direct connection
DC offset across load
Speaker Output Driver (SPKOUTLP, SPKOUTLN, SPKOUTRP, SPKOUTRN)
Stereo Mode (SPKMODE=0), Class AB
DC offset across load
SPKVDD leakage current
Stereo Mode (SPKMODE=0), Class D
Mono Mode (SPKMODE=1)
Sum of I
SPKVDD1
+ I
SPKVDD2
8
4
4
200 pF
±5 mV
Ω
TBD pF
±5 mV
1 µA
Note 6: In typical applications, the PCB trace resistance, jack contact resistance and ESR of any series passive components
(eg. inductor or ferrite bead) are sufficient to provide this minimum resistance; additional series components are not required. w PP, August 2012, Rev 3.4
19
WM8958
Pre-Production
ADC INPUT PATH PERFORMANCE
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
D1
D2
D3
D4
Line Inputs to ADC via MIXINL and MIXINR
SNR
THD
THD+N
A-weighted
-1dBV input
-1dBV input
Channel Separation
(L/R)
PSRR (all supplies) 100mV (pk-pk)
217Hz
Record Path (DACs to ADCs via MIXINL and MIXINR)
SNR
THD
THD+N
Channel Separation
(L/R)
A-weighted
-1dBFS input
-1dBFS input
Input PGAs to ADC via MIXINL or MIXINR
SNR A-weighted
THD
THD+N
-1dBV input
-1dBV input
Channel Separation
(L/R)
PSRR (AVDD1) 100mV (pk-pk)
217Hz
RXVOICE to ADCL or ADCR
SNR
THD
THD+N
A-weighted
-1dBV input
-1dBV input
94 dB
-83 dB
-81 dB
73 dB
92 dB
-74 dB
-72 dB
84 95 dB
-82 -72 dB
-80 -70 dB
97 dB
94 dB
-84 dB
-82 dB w PP, August 2012, Rev 3.4
20
Pre-Production
WM8958
DAC OUTPUT PATH PERFORMANCE
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
E1
E2
E5
E6
DAC to Single-Ended Line Output (Load = 10k // 50pF)
SNR A-weighted
THD
THD+N
0dBFS input
0dBFS input
Channel Separation
(L/R)
PSRR (all supplies) 100mV (pk-pk)
217Hz
LINEOUTFB rejection LINEOUTn_FB=1,
100mV (pk-pk)
217Hz
DAC to Differential Line Output (Load = 10k // 50pF)
SNR A-weighted
THD
THD+N
0dBFS input
0dBFS input
Channel Separation
(L/R)
PSRR (all supplies) 100mV (pk-pk)
217Hz
DAC to Headphone on HPOUT1L or HPOUT1R (Load = 32 )
SNR (A-weighted) DAC_OSR128=1
DAC_OSR128=0
THD P
O
=20mW
THD+N P
O
=20mW
THD P
O
=5mW
THD+N P
O
=5mW
Channel Separation
(L/R)
PSRR (all supplies) 100mV (pk-pk)
217Hz
HPOUT1FB rejection 100mV (pk-pk)
217Hz
DAC to Headphone on HPOUT1L or HPOUT1R (Load = 16 )
SNR (A-weighted) DAC_OSR128=1
DAC_OSR128=0
THD P
O
=20mW
THD+N P
O
=20mW
THD P
O
=5mW
THD+N P
O
=5mW
Channel Separation
(L/R)
PSRR (all supplies)
HPOUT1FB rejection
100mV (pk-pk)
217Hz
100mV (pk-pk)
217Hz
93 dB
-75
-73 dB dB
36 dB
38 dB
97 dB
-76 dB
-75 dB
51 dB
100 dB
97 dB
-72 dB
-76 dB
-74 dB
95 dB
51 dB
29 dB
90 100 dB
97 dB
-82 dB
-80 dB
95 dB
51 dB
29 dB w PP, August 2012, Rev 3.4
21
WM8958
Pre-Production
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
E9 DAC to Earpiece Driver (Load = 16 BTL)
SNR A-weighted
THD P
O
=50mW
THD+N P
O
=50mW
PSRR (all supplies) 100mV (pk-pk)
217Hz
97 dB
-71 dB
-69 dB
51 dB
E12 DAC to Speaker Outputs (Load = 8 + 22H BTL, Stereo Mode)
Class D Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted
THD P
O
=0.5W
THD+N P
O
=0.5W
THD P
O
=1.0W
THD+N P
O
=1.0W
PSRR (all supplies) 100mV (pk-pk)
217Hz
Channel Separation
(L/R)
DAC to Speaker Outputs (Load = 8 + 22H BTL, Stereo Mode)
Class AB Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted
THD P
O
=0.5W
THD+N P
O
=0.5W
THD P
O
=1.0W
THD+N P
O
=1.0W
PSRR (all supplies) 100mV (pk-pk)
217Hz
Channel Separation
(L/R)
DAC to Speaker Outputs (Load = 4 + 22H BTL, Stereo Mode)
Class D Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted
THD P
O
=0.5W
THD+N P
O
=0.5W
THD P
O
=1.0W
THD+N P
O
=1.0W
THD P
O
=2.0W
THD+N P
O
=2.0W
PSRR (all supplies)
Channel Separation
(L/R)
100mV (pk-pk)
217Hz
85 94 dB
-65 dB
-70 dB
-68 dB
43 dB
80 dB
96 dB
-67 dB
-65 dB
-64 dB
-62 dB
43 dB
80 dB
93 dB
dB
-63 dB
dB
-63 dB
dB
-66 dB
dB
dB w PP, August 2012, Rev 3.4
22
Pre-Production
WM8958
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
E13 Speaker Output Power (Load = 8 + 22H BTL, Stereo Mode)
Output Power SPKVDD1=
SPKVDD2=5.0V
THD+N ≤ 1%
SPKVDD1=
SPKVDD2=4.2V
THD+N ≤ 1%
Class AB
Class D
Class AB
Class D
SPKVDD1=
SPKVDD2=3.7V
THD+N ≤ 1%
Class AB
Class D
Note that the maximum recommended speaker output power is 1W per channel into 8Ω.
Output levels that exceed this limit are not guaranteed and may cause damage to the WM8958.
Speaker Output Power (Load = 4 + 22H BTL, Stereo Mode)
1
1
0.95
0.95
0.75
0.75
W
W
W
Output Power SPKVDD1=
SPKVDD2=5.0V
THD+N ≤ 1%
SPKVDD1=
SPKVDD2=4.2V
THD+N ≤ 1%
Class D
(see note below)
Class D
2.3 W
1.6 W
SPKVDD1=
SPKVDD2=3.7V
THD+N ≤ 1%
Speaker Output Power (Load = 4 + 22H BTL, Mono Mode)
Output Power
Class D
SPKVDD1=
SPKVDD2=5.0V
THD+N ≤ 1%
Class AB
(see note below)
Class D
(see note below)
Note that the maximum recommended speaker output power is 2W per channel into 4Ω.
Output levels that exceed this limit are not guaranteed and may cause damage to the WM8958.
1.2 W
2.7 W
2.7 w PP, August 2012, Rev 3.4
23
WM8958
Pre-Production
BYPASS PATH PERFORMANCE
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
F1 Input PGA to Differential Line Output (Load = 10k // 50pF)
SNR A-weighted 100 dB
-90 dB
-87 dB
F3
F2
F4
Input PGA to Headphone via MIXOUTL or MIXOUTR (Load = 16 )
SNR
THD
THD+N
THD
THD+N
A-weighted
P
P
O
P
O
O
P
O
PSRR (all supplies) 100mV (pk-pk)
217Hz
Channel Separation
(L/R)
Line Input (IN2LP or IN2RP) to Headphone via MIXOUTL or MIXOUTR (Load = 16 )
SNR A-weighted
THD P
O
=20mW
THD+N P
O
=20mW
THD P
O
=5mW
THD+N P
O
=5mW
PSRR (all supplies) 100mV (pk-pk)
217Hz
Line Input (IN2LN or IN2RN) to Headphone via MIXOUTL or MIXOUTR (Load = 16 )
SNR
THD
THD+N
THD
A-weighted
P
P
O
P
O
O
THD+N
PSRR (all supplies)
P
O
100mV (pk-pk)
217Hz
Channel Separation
(L/R)
98 dB
49 dB
100 dB
-86 dB
-84 dB
-84 dB
-82 dB
49 dB
100 dB
49 dB w PP, August 2012, Rev 3.4
24
Pre-Production
WM8958
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
F5
F6
F7
Direct Voice Path to Earpiece Driver (Load = 16 BTL)
SNR A-weighted
THD P
O
=50mW
THD+N P
O
=50mW
PSRR (all supplies) 100mV (pk-pk)
217Hz
90 104 dB
-70 dB
91 dB
Direct Voice Path to Speaker Outputs (Load = 8 + 22H BTL, Stereo Mode)
Class D Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted
THD P
THD+N P
O
O
=0.5W
=0.5W
THD P
O
=1.0W
THD+N P
O
=1.0W
PSRR (all supplies) 100mV (pk-pk)
217Hz
Direct Voice Path to Speaker Outputs (Load = 8 + 22H BTL, Stereo Mode)
Class AB Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted
97 dB
-62 dB
-60 dB
-67 dB
-65 dB
63 dB
103 dB
THD P
THD+N P
O
O
=0.5W
=0.5W
-62 dB
-60 dB
THD P
O
=1.0W
THD+N P
O
=1.0W
PSRR (all supplies) 100mV (pk-pk)
217Hz
-64 dB
-62 dB
67 dB
Line Input to Speaker Outputs via SPKMIXL or SPKMIXR (Load = 8 + 22H BTL, Stereo Mode)
Class D Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted 93 dB
THD P
O
=0.5W
THD+N P
O
=0.5W
-62 dB
-60 dB
THD P
O
=1.0W
THD+N P
O
=1.0W
-67 dB
-65 dB
PSRR (all supplies) 100mV (pk-pk)
217Hz
47 dB
Line Input to Speaker Outputs via SPKMIXL or SPKMIXR (Load = 8 + 22H BTL, Stereo Mode)
Class AB Mode, +12dB boost (SPKOUTx_BOOST = 111)
SNR A-weighted 96 dB
THD P
O
=0.5W
THD+N P
O
=0.5W
THD P
O
=1.0W
THD+N P
O
=1.0W
PSRR (all supplies) 100mV (pk-pk)
217Hz
-72 dB
-68 dB
-64 dB
-62 dB
47 dB w PP, August 2012, Rev 3.4
25
WM8958
Pre-Production
MULTI-PATH CROSSTALK
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
85 dB G1
G2
Headset Voice Call:
DAC/Headset to Tx Voice
Separation
1kHz 0dBFS DAC playback direct to HPOUT1L and HPOUT1R;
Quiescent input on IN1LN/P or
IN1RN/P (Gain=+12dB), differential line output; Measure crosstalk at differential line output
Speakerphone Voice Call:
DAC/Speaker to Tx Voice
Separation
1kHz 0dBFS DAC playback to speakers, 1W/chan output;
Quiescent input on IN1LN/P or
IN1RN/P (Gain=+12dB), differential line output; Measure crosstalk at differential line output
CR
OS
ST
AL
K
100 dB
G3
G4
G5
Earpiece PCM Voice Call:
RXVOICE to Tx Voice Separation fs=8kHz for ADC and DAC,
DAC_SB_FILT=1; -5dBFS, DAC output to HPOUT2P-HPOUT2N;
Quiescent input on input PGA
(Gain=+12dB) to ADC via MIXINL or MIXINR; Measure crosstalk at
ADC output
Speakerphone PCM Voice Call:
DAC/Speaker to ADC Separation fs=8kHz for ADC and DAC,
DAC_SB_FILT=1; 0dBFS DAC output to speaker (1W output);
ADC record from input PGA
(Gain=+30dB); Measure crosstalk on ADC output
Speakerphone PCM Voice Call:
ADC to DAC/Speaker Separation fs=8kHz for ADC and DAC,
DAC_SB_FILT=1; Quiescent DAC output to speaker; ADC record from input PGA (Gain=+30dB +
30dB boost); Measure crosstalk on speaker output
110 dB
90 dB
95 dB w PP, August 2012, Rev 3.4
26
Pre-Production
WM8958
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
G6
G7
G8
Earpiece Speaker Voice Call:
Tx Voice and RXVOICE
Separation
1kHz Full scale differential input on VRXP-VRXN, output to
HPOUT2P-HPOUT2N; Quiescent input on IN1LN/P or IN1RN/P
(Gain=+12dB), differential line output; Measure crosstalk at differential line output
Headset Voice Call:
Tx Voice and RXVOICE
Separation
1kHz full scale differential input on
VRXP-VRXN via RXVOICE to
MIXOUTL and MIXOUTR, output to HPOUT1L and HPOUT1R;
Quiescent input on IN1LN/P or
IN1RN/P (Gain=+12dB), differential line output; Measure crosstalk at differential line output
Stereo Line Record and Playback:
DAC/Headset to ADC Separation
-5dBFS input to DACs, playback to
HPOUT1L and HPOUT1R; ADC record from line input; Measure crosstalk on ADC output
IN1LN or
IN1RN
IN1LP or
IN1RP
-
+12dB
+
Quiescent input
IN1L or IN1R
(Single-ended or differential mode)
VRXN
Full scale input
VRXP
-
+
RXVOICE
(MIXINL or
MIXINR)
IN1LP or
IN1RP
Quiescent input
CROSSTALK
LINEOUT1NMIX or
LINEOUT2NMIX
LINEOUT1PMIX or
LINEOUT2PMIX
MIXINL or
MIXINR
+
DACL
MIXOUTL
+
0dB
HPOUT1LVOL
0dB
+
MIXOUTR HPOUT1RVOL
ADCL or
ADCR
0dB
0dB
0dB
0dB
LINEOUT1N or
LINEOUT2N
LINEOUT1P or
LINEOUT2P
HPOUT1LVOL
HPOUT1L
HPOUT1R
HPOUT1L
HPOUT1R
DACR
HPOUT1RVOL
100 dB
90 dB
95 dB w PP, August 2012, Rev 3.4
27
WM8958
Pre-Production
DIGITAL INPUT / OUTPUT
Test Conditions
The following electrical characteristics are valid across the full range of recommended operating conditions.
Digital Input / Output (except DMICDATn and DMICCLK)
Digital I/O is referenced to DBVDD1, DBVDD2 or DBVDD3. See “Pin Description” for the domain applicable to each pin.
H16 Input HIGH Level, V
IH
V
DBVDDn
H17 Input LOW Level, V
IL
DBVDDn
Note that digital input pins should not be left unconnected / floating.
H18 Output HIGH Level, V
OH
I
OH
DBVDDn
V
H19 Output LOW Level, V
OL
I
OL
=-1mA
DBVDDn
-0.9
Digital Microphone Input / Output (DMICDATn and DMICCLK)
H22 DMICDATn input HIGH Level, V
IH
MICBIAS1
H23 DMICDATn input LOW Level, V
IL
MICBIAS1
H24
H25
DMICCLK output HIGH Level, V
DMICCLK output LOW Level, V
OH
OL
I
OH
MICBIAS1
I
OL
=-1mA
MICBIAS1
-0.9
V
V
DIGITAL FILTER CHARACTERISTICS
Test Conditions
The following electrical characteristics are valid across the full range of recommended operating conditions.
ADC Decimation Filter
Passband +/- 0.05dB
-6dB
0 0.454 fs
Stopband f > 0.546 fs -85 dB
DAC Interpolation Filter
Passband
Stopband
+/- 0.05dB
-6dB
0.454 fs f > 0.546 fs
0
-85
0.454 fs
+/- 0.05 dB dB w PP, August 2012, Rev 3.4
28
Pre-Production
WM8958
MICROPHONE BIAS CHARACTERISTICS
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, unless otherwise stated.
Microphone Bias (MICBIAS1 and MICBIAS2)
Note - No capacitor on MICBIASn
Note - In regulator mode, it is required that AVDD1 - V
MICBIASn
> 200mV
MICBn_LVL = 000
Regulator mode (MICBn_MODE=0)
Load current ≤ 1.0mA
MICBn_LVL = 001
MICBn_LVL = 010
MICBn_LVL = 011
MICBn_LVL = 100
MICBn_LVL = 101
MICBn_LVL = 110
MICBn_LVL = 111
Bias Voltage
Bypass mode (MICBn_MODE=1)
Load current ≤ 3.6mA
H4
H5
Output Noise Density
Integrated Noise Voltage
Output discharge resistance
Regulator mode
(MICBn_MODE=0)
Bypass mode
(MICBn_MODE=1)
Regulator mode
(MICBn_MODE=0),
MICBn_LVL = 100,
Load current = 1mA,
Measured at 1kHz
Regulator mode
(MICBn_MODE=0),
MICBn_LVL = 100,
Load current = 1mA,
100Hz to 7kHz, A-weighted
100mV (pk-pk) 217Hz
100mV (pk-pk) 10kHz
Regulator mode
(MICBn_MODE=0)
MICBn_ENA=0,
MICBn_DISCH=1
-5%
-5%
-5%
-5%
-5%
-5%
-5%
-5%
80mV
1.5
1.8
1.9
2.0
2.2
2.4
2.5
2.6
100
80
+5%
+5%
+5%
+5%
+5%
+5%
+5%
+5%
V
3.6
µV
RMS dB w PP, August 2012, Rev 3.4
29
WM8958
Pre-Production
MISCELLANEOUS CHARACTERISTICS
Test Conditions
AVDD1=3.0V (powered from LDO1), DCVDD=1.2V (powered from LDO2), AVDD2=DBVDD1=DBVDD2=DBVDD3=CPVDD=1.8V,
LDO1VDD=SPKVDD1=SPKVDD2=5V, DGND=AGND=CPGND=SPKGND1=SPKGND2=HP2GND=0V,
T
A
= +25 o C, 1kHz sinusoidal signal, fs = 48kHz, PGA gain = 0dB, 24-bit audio data unless otherwise stated.
Analogue Reference Levels
H1 VMID Midrail Reference Voltage
VMID Start-Up time
VMID_SEL = 01,
4.7
F capacitor on VMIDC
VMID_SEL = 01,
VMID_RAMP = 11,
4.7
F capacitor on VMIDC
External Accessory Detection
Load impedance detection range
(MICDET)
2.2kΩ (2%) MICBIAS2 resistor.
Note these characteristics assume no other component is connected to
MICDET. See “Applications Information” for recommended external components when a typical microphone is present. for MICD_LVL[0] = 1 for MICD_LVL[1] = 1 for MICD_LVL[2] = 1 for MICD_LVL[3] = 1 for MICD_LVL[4] = 1 for MICD_LVL[5] = 1 for MICD_LVL[6] = 1 for MICD_LVL[7] = 1
Frequency Locked Loops (FLLs)
H30 Free-running mode start-up time
H31 Free-running mode frequency accuracy
F
REF
=32kHz,
F
OUT
=12.288MHz
F
REF
=12MHz,
F
OUT
=12.288MHz
Reference supplied initially
No reference provided
LDO Regulators
H38 LDO1 Start-Up Time 4.7
F capacitor on AVDD1
1 F capacitor on VREFC
LDO1 Drop-Out voltage
(LDO1VDD - AVDD1)
LDO1 PSRR (LDO1VDD)
H42 LDO2 Start-Up Time
LDO2 PSRR (DBVDD1)
100mV (pk-pk) 217Hz
2.2
F capacitor on DCVDD
1 F capacitor on VREFC
100mV (pk-pk) 217Hz
-3% AVDD1/2 +3% V
0
13.33
27.16
42.48
65
114
191
475
+/-10
+/-30
TBD
TBD
3
15.27
30.96
49.47
85
155.24
329.87
30000
Ω
%
% dB dB w PP, August 2012, Rev 3.4
30
Pre-Production
WM8958
TERMINOLOGY
1. Signal-to-Noise Ratio (dB) – SNR is a measure of the difference in level between the maximum full scale output signal and the output with no input signal applied.
2. Total Harmonic Distortion (dB) – THD is the level of the rms value of the sum of harmonic distortion products relative to the amplitude of the measured output signal.
3. Total Harmonic Distortion plus Noise (dB) – THD+N is the level of the rms value of the sum of harmonic distortion products plus noise in the specified bandwidth relative to the amplitude of the measured output signal.
4. Power Supply Rejection Ratio (dB) - PSRR is the ratio of a specified power supply variation relative to the output signal that results from it. PSRR is measured under quiescent signal path conditions.
5. Common Mode Rejection Ratio (dB) – CMRR is the ratio of a specified input signal (applied to both sides of a differential input), relative to the output signal that results from it.
6. Channel Separation (L/R) (dB) – left-to-right and right-to-left channel separation is the difference in level between the active channel (driven to maximum full scale output) and the measured signal level in the idle channel at the test signal frequency. The active channel is configured and supplied with an appropriate input signal to drive a full scale output, with signal measured at the output of the associated idle channel.
7. Multi-Path Crosstalk (dB) – is the difference in level between the output of the active path and the measured signal level in the idle path at the test signal frequency. The active path is configured and supplied with an appropriate input signal to drive a full scale output, with signal measured at the output of the specified idle path.
8. Mute Attenuation – This is a measure of the difference in level between the full scale output signal and the output with mute applied.
9. All performance measurements carried out with 20kHz low pass filter, and where noted an A-weighted filter. Failure to use such a filter will result in higher THD and lower SNR readings than are found in the Electrical Characteristics. The low pass filter removes out of band noise; although it is not audible it may affect dynamic specification values. w PP, August 2012, Rev 3.4
31
WM8958
Pre-Production
TYPICAL PERFORMANCE
TYPICAL POWER CONSUMPTION
OPERATING MODE TEST CONDITIONS SPKVDD
(Note 3)
LDO1VDD AVDD2 CPVDD DBVDD
(Note 4)
TOTAL
Off (Battery Leakage only)
LDO1 disabled, LDO2 disabled
Standby
LDO1 disabled, LDO2 enabled
Standby
LDO1 enabled, LDO2 enabled
All supplies present,
No clocks,
Default register settings
Music playback to Headphone (32ohm load)
AIF1 to DAC to
HPOUT1 (stereo)
All supplies present,
No clocks,
Default register settings
AIF1 to DAC to
HPOUT1 (stereo)
LDOs disabled,
See Note 7 fs=44.1kHz,
Clocking rate=256fs,
24-bit I2S, Slave mode fs=44.1kHz,
Clocking rate=128fs,
24-bit I2S, Slave mode,
Class W
4.2V
1.1
A
4.2V
1.8
A
4.2V
1.8
A
4.2V
0.0mA
4.2V
0.4
4.2V
1
4.2V
89
A
A
A
4.2V
2.05mA
0.0V
5.5
1.8V
60
1.8V
65
A
A
A
1.8V
0.32mA
0.0V
5 A
1.8V
5 A
1.8V
5 A
1.8V
0.48mA
0.0V
9.5
1.8V
62
72
A
A
1.8V
A
1.8V
1.13mA
3.6V
0.0mA
AVDD1=
2.4V
1.43mA
1.8V
0.21mA
1.8V
0.21mA
DBVDD=
1.8V
0.01mA
DCVDD=
1.2V
0.94mA
Music playback to Class D speaker output (8ohm, 22 H load)
AIF1 to DAC to
SPKOUT (stereo) fs=44.1kHz,
Clocking rate=256fs,
24-bit I2S, Slave mode,
4.2V
1.65mA
+7.5dB Class D boost
AIF1 to DAC to
SPKOUT (Left) fs=44.1kHz,
Clocking rate=256fs,
24-bit I2S, Slave mode,
+0.0dB Class D boost
4.2V
0.74mA
AIF1 to AIF3 Mono Digital Bypass (eg. Bluetooth video call)
AIF1(L) to AIF3(L),
AIF3(L) to AIF1(L) fs=8kHz,
Clocking rate=256fs,
24-bit I2S, Slave mode
4.2V
0.0mA
AIF2 to AIF3 Mono Digital Bypass (eg. Bluetooth voice call)
AIF2(L) to AIF3(L),
AIF3(L) to AIF2(L) fs=8kHz,
Clocking rate=256fs,
24-bit I2S, Slave mode
4.2V
0.002mA
4.2V
2.36mA
4.2V
2.34mA
4.2V
0.09mA
4.2V
0.089mA
1.8V
1.24mA
1.8V
0.79mA
1.8V
0.07mA
1.8V
0.065mA
1.8V
0.01mA
1.8V
0.01mA
1.8V
0.01mA
1.8V
0.003mA
1.8V
1.13mA
1.8V
1.13mA
1.8V
0.41mA
1.8V
0.311mA
Notes:
1. AVDD1 = 3.0V, generated by LDO1.
2. DCVDD = 1.2V, generated by LDO2.
3. SPKVDD = SPKVDD1 = SPKVDD2.
4. DBVDD = DBVDD1 = DBVDD2 = DBVDD3.
5. I
SPKVDD
= I
SPKVDD1
+ I
SPKVDD2
.
6. I
DBVDD
= I
DBVDD1
+ I
DBVDD2
+ I
DBVDD3
.
7. Power consumption for music playback with LDOs disabled requires an external supply for AVDD1 and DCVDD
0.01mW
0.2mW
0.6mW
12.1mW
5.34mW
21.1mW
16.4mW
1.2mW
1.1mW w PP, August 2012, Rev 3.4
32
Pre-Production
TYPICAL SIGNAL LATENCY
OPERATING MODE
WM8958
TEST CONDITIONS LATENCY
AIF2 to DAC Stereo Path
AIF2 EQ enabled,
AIF2 3D enabled,
AIF2 DRC enabled,
SRC enabled fs=8kHz,
Clock rate = 256fs fs=48kHz,
Clock rate = 256fs fs=8kHz,
Clock rate = 256fs fs=48kHz,
Clock rate = 256fs fs=8kHz,
Clock rate = 1536fs fs=8kHz,
Clock rate = 1536fs fs=8kHz,
Clock rate = 256fs fs=8kHz,
Clock rate = 256fs
SYSCLK=AIF1CLK 1.4ms
SYSCLK=AIF1CLK 1.3ms
SYSCLK=AIF1CLK 1.7ms
SYSCLK=AIF1CLK 1.4ms
ADC to AIF2 Stereo Path
Digital Sidetone HPF enabled,
AIF2 DRC enabled,
AIF2 HPF enabled,
SRC enabled
Digital Sidetone HPF disabled,
AIF2 DRC disabled,
AIF2 HPF disabled,
SRC disabled
Digital Sidetone HPF disabled,
AIF2 DRC disabled,
AIF2 HPF disabled,
SRC enabled fs=8kHz,
Clock rate = 256fs fs=48kHz,
Clock rate = 256fs fs=48kHz,
Clock rate = 256fs fs=8kHz,
Clock rate = 256fs fs=8kHz,
Clock rate = 256fs
Clock rate = 1536fs fs=8kHz,
Clock rate = 1536fs
SYSCLK=AIF1CLK 2.2ms
SYSCLK=AIF1CLK 1.2ms
SYSCLK=AIF1CLK 1.1ms
Notes:
1. These figures are relevant to typical voice call modes, assuming AIF2 is connected to the baseband processor
2. The SRC (Sample Rate Converter) is enabled automatically whenever required w PP, August 2012, Rev 3.4
33
WM8958
Pre-Production
SPEAKER DRIVER PERFORMANCE
Typical speaker driver THD+N performance is shown below for Class D and Class AB modes. Curves are shown for typical
SPKVDD supply voltage, gain and load conditions.
10
THD+N vs. Output Power
Class D, Mono (SPKMODE=1), 8Ω + 10µH
SPKVDD = 3.3V
10
THD+N vs. Output Power
Class D, Mono (SPKMODE=1), 4Ω + 10µH
SPKVDD = 3.3V
SPKVDD = 3.7V
1
SPKVDD = 5.0V
1
SPKVDD = 3.7V
SPKVDD = 4.2V
SPKVDD = 4.5V
SPKVDD = 5.0V
0.1
0.1
SPKVDD = 4.2V
0.01
0
10
0.2
0.4
0.6
0.8
1 1.2
Output Power (W)
1.4
THD+N vs. Output Power
Class AB, Mono (SPKMODE=1), 8Ω + 10µH
SPKVDD = 3.3V
1.6
1.8
2
0.01
0
10
SPKVDD = 4.5V
0.5
1 1.5
Output Power (W)
2
THD+N vs. Output Power
Class AB, Mono (SPKMODE=1), 4Ω + 10µH
SPKVDD = 3.3V
2.5
3
SPKVDD = 3.7V
SPKVDD = 5.0V
1 1 SPKVDD = 3.7V
SPKVDD = 5.0V
SPKVDD = 4.5V
0.1
0.1
SPKVDD = 4.2V
0.01
0
10
0.2
0.4
0.6
0.8
1 1.2
Output Power (W)
1.4
THD+N vs. Output Power
Class D, Stereo (SPKMODE=0), 8Ω + 10µH
SPKVDD = 3.3V
1.6
1.8
2
0.01
0
10
SPKVDD = 4.2V
SPKVDD = 4.5V
0.5
1 1.5
Output Power (W)
2
THD+N vs. Output Power
Class D, Stereo (SPKMODE=0), Load = 4Ω + 10µH
SPKVDD = 3.3V
2.5
3
SPKVDD = 3.7V
1 SPKVDD = 3.7V
0.1
SPKVDD = 5.0V
SPKVDD = 4.5V
SPKVDD = 4.2V
1
0.1
SPKVDD = 5.0V
SPKVDD = 4.5V
0.01
0
10
0.2
0.4
0.6
0.8
1 1.2
Output Power (W)
1.4
THD+N vs. Output Power
Class AB, Stereo (SPKMODE=0), 8Ω + 10µH
SPKVDD = 3.3V
1.6
1.8
2
0.01
0 0.5
1 1.5
Output Power (W)
2
SPKVDD = 4.2V
2.5
3
1
SPKVDD = 3.7V
SPKVDD = 5.0V
SPKVDD = 4.5V
0.1
SPKVDD = 4.2V
0.01
0 0.2
0.4
0.6
0.8
1 1.2
Output Power (W)
1.4
1.6
1.8
2 w PP, August 2012, Rev 3.4
34
Pre-Production
SIGNAL TIMING REQUIREMENTS
SYSTEM CLOCKS & FREQUENCY LOCKED LOOP (FLL)
WM8958
Figure 2 Master Clock Timing
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
PARAMETER SYMBOL CONDITIONS MIN
Master Clock Timing (MCLK1 and MCLK2)
MCLK as input to FLL,
FLLn_REFCLK_DIV = 01, 10, 11
37 ns
MCLK cycle time T
MCLKY
74
MCLK duty cycle
(= T
MCLKH
: T
MCLKL
)
Frequency Locked Loops (FLL1 and FLL2)
FLL Input Frequency
MCLK as input to FLL,
FLLn_REFCLK_DIV = 00
FLL not used, AIFnCLK_DIV = 1
FLL not used, AIFnCLK_DIV = 0
40
80
60:40 40:60
MHz FLLn_REFCLK_DIV = 00
FLLn_REFCLK_DIV = 01
FLLn_REFCLK_DIV = 10
FLLn_REFCLK_DIV = 11
0.032
0.064
0.128
0.256
13.5
27
27
27
Internal Clocking
SYSCLK frequency
AIF1CLK frequency
AIF2CLK frequency
DSP2CLK frequency w PP, August 2012, Rev 3.4
35
WM8958
Pre-Production
AUDIO INTERFACE TIMING
DIGITAL MICROPHONE (DMIC) INTERFACE TIMING
Figure 3 Digital Microphone Interface Timing
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
Digital Microphone Interface Timing
DMICCLK cycle time
DMICCLK duty cycle
DMICDAT (Left) setup time to falling DMICCLK edge
DMICDAT (Left) hold time from falling DMICCLK edge
DMICDAT (Right) setup time to rising DMICCLK edge
DMICDAT (Right) hold time from rising DMICCLK edge t t
CY
ns
LSU
45:55 55:45 %
15 ns ns t
RSU
15 ns t t
LH
0
RH
0 ns w PP, August 2012, Rev 3.4
36
Pre-Production
DIGITAL AUDIO INTERFACE - MASTER MODE
WM8958
BCLK
(output)
V
OH
V
OL t
BCY
LRCLK
(output)
V
OH
V
OL t
DL
ADCDAT
(output)
V
OH
V
OL t
DDA
DACDAT
(input)
V
IH
V
IL t
DST t
DHT
Figure 4 Audio Interface Timing - Master Mode
Note that BCLK and LRCLK outputs can be inverted if required; Figure 4 shows the default, noninverted polarity of these signals.
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
Audio Interface Timing - Master Mode
BCLK cycle time
LRCLK propagation delay from BCLK falling edge
ADCDAT propagation delay from BCLK falling edge
DACDAT setup time to BCLK rising edge
DACDAT hold time from BCLK rising edge
Audio Interface Timing - Ultrasonic (4FS) Master Mode t t t t t
BCY
DL
DDA
DST
DHT
BCLK cycle time
ADCDAT propagation delay from BCLK falling edge t
BCY t
DDA
Note that the descriptions above assume non-inverted polarity of BCLK and LRCLK. w PP, August 2012, Rev 3.4
37
WM8958
Pre-Production
DIGITAL AUDIO INTERFACE - SLAVE MODE
Figure 5 Audio Interface Timing - Slave Mode
Note that BCLK and LRCLK inputs can be inverted if required; Figure 5 shows the default, noninverted polarity.
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
Audio Interface Timing - Slave Mode
BCLK cycle time
BCLK pulse width high
BCLK pulse width low
LRCLK set-up time to BCLK rising edge
LRCLK hold time from BCLK rising edge
DACDAT hold time from BCLK rising edge
ADCDAT propagation delay from BCLK falling edge
DACDAT set-up time to BCLK rising edge t
BCY
160 t
BCH
64 t
BCL
64 t
LRSU
10 t
LRH
10 t
DH
10 t
DD t
DS
32
Note that the descriptions above assume non-inverted polarity of BCLK and LRCLK. ns ns ns ns ns ns ns w PP, August 2012, Rev 3.4
38
Pre-Production
WM8958
DIGITAL AUDIO INTERFACE - TDM MODE
When TDM operation is used on the ADCDATn pins, it is important that two devices do not attempt to drive the ADCDATn pin simultaneously. To support this requirement, the ADCDATn pins can be configured to be tri-stated when not outputting data.
The timing of the WM8958 ADCDATn tri-stating at the start and end of the data transmission is described in Figure 6 below.
Figure 6 Audio Interface Timing - TDM Mode
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
TDM Timing - Master Mode
ADCDAT setup time from BCLK falling edge
ADCDAT release time from BCLK falling edge
TDM Timing - Slave Mode
ADCDAT setup time from BCLK falling edge
ADCDAT release time from BCLK falling edge
0 ns
5 ns w PP, August 2012, Rev 3.4
39
WM8958
CONTROL INTERFACE TIMING
Pre-Production
Figure 7 Control Interface Timing
Test Conditions
The following timing information is valid across the full range of recommended operating conditions.
SCLK Frequency
SCLK Low Pulse-Width
SCLK High Pulse-Width
Hold Time (Start Condition)
Setup Time (Start Condition)
Data Setup Time
SDA, SCLK Rise Time
SDA, SCLK Fall Time
Setup Time (Stop Condition)
Data Hold Time
Pulse width of spikes that will be suppressed t t t t
1 t
2
3 t
4
600 t
5
100 t
6 t
7 t
8
9 ps
1300
600
600
600
0 ns ns ns ns ns ns w PP, August 2012, Rev 3.4
40
Pre-Production
WM8958
DEVICE DESCRIPTION
INTRODUCTION
The WM8958 is a low power, high quality audio codec designed to interface with a wide range of processors and analogue components. A high level of mixed-signal integration in a very small footprint makes it ideal for portable applications such as mobile phones. Fully differential internal architecture and on-chip RF noise filters ensure a very high degree of noise immunity.
Three sets of audio interface pins are available in order to provide independent and fully asynchronous connections to multiple processors, typically an application processor, baseband processor and wireless transceiver. Any two of these interfaces can operate totally independently and asynchronously while the third interface can be synchronised to either of the other two and can also provide ultra low power loopback modes to support, for example, wireless headset voice calls.
Four digital microphone input channels are available to support advanced multi-microphone applications such as noise cancellation. An integrated microphone activity monitor is available to enable the processor to sleep during periods of microphone inactivity, saving power.
Four DAC channels are available to support use cases requiring up to four simultaneous digital audio streams to the output drivers.
Eight highly flexible analogue inputs allow interfacing to up to four microphone inputs (single-ended or differential), plus multiple stereo or mono line inputs. Connections to an external voice CODEC, FM radio, line input, handset MIC and headset MIC are all fully supported. Signal routing to the output mixers and within the CODEC has been designed for maximum flexibility to support a wide variety of usage modes. A ‘Direct Voice’ path from a voice CODEC directly to the Speaker or Earpiece output drivers is included.
Impedance sensing and measurement for external accessories is provided, for detection of the insertion or removal of microphones and other accessories. Push-button detection of up to 7 inputs can be supported using this feature.
Nine analogue output drivers are integrated, including a stereo pair of high power, high quality
Class D/AB switchable speaker drivers; these can support 2W each in stereo mode. It is also possible to configure the speaker drivers as a mono output, giving enhanced performance. A mono earpiece driver is provided, providing output from the output mixers or from the low-power differential ‘Direct
Voice’ path.
One pair of ground-referenced headphone outputs is provided; these are powered from an integrated
Charge Pump, enabling high quality, power efficient headphone playback without any requirement for
DC blocking capacitors. A DC Servo circuit is available for DC offset correction, thereby suppressing pops and reducing power consumption. Four line outputs are provided, with multiple configuration options including 4 x single-ended output or 2 x differential outputs. The line outputs are suitable for output to a voice CODEC, an external speaker driver or line output connector. Ground loop feedback is available on the headphone outputs and the line outputs, providing rejection of noise on the ground connections. All outputs have integrated pop and click suppression features.
Internal differential signal routing and amplifier configurations have been optimised to provide the highest performance and lowest possible power consumption for a wide range of usage scenarios, including voice calls and music playback. The speaker drivers offer low leakage and high PSRR; this enables direct connection to a Lithium battery. The speaker drivers provide eight levels of AC and DC gain to allow output signal levels to be maximised for many commonly-used SPKVDD/AVDD1 combinations.
The ADCs and DACs are of hi-fi quality, using a 24-bit low-order oversampling architecture to deliver optimum performance. A flexible clocking arrangement supports mixed sample rates, whilst integrated ultra-low power dual FLLs provide additional flexibility. A high pass filter is available in all ADC and digital MIC paths for removing DC offsets and suppressing low frequency noise such as mechanical vibration and wind noise. A digital mixing path from the ADC or digital MICs to the DAC provides a sidetone of enhanced quality during voice calls. DAC soft mute and un-mute is available for pop-free music playback. w PP, August 2012, Rev 3.4
41
WM8958
Pre-Production
The integrated Multiband Compressors (MBC), Dynamic Range Controllers (DRC) and ReTune TM
Mobile 5-band parametric equaliser (EQ) provide further processing capability of the digital audio paths. The MBC enables the loudness of the digital playback path to be maximised without overdriving the loudspeakers. The RMS Limiter within the MBC function enables the maximum signal level to be matched to the application requirements and/or power rating of the loudspeaker. The DRC provides compression and signal level control to improve the handling of unpredictable signal levels.
‘Anti-clip’ and ‘quick release’ algorithms improve intelligibility in the presence of transients and impulsive noises. The EQ provides the capability to tailor the audio path according to the frequency characteristics of an earpiece or loudspeaker, and/or according to user preferences.
I
The WM8958 has highly flexible digital audio interfaces, supporting a number of protocols, including
2 S, DSP, MSB-first left/right justified, and can operate in master or slave modes. PCM operation is supported in the DSP mode. A-law and -law companding are also supported. Time division multiplexing (TDM) is available to allow multiple devices to stream data simultaneously on the same bus, saving space and power. The four digital MIC and ADC channels and four DAC channels are available via four TDM channels on Digital Audio Interface 1 (AIF1).
A powerful digital mixing core allows data from each TDM channel of each audio interface and from the ADCs and digital MICs to be mixed and re-routed back to a different audio interface and to the 4
DAC output channels. The digital mixing core can operate synchronously with either Audio Interface 1 or Audio Interface 2, with asynchronous stereo full duplex sample rate conversion performed on the other audio interface as required.
The system clock (SYSCLK) provides clocking for the ADCs, DACs, DSP core, digital audio interface and other circuits. SYSCLK can be derived directly from one of the MCLK1 or MCLK2 pins or via one of two integrated FLLs, providing flexibility to support a wide range of clocking schemes, including self-clocking FLL modes. Typical portable system MCLK frequencies, and sample rates from 8kHz to
96kHz are all supported. A low frequency (eg. 32.768kHz) clock can be used as the input reference to the FLLs, providing further flexibility. Automatic configuration of the clocking circuits is available, derived from the sample rate and from the MCLK / SYSCLK ratio.
The WM8958 uses a standard 2-wire control interface, providing full software control of all features, together with device register readback. An integrated Control Write Sequencer enables automatic scheduling of control sequences; commonly-used signal configurations may be selected using readyprogrammed sequences, including time-optimised control of the WM8958 pop suppression features. It is an ideal partner for a wide range of industry standard microprocessors, controllers and DSPs.
Unused circuitry can be disabled under software control, in order to save power; low leakage currents enable extended standby/off time in portable battery-powered applications.
Versatile GPIO functionality is provided, with support for button/accessory detect inputs, or for clock, system status, or programmable logic level output for control of additional external circuitry. Interrupt logic, status readback and de-bouncing options are supported within this functionality. w PP, August 2012, Rev 3.4
42
Pre-Production
WM8958
ANALOGUE INPUT SIGNAL PATH
The WM8958 has eight highly flexible analogue input channels, configurable in a large number of combinations:
1. Up to four fully differential or single-ended microphone inputs
2. Up to eight mono line inputs or 4 stereo line inputs
3. A dedicated mono differential input from external voice CODEC
These inputs may be mixed together or independently routed to different combinations of output drivers. An internal record path is provided at the input mixers to allow DAC output to be mixed with the input signal path (e.g. for voice call recording).
The WM8958 input signal paths and control registers are illustrated in Figure 8.
Figure 8 Control Registers for Input Signal Path w PP, August 2012, Rev 3.4
43
WM8958
Pre-Production
MICROPHONE INPUTS
Up to four analogue microphones can be connected to the WM8958, either in single-ended or differential mode. A dedicated PGA is provided for each microphone input. Two low noise microphone bias circuits are provided, reducing the need for external components.
For single-ended microphone inputs, the microphone signal is connected to the inverting input of the
PGAs (IN1LN, IN2LN, IN1RN or IN2RN). The non-inverting inputs of the PGAs are internally connected to VMID in this configuration. The non-inverting input pins IN1LP, IN2LP, IN1RP and
IN2RP are free to be used as line connections to the input or output mixers in this configuration.
For differential microphone inputs, the non-inverted microphone signal is connected to the noninverting input of the PGAs (IN1LP, IN2LP, IN1RP or IN2RP), whilst the inverted (or ‘noisy ground’) signal is connected to the inverting input pins (IN1LN, IN2LN, IN1RN and IN2RN).
The gain of the input PGAs is controlled via register settings, as defined in Table 4. Note that the input impedance of both inverting and non-inverting inputs changes with the input PGA gain setting, as described under “Electrical Characteristics”. See also the “Applications Information” for details of input resistance at all PGA Gain settings.
The microphone input configurations are illustrated in Figure 9 and Figure 10. Note that any PGA input pin that is used in either microphone configuration is not available for use as a line input path at the same time.
Figure 9 Single-Ended Microphone Input Figure 10 Differential Microphone Input w
MICROPHONE BIAS CONTROL
There are two MICBIAS generators which provide low noise reference voltages suitable for powering silicon (MEMS) microphones or biasing electret condenser (ECM) type microphones via an external resistor. Refer to the “Applications Information” section for recommended external components.
The MICBIAS outputs can be independently enabled using the MICB1_ENA and MICB2_ENA register bits. Under default conditions, a smooth pop-free profile of the MICBIAS outputs is implemented when MICB1_ENA or MICB2_ENA is enabled or disabled; a faster transition can be selected by setting the MICB1_RATE and MICB2_RATE registers as described in Table 1.
When a MICBIAS output is disabled, the output pin can be configured to be floating or to be actively discharged. This is selected using the MICB1_DISCH and MICB2_DISCH register bits.
The MICBIAS generators can each operate as a voltage regulator or in bypass mode.
In Regulator mode, the output voltage is selected using the MICB1_LVL and MICB2_LVL register bits. In this mode, AVDD1 must be at least 200mV greater than the required MICBIAS output voltages. The MICBIAS outputs are powered from the AVDD1 supply pin, and use the internal bandgap circuit as a reference.
Note that, in Regulator mode, the MICBIAS regulators are designed to operate without external decoupling capacitors. It is important that parasitic capacitances on the MICBIAS1 or MICBIAS2 pins do not exceed the specified limit in Regulator mode (see “Electrical Characteristics”).
In Bypass mode, the output pin (MICBIAS1 or MICBIAS2) is connected directly to AVDD1. This enables a low power operating state. Note that, if a capacitive load is connected to MICBIAS1 or
MICBIAS2 (eg. for a digital microphone supply), then the respective MICBIAS generator must be configured in Bypass mode.
PP, August 2012, Rev 3.4
44
Pre-Production
The MICBIAS configuration is illustrated in Figure 11.
WM8958 w
Figure 11 MICBIAS Generator
REGISTER
ADDRESS
R1
(0001h)
Power
Managem ent (1)
5 MICB2_ENA
4 MICB1_ENA
R61
(003Dh)
MICBIAS
1
5 MICB1_RATE
0
0
1
4 MICB1_MODE 1
0 MICB1_DISCH 1
R62
(003Eh)
MICBIAS
2
5 MICB2_RATE 1
4 MICB2_MODE 1
DESCRIPTION
Microphone Bias 2 Enable
0 = Disabled
1 = Enabled
Microphone Bias 1 Enable
0 = Disabled
1 = Enabled
Microphone Bias 1 Rate
0 = Fast start-up / shut-down
1 = Pop-free start-up / shut-down
Microphone Bias 1 Mode
0 = Regulator mode
1 = Bypass mode
Microphone Bias 1 Voltage Control
(when MICB1_MODE = 0)
000 = 1.5V
001 = 1.8V
010 = 1.9V
011 = 2.0V
100 = 2.2V
101 = 2.4V
110 = 2.5V
111 = 2.6V
Microphone Bias 1 Discharge
0 = MICBIAS1 floating when disabled
1 = MICBIAS1 discharged when disabled
Microphone Bias 2 Rate
0 = Fast start-up / shut-down
1 = Pop-free start-up / shut-down
Microphone Bias 2 Mode
0 = Regulator mode
1 = Bypass mode
PP, August 2012, Rev 3.4
45
WM8958 w
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
0 MICB2_DISCH 1
Microphone Bias 2 Voltage Control
(when MICB2_MODE = 0)
000 = 1.5V
001 = 1.8V
010 = 1.9V
011 = 2.0V
100 = 2.2V
101 = 2.4V
110 = 2.5V
111 = 2.6V
Microphone Bias 2 Discharge
0 = MICBIAS2 floating when disabled
1 = MICBIAS2 discharged when disabled
Table 1 Microphone Bias Control
Note that the maximum source current capability for MICBIAS1 and MICBIAS2 is 2.4mA each in
Regulator mode. The external biasing resistance must be large enough to limit each MICBIAS current to 2.4mA across the full microphone impedance range. The maximum source current for MICBIAS1 and MICBIAS2 is 3.6mA each in Bypass mode, as described in the “Electrical Characteristics”.
MICROPHONE ACCESSORY DETECT
The WM8958 provides a microphone detection function, which uses impedance measurement to detect one or more different external accessory connections. This feature is described in the “External
Accessory Detection” section.
LINE AND VOICE CODEC INPUTS
All eight analogue input pins may be used as line inputs. Each line input has different signal path options, providing flexibility, high performance and low power consumption for many different usage modes.
IN1LN and IN1RN can operate as single-ended line inputs to the input PGAs IN1L and IN1R respectively. These inputs provide a high gain path if required for low input signal levels.
IN2LN and IN2RN can operate as single-ended line inputs to the input PGAs IN2L and IN2R respectively, providing further high gain signal paths. These pins can also be connected to either of the output mixers MIXOUTL and MIXOUTR.
IN1LP and IN1RP can operate as single-ended line inputs to the input mixers MIXINL and MIXINR, or to the speaker mixers SPKMIXL and SPKMIXR. These signal paths enable power consumption to be reduced, by allowing the input PGAs and other circuits to be disabled if not required.
IN2LP/VRXN and IN2RP/VRXP can operate in three different ways:
Mono differential ’RXVOICE’ input (e.g. from an external voice CODEC) to the input mixers
MIXINL and MIXINR.
Single-ended line inputs to either of the output mixers MIXOUTL and MIXOUTR.
Ultra-low power mono differential ‘Direct Voice’ input (e.g. from an external voice CODEC) to the ear speaker driver on HPOUT2, or to either of the speaker drivers on SPKOUTL and
SPKOUTR.
Signal path configuration to the input PGAs and input mixers is detailed later in this section. Signal path configuration to the output mixers and speaker mixers is described in “Analogue Output Signal
Path”.
PP, August 2012, Rev 3.4
46
Pre-Production
WM8958
The line input and voice CODEC input configurations are illustrated in Figure 12 through to Figure 15.
MIXOUTL/R
Line Input
IN2LN,
IN2RN
-
+
PGA
MIXINL/R
Figure 12 IN1LN or IN1RN as Line Inputs
VMID
Figure 13 IN2LN or IN2RN as Line Inputs w
Figure 14 IN1LP or IN1RP as Line Inputs Figure 15 IN2LP or IN2RP as Line Inputs
INPUT PGA ENABLE
The Input PGAs are enabled using register bits IN1L_ENA, IN2L_ENA, IN1R_ENA and IN2R_ENA, as described in Table 2. The Input PGAs must be enabled for microphone input on the respective input pins, or for line input on the inverting input pins IN1LN, IN1RN, IN2LN, IN2RN.
REGISTER
ADDRESS
DESCRIPTION
R2 (0002h)
Power
Management
(2)
7
6
5
4
IN2L_ENA
IN1L_ENA
IN2R_ENA
IN1R_ENA
0
0
0
0
IN2L Input PGA Enable
0 = Disabled
1 = Enabled
IN1L Input PGA Enable
0 = Disabled
1 = Enabled
IN2R Input PGA Enable
0 = Disabled
1 = Enabled
IN1R Input PGA Enable
0 = Disabled
1 = Enabled
Table 2 Input PGA Enable
For normal operation of the input PGAs, the reference voltage VMID and the bias current must also be enabled. See “Reference Voltages and Master Bias” for details of the associated controls
VMID_SEL and BIAS_ENA.
PP, August 2012, Rev 3.4
47
WM8958 w
Pre-Production
INPUT PGA CONFIGURATION
Each of the Input PGAs can operate in a single-ended or differential mode. In differential mode, both inputs to the PGA are connected to the input source. In single-ended mode, the non-inverting input to the PGA must be connected to VMID. Configuration of the PGA inputs to the WM8958 input pins is controlled using the register bits shown in Table 3.
Single-ended microphone operation is configured by connecting the input source to the inverting input of the applicable PGA. The non-inverting input of the PGA must be connected to the buffered VMID reference. Note that the buffered VMID reference must be enabled, using the VMID_BUF_ENA register, as described in “Reference Voltages and Master Bias”.
Differential microphone operation is configured by connecting the input source to both inputs of the applicable PGA.
Line inputs to the input pins IN1LN, IN2LN, IN1RN and IN2RN must be connected to the applicable
PGA. The non-inverting input of the PGA must be connected to VMID.
Line inputs to the input pins IN1LP, IN2LP, IN1RP or IN2RP do not connect to the input PGAs. The non-inverting inputs of the associated PGAs must be connected to VMID. The inverting inputs of the associated PGAs may be used as separate mic/line inputs if required.
The maximum available attenuation on any of these input paths is achieved by using register bits shown in Table 3 to disconnect the input pins from the applicable PGA.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R40 (0028h)
Input Mixer
(2)
7
6
5
4
3
2
1
0
IN2LP_TO_IN2L
IN2LN_TO_IN2L
IN1LP_TO_IN1L
IN1LN_TO_IN1L
IN2RP_TO_IN2R
IN2RN_TO_IN2R
IN1RP_TO_IN1R
IN1RN_TO_IN1R
0
0
0
0
0
0
0
0
IN2L PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN2LP
Note that VMID_BUF_ENA must be set when using IN2L connected to
VMID.
IN2L PGA Inverting Input Select
0 = Not connected
1 = Connected to IN2LN
IN1L PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN1LP
Note that VMID_BUF_ENA must be set when using IN1L connected to
VMID.
IN1L PGA Inverting Input Select
0 = Not connected
1 = Connected to IN1LN
IN2R PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN2RP
Note that VMID_BUF_ENA must be set when using IN2R connected to
VMID.
IN2R PGA Inverting Input Select
0 = Not connected
1 = Connected to IN2RN
IN1R PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN1RP
Note that VMID_BUF_ENA must be set when using IN1R connected to
VMID.
IN1R PGA Inverting Input Select
0 = Not connected
1 = Connected to IN1RN
Table 3 Input PGA Configuration
PP, August 2012, Rev 3.4
48
Pre-Production
WM8958
INPUT PGA VOLUME CONTROL
Each of the four Input PGAs has an independently controlled gain range of -16.5dB to +30dB in 1.5dB steps. The gains on the inverting and non-inverting inputs to the PGAs are always equal. Each Input
PGA can be independently muted using the PGA mute bits as described in Table 4, with maximum mute attenuation achieved by simultaneously disconnecting the corresponding inputs described in
Table 3.
Note that, under default conditions (following power-up or software reset), the PGA mute register bits are set to ‘1’, but the mute functions will only become effective after the respective bit has been toggled to ‘0’ and then back to ‘1’. The Input PGAs will be un-muted (Mute disabled) after power-up or software reset, regardless of the readback value of the respective PGA mute bits.
To prevent "zipper noise", a zero-cross function is provided on the input PGAs. When this feature is enabled, volume updates will not take place until a zero-crossing is detected. In the case of a long period without zero-crossings, a timeout function is provided. When the zero-cross function is enabled, the volume will update after the timeout period if no earlier zero-cross has occurred. The timeout clock is enabled using TOCLK_ENA, the timeout period is set by TOCLK_DIV. See “Clocking and Sample Rates” for more information on these fields.
The IN1_VU and IN2_VU bits control the loading of the input PGA volume data. When IN1_VU and
IN2_VU are set to 0, the PGA volume data will be loaded into the respective control register, but will not actually change the gain setting. The IN1L and IN1R volume settings are both updated when a 1 is written to IN1_VU; the IN2L and IN2R volume settings are both updated when a 1 is written to
IN2_VU. This makes it possible to update the gain of the left and right signal paths simultaneously. w PP, August 2012, Rev 3.4
49
WM8958 w
Pre-Production
The Input PGA Volume Control register fields are described in Table 4 and Table 5.
REGISTER
ADDRESS
R24 (0018h)
Left Line Input
1&2 Volume
R25 (0019h)
Left Line Input
3&4 Volume
R26 (001Ah)
Right Line
Input 1&2
Volume
R27 (001Bh)
Right Line
Input 3&4
Volume
BIT LABEL DEFAULT
8
7
6
4:0
8
7
6
4:0
8
7
6
4:0
8
7
6
4:0
IN1_VU
IN1L_MUTE
IN1L_ZC
IN1L_VOL
[4:0]
IN2_VU
IN2L_MUTE
IN2L_ZC
IN2L_VOL
[4:0]
IN1_VU
IN1R_MUTE
IN1R_ZC
IN1R_VOL
[4:0]
IN2_VU
IN2R_MUTE
IN2R_ZC
IN2R_VOL
[4:0]
N/A
1
0
01011
(0dB)
N/A
1
0
01011
(0dB)
N/A
1
0
01011
(0dB)
N/A
1
0
01011
(0dB)
DESCRIPTION
Input PGA Volume Update
Writing a 1 to this bit will cause IN1L and
IN1R input PGA volumes to be updated simultaneously
IN1L PGA Mute
0 = Disable Mute
1 = Enable Mute
IN1L PGA Zero Cross Detector
0 = Change gain immediately
1 = Change gain on zero cross only
IN1L Volume
-16.5dB to +30dB in 1.5dB steps
(See Table 5 for volume range)
Input PGA Volume Update
Writing a 1 to this bit will cause IN2L and
IN2R input PGA volumes to be updated simultaneously
IN2L PGA Mute
0 = Disable Mute
1 = Enable Mute
IN2L PGA Zero Cross Detector
0 = Change gain immediately
1 = Change gain on zero cross only
IN2L Volume
-16.5dB to +30dB in 1.5dB steps
(See Table 5 for volume range)
Input PGA Volume Update
Writing a 1 to this bit will cause IN1L and
IN1R input PGA volumes to be updated simultaneously
IN1R PGA Mute
0 = Disable Mute
1 = Enable Mute
IN1R PGA Zero Cross Detector
0 = Change gain immediately
1 = Change gain on zero cross only
IN1R Volume
-16.5dB to +30dB in 1.5dB steps
(See Table 5 for volume range)
Input PGA Volume Update
Writing a 1 to this bit will cause IN2L and
IN2R input PGA volumes to be updated simultaneously
IN2R PGA Mute
0 = Disable Mute
1 = Enable Mute
IN2R PGA Zero Cross Detector
0 = Change gain immediately
1 = Change gain on zero cross only
IN2R Volume
-16.5dB to +30dB in 1.5dB steps
(See Table 5 for volume range)
Table 4 Input PGA Volume Control
PP, August 2012, Rev 3.4
50
Pre-Production
IN1L_VOL[4:0], IN2L_VOL[4:0],
IN1R_VOL[4:0], IN2R_VOL[4:0]
VOLUME
(dB)
00000 -16.5
00001 -15.0
00010 -13.5
00011 -12.0
00100 -10.5
00101 -9.0
00110 -7.5
00111 -6.0
01000 -4.5
01001 -3.0
01010 -1.5
01011 0
01100 +1.5
01101 +3.0
01110 +4.5
01111 +6.0
10000 +7.5
10001 +9.0
10010 +10.5
10011 +12.0
10100 +13.5
10101 +15.0
10110 +16.5
10111 +18.0
11000 +19.5
11001 +21.0
11010 +22.5
11011 +24.0
11100 +25.5
11101 +27.0
11110 +28.5
11111 +30.0
Table 5 Input PGA Volume Range
WM8958 w PP, August 2012, Rev 3.4
51
WM8958 w
Pre-Production
INPUT MIXER ENABLE
The WM8958 has two analogue input mixers which allow the Input PGAs and Line Inputs to be combined in a number of ways and output to the ADCs, Output Mixers, or directly to the output drivers via bypass paths.
The input mixers MIXINL and MIXINR are enabled by the MIXINL_ENA and MIXINR_ENA register bits, as described in Table 6. These control bits also enable the RXVOICE input path, described in the following section.
For normal operation of the input mixers, the reference voltage VMID and the bias current must also be enabled. See “Reference Voltages and Master Bias” for details of the associated controls
VMID_SEL and BIAS_ENA.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R2 (0002h)
Power
Management
(2)
9
8
MIXINL_ENA
MIXINR_ENA
0
0
Left Input Mixer Enable
(Enables MIXINL and RXVOICE input to
MIXINL)
0 = Disabled
1 = Enabled
Right Input Mixer Enable
(Enables MIXINR and RXVOICE input to
MIXINR)
0 = Disabled
1 = Enabled
Table 6 Input Mixer Enable
INPUT MIXER CONFIGURATION AND VOLUME CONTROL
The left and right channel input mixers MIXINL and MIXINR can be configured to take input from up to five sources:
1. IN1L or IN1R Input PGA
2. IN2L or IN2R Input PGA
3. IN1LP or IN1RP pin (PGA bypass)
4. RXVOICE mono differential input from IN2LP/VRXN and IN2RP/VRXP
5. MIXOUTL or MIXOUTR Output Mixer (Record path)
The Input Mixer configuration and volume controls are described in Table 7 for the Left input mixer
(MIXINL) and Table 8 for the Right input mixer (MIXINR). The signal levels from the Input PGAs may be set to Mute, 0dB or 30dB boost. Gain controls for the PGA bypass, RXVOICE and Record paths provide adjustment from -12dB to +6dB in 3dB steps.
When using the IN1LP or IN1RP signal paths direct to the input mixers (PGA bypass paths), a signal gain of +15dB can be selected using the IN1RP_MIXINR_BOOST or IN1LP_MIXINL_BOOST register bits. See Table 7 and Table 8 for further details.
When using the IN1LP or IN1RP signal paths direct to the input mixers (PGA bypass paths), the buffered VMID reference must be enabled, using the VMID_BUF_ENA register, as described in
“Reference Voltages and Master Bias”.
To prevent pop noise, it is recommended that gain and mute controls for the input mixers are not modified while the signal paths are active. If volume control is required on these signal paths, it is recommended that this is implemented using the input PGA volume controls or the ADC volume controls. The ADC volume controls are described in the “Analogue to Digital Converter (ADC)” section.
PP, August 2012, Rev 3.4
52
Pre-Production
REGISTER
ADDRESS
R21
(0015h)
Input Mixer
(1)
R41
(0029h)
Input Mixer
(3)
R43
(002Bh)
Input Mixer
(5)
WM8958
BIT LABEL DEFAULT DESCRIPTION
7
8
7
5
4
IN1LP_MIXINL_BOOST
IN2L_TO_MIXINL
IN2L_MIXINL_VOL
IN1L_TO_MIXINL
IN1L_MIXINL_VOL
2:0 MIXOUTL_MIXINL_VOL
[2:0]
8:6 IN1LP_MIXINL_VOL
[2:0]
0
0
0
0
0
000
(Mute)
000
(Mute)
IN1LP Pin (PGA Bypass) to
MIXINL Gain Boost.
This bit selects the maximum gain setting of the IN1LP_MIXINL_VOL register.
0 = Maximum gain is +6dB
1 = Maximum gain is +15dB
IN2L PGA Output to MIXINL Mute
0 = Mute
1 = Un-Mute
IN2L PGA Output to MIXINL Gain
0 = 0dB
1 = +30dB
IN1L PGA Output to MIXINL Mute
0 = Mute
1 = Un-Mute
IN1L PGA Output to MIXINL Gain
0 = 0dB
1 = +30dB
Record Path MIXOUTL to MIXINL
Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB
IN1LP Pin (PGA Bypass) to
MIXINL Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB (see note below).
When IN1LP_MIXINL_BOOST is set, then the maximum gain setting is increased to +15dB, ie.
111 = +15dB.
Note that VMID_BUF_ENA must be set when using the IN1LP (PGA
Bypass) input to MIXINL. w PP, August 2012, Rev 3.4
53
WM8958
REGISTER
ADDRESS
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
2:0 IN2LRP_MIXINL_VOL
[2:0]
000
(Mute)
RXVOICE Differential Input
(VRXP-VRXN) to MIXINL Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB
Table 7 Left Input Mixer (MIXINL) Volume Control
REGISTER
ADDRESS
R21 (0015h)
Input Mixer
(1)
8 IN1RP_MIXINR_BOOST 0
DESCRIPTION
R42 (002A)
Input Mixer
(4)
8
7
5
4
2:0
IN2R_TO_MIXINR
IN2R_MIXINR_VOL
IN1R_TO_MIXINR
IN1R_MIXINR_VOL
MIXOUTR_MIXINR_VOL
[2:0]
0
0
0
0
000
(Mute)
IN1RP Pin (PGA Bypass) to
MIXINR Gain Boost.
This bit selects the maximum gain setting of the
IN1RP_MIXINR_VOL register.
0 = Maximum gain is +6dB
1 = Maximum gain is +15dB
IN2R PGA Output to MIXINR Mute
0 = Mute
1 = Un-Mute
IN2R PGA Output to MIXINR Gain
0 = 0dB
1 = +30dB
IN1R PGA Output to MIXINR Mute
0 = Mute
1 = Un-Mute
IN1R PGA Output to MIXINR Gain
0 = 0dB
1 = +30dB
Record Path MIXOUTR to MIXINR
Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB w PP, August 2012, Rev 3.4
54
Pre-Production
REGISTER
ADDRESS
R44
(002Ch)
Input Mixer
(6)
8:6
2:0
IN1RP_MIXINR_VOL
[2:0]
IN2LRP_MIXINR_VOL
[2:0]
Table 8 Right Input Mixer (MIXINR) Volume Control
WM8958
DESCRIPTION
000
(Mute)
000
(Mute)
IN1RP Pin (PGA Bypass) to
MIXINR Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB (see note below).
When IN1RP_MIXINR_BOOST is set, then the maximum gain setting is increased to +15dB, ie.
111 = +15dB.
Note that VMID_BUF_ENA must be set when using the IN1RP
(PGA Bypass) input to MIXINR.
RXVOICE Differential Input
(VRXP-VRXN) to MIXINR Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB w PP, August 2012, Rev 3.4
55
WM8958
Pre-Production
DIGITAL MICROPHONE INTERFACE
The WM8958 supports a four-channel digital microphone interface. Two channels of audio data are multiplexed on the DMICDAT1 pin and a further two channels are multiplexed on the DMICDAT2 pin.
All four channels are clocked using the DMICCLK output pin.
The DMICDAT1 function is shared with the IN2LN pin; the analogue signal paths from IN2LN cannot be used when this pin is used for DMICDAT1 digital microphone input.
The DMICDAT2 function is shared with the IN2RN pin; the analogue signal paths from IN2RN cannot be used when this pin is used for DMICDAT2 digital microphone input.
The digital microphone interface is referenced to the MICBIAS1 voltage domain; the MICBIAS1 output must be enabled (MICB1_ENA = 1) when using the digital microphone interface.
The MICBIAS1 generator is suitable for use as a low noise supply for the digital microphones. Note that, if the capacitive load on the MICBIAS1 generator exceeds the specified limit (eg. due to a decoupling capacitor or long PCB trace), then the MICBIAS1 generator must be configured in Bypass mode. See “Analogue Input Signal Path” for details of the MICBIAS1 generator.
When digital microphone input is enabled, the WM8958 outputs a clock signal on the DMICCLK pin.
A pair of digital microphones is connected as illustrated in Figure 16. The microphones must be configured to ensure that the Left mic transmits a data bit when DMICCLK is high, and the Right mic transmits a data bit when DMICCLK is low. The WM8958 samples the digital microphone data at the end of each DMICCLK phase. Each microphone must tri-state its data output when the other microphone is transmitting. w
Figure 16 Digital Microphone Input
The DMICDAT1 digital microphone channels are enabled using DMIC1L_ENA and DMIC1R_ENA.
When these signal paths are enabled, the respective ADC path is disconnected and the digital microphone data is routed to the digital mixing input bus, as illustrated in “Digital Mixing”.
The DMICDAT2 digital microphone channels are enabled using DMIC2L_ENA and DMIC2R_ENA.
When these signal paths are enabled, the digital microphone data is routed to the digital mixing input bus, as illustrated in “Digital Mixing”.
Two microphone channels are interleaved on DMICDAT1; another two channels are interleaved on
DMICDAT2. The timing is illustrated in Figure 17. Each microphone must tri-state its data output when the other microphone is transmitting.
PP, August 2012, Rev 3.4
56
Pre-Production
WM8958 w
Figure 17 Digital Microphone Interface Timing
The four digital microphone channels can be routed to one of the four timeslots on AIF1. The
DMICDAT1 microphones, when enabled, are routed to the Left/Right channels of AIF1 Timeslot 0.
The DMICDAT2 microphones, when enabled, are routed to the Left/Right channels of AIF1 Timeslot
1.
Digital volume control of the digital microphone channels in the AIF1 signal paths is provided using the registers described in the “Digital Volume and Filter Control” section.
The digital microphone channels can be routed, in a limited number of configurations, to the digital mixing output bus, via the digital sidetone signal paths. See “Digital Mixing” for further details.
Digital volume control of the digital microphone channels in the digital sidetone signal paths is provided using the registers described in the “Digital Mixing” section.
The digital microphone interface control fields are described in Table 9.
REGISTER
ADDRESS
R4 (0004h)
Power
Management
(4)
5 DMIC2L_ENA 0
4
3
2
DMIC2R_ENA
DMIC1L_ENA
DMIC1R_ENA
0
0
0
DESCRIPTION
Digital microphone DMICDAT2
Left channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT2
Right channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT1
Left channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT1
Right channel enable
0 = Disabled
1 = Enabled
Table 9 Digital Microphone Interface Control
PP, August 2012, Rev 3.4
57
WM8958 w
Pre-Production
Clocking for the Digital Microphone interface is derived from SYSCLK. The DMICCLK frequency is configured automatically, according to the AIFn_SR, AIFnCLK_RATE and ADC_OSR128 registers.
(See “Clocking and Sample Rates” for further details of the system clocks and control registers.)
The DMICCLK is enabled whenever a digital microphone input path is enabled on the DMICDAT1 or
DMICDAT2 pin(s). Note that the SYSDSPCLK_ENA register must also be set.
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then the DMICCLK frequency is controlled by the AIF1_SR and AIF1CLK_RATE registers.
When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then the DMICCLK frequency is controlled by the AIF2_SR and AIF2CLK_RATE registers.
The DMICCLK frequency is as described in Table 10 (for ADC_OSR128=1) and Table 11 (for
ADC_OSR128=0). The ADC_OSR128 bit is set by default, giving best audio performance. Note that the only valid DMICCLK configurations are the ones listed in Table 10 and Table 11.
The applicable clocks (SYSCLK, and AIF1CLK or AIF2CLK) must be present and enabled when using the digital microphone interface.
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
2.048
2.8224
3.072
2.8224
3.072
2.048 2.048
2.8224 2.8224 22.05
24
32
44.1
48
3.072 3.072
2.048
2.8224
3.072
88.2
96
Note that, when ADC_OSR128=1, digital microphone operation is only supported for the above
DMICCLK configurations.
Table 10 DMICCLK Frequency (MHz) - ADC_OSR128 = 1 (Default)
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
1.024 1.024
1.4112 1.4112 1.4112
1.536 1.536 1.536
1.024 1.024 1.024 1.024
1.4112 1.4112 1.4112
1.536 1.536 1.536
22.05
24
32
44.1
48
88.2
2.8224
3.072
96
Note that, when ADC_OSR128=0, digital microphone operation is only supported for the above
DMICCLK configurations.
Table 11 DMICCLK Frequency (MHz) - ADC_OSR128 = 0
PP, August 2012, Rev 3.4
58
Pre-Production
WM8958
DIGITAL PULL-UP AND PULL-DOWN
The WM8958 provides integrated pull-up and pull-down resistors on the DMICDAT1 and DMICDAT2 pins. This provides a flexible capability for interfacing with other devices. Each of the pull-up and pulldown resistors can be configured independently using the register bits described in Table 12.
Note that, if the DMICDAT1 or DMICDAT2 digital microphone channels are disabled, or if
DMICDATn_PU and DMICDATn_PD are both set, then the pull-up and pull-down will be disabled on the respective pin.
DESCRIPTION REGISTER
ADDRESS
R1824
(0720h)
Pull Control
(1)
11
10
9
8
DMICDAT2_PU
DMICDAT2_PD
DMICDAT1_PU
DMICDAT1_PD
0
0
0
0
DMICDAT2 Pull-Up enable
0 = Disabled
1 = Enabled
DMICDAT2 Pull-Down enable
0 = Disabled
1 = Enabled
DMICDAT1 Pull-Up enable
0 = Disabled
1 = Enabled
DMICDAT1 Pull-Down enable
0 = Disabled
1 = Enabled
Table 12 Digital Pull-Up and Pull-Down Control
ANALOGUE TO DIGITAL CONVERTER (ADC)
The WM8958 uses stereo 24-bit sigma-delta ADCs. The use of multi-bit feedback and high oversampling rates reduces the effects of jitter and high frequency noise. The oversample rate can be adjusted, if required, to reduce power consumption - see “Clocking and Sample Rates” for details.
The ADC full scale input level is proportional to AVDD1 - see “Electrical Characteristics”. Any input signal greater than full scale may overload the ADC and cause distortion.
The ADCs are enabled by the ADCL_ENA and ADCR_ENA register bits.
REGISTER
ADDRESS
R4 (0004h)
Power
Management (4)
BIT LABEL DEFAULT
1
0
ADCL_ENA
ADCR_ENA
0
0
DESCRIPTION
Left ADC Enable
0 = Disabled
1 = Enabled
Right ADC Enable
0 = Disabled
1 = Enabled
Table 13 ADC Enable Control
The outputs of the ADCs can be routed to the Left/Right channels of AIF1 (Timeslot 0).
Digital volume control of the ADC outputs in the AIF1 signal paths is provided using the registers described in the “Digital Volume and Filter Control” section.
The outputs of the ADCs can be routed, in a limited number of configurations, to the digital mixing output bus, via the digital sidetone signal paths. See “Digital Mixing” for further details.
Digital volume control of the ADC outputs in the digital sidetone signal paths is provided using the registers described in the “Digital Mixing” section. w PP, August 2012, Rev 3.4
59
WM8958 w
Pre-Production
ADC CLOCKING CONTROL
Clocking for the ADCs is derived from SYSCLK. The required clock is enabled when the
SYSDSPCLK_ENA register is set.
The ADC clock rate is configured automatically, according to the AIFn_SR, AIFnCLK_RATE and
ADC_OSR128 registers. (See “Clocking and Sample Rates” for further details of the system clocks and control registers.)
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then the ADC clocking is controlled by the AIF1_SR and AIF1CLK_RATE registers.
When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then the ADC clocking is controlled by the AIF2_SR and AIF2CLK_RATE registers.
The supported ADC clocking configurations are described in Table 14 (for ADC_OSR128=1) and
Table 15 (for ADC_OSR128=0). The ADC_OSR128 bit is set by default, giving best audio performance.
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
22.05
24
32
44.1
48
88.2
96
When ADC_OSR128=1, ADC operation is only supported for the configurations indicated above
Table 14 ADC Clocking - ADC_OSR128 = 1 (Default)
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
22.05
24
32
44.1
48
88.2
96
When ADC_OSR128=0, ADC operation is only supported for the configurations indicated above
Table 15 ADC Clocking - ADC_OSR128 = 0
The clocking requirements in Table 14 and Table 15 are only applicable to the AIFnCLK that is selected as the SYSCLK source. Note that both clocks (AIF1CLK and AIF2CLK) must satisfy the requirements noted in the “Clocking and Sample Rates” section.
The applicable clocks (SYSCLK, and AIF1CLK or AIF2CLK) must be present and enabled when using the Analogue to Digital Converters (ADCs).
PP, August 2012, Rev 3.4
60
Pre-Production
WM8958
DIGITAL CORE ARCHITECTURE
The WM8958 Digital Core provides an extensive set of mixing and signal processing features. The
Digital Core Architecture is illustrated in Figure 18, which also identifies the datasheet sections applicable to each portion of the Digital Core.
Audio Interface 1 (AIF1) supports audio input and output on two stereo timeslots simultaneously, making a total of four inputs and four outputs. The mixing of the four AIF1 output paths is described in
“Audio Interface 1 (AIF1) Output Mixing”.
A digital mixing path from the ADCs or Digital Microphones to the DAC output paths provides a high quality sidetone for voice calls or other applications. The sidetone configuration is described in “Digital
Sidetone Mixing”; the associated filter and volume control is described in “Digital Sidetone Volume and Filter Control”.
Each of the four hi-fi DACs has a dedicated mixer for controlling the signal paths to that DAC. The configuration of these signal paths is described in “DAC Output Digital Mixing”.
Each DAC is provided with digital volume control, soft mute / un-mute and a low pass filter. The associated controls are defined in the “Digital to Analogue Converter (DAC)” section.
Digital processing can be applied to the four input channels of AIF1 and the two input channels of
AIF2. The available features include multiband compression (MBC), 5-band equalization (EQ), 3D stereo expansion and dynamic range control (DRC).
The MBC provides a function to maximise the loudness of the audio signal, using independent compression and boost of different frequency bands without overdriving the loudspeakers. The RMS
Limiter within the MBC function enables the maximum signal level to be matched to the application requirements and/or power rating of the loudspeaker. The MBC controls are described in “Multiband
Compressor”. The EQ provides the capability to tailor the audio path according to the frequency characteristics of an earpiece or loudspeaker, and/or according to user preferences. The EQ controls are described in “ReTune TM Mobile Parametric Equalizer (EQ)”.
The DRC provides adaptive signal level control to improve the handling of unpredictable signal levels and to improve intelligibility in the presence of transients and impulsive noises. The DRC controls are described in “Dynamic Range Control (DRC)”. 3D stereo expansion provides a stereo enhancement effect; the depth of the effect is programmable, as described in “3D Stereo Expansion”.
The input channels of AIF1 and AIF2 are also equipped with digital volume control and soft mute / unmute control; see “Digital Volume and Filter Control” for details of these features.
The output channels of AIF1 and AIF2 can be configured using the digital volume control and a programmable high-pass filter (HPF). The Dynamic Range Control (DRC) circuit can also be applied here, with the restriction that a DRC cannot be enabled in the input and output path of one AIF channel at the same time. The AIF output volume and filter controls are described in “Digital Volume and Filter Control”.
The WM8958 provides an ultrasonic mode on the output paths of AIF1, allowing high frequency signals (such as ultrasonic microphone signals) to be output. See “Ultrasonic (4FS) AIF Output Mode” for further details.
The WM8958 provides two full audio interfaces, AIF1 and AIF2. Each interface supports a number of protocols, including I 2 S, DSP, MSB-first left/right justified, and can operate in master or slave modes.
PCM operation is supported in the DSP mode. A-law and -law companding are also supported. Time division multiplexing (TDM) is available to allow multiple devices to stream data simultaneously on the same bus, saving space and power.
Four-channel input and output is supported using TDM on AIF1. Two-channel input and output is supported on AIF2. A third interface, AIF3, is partially supported, using multiplexers to re-configure alternate connections to AIF1 or AIF2.
Signal mixing between audio interfaces is possible. The WM8958 performs stereo full-duplex sample rate conversion between the audio interfaces as required. (Note that sample rate conversion is not supported on some signal paths, as noted in Figure 18.)
The audio interfaces AIF1, AIF2 and AIF3 are referenced to DBVDD1, DBVDD2 and DBVDD3 respectively; this provides additional capability to interface between different sub-systems within an application. w PP, August 2012, Rev 3.4
61
WM8958
DMICDAT2
DMICDAT1
ADC
L
ADC
R
Digital Sidetone
Mixing
Pre-Production
Digital Sidetone Volume and Filter Control
DAC Output
Digital Mixing
G G
G G
G G
[Code]
Gain Codes
V = Full volume control
(-71.625dB to 12dB, 0.375dB steps for DAC
-71.625dB to 17.625dB, 0.375dB steps for ADC/MICs)
S = Softmute/un-mute
NG = Digital Noise Gate
G = Fixed gain control
(-36dB to 0dB, 3dB steps)
+
+
+
+
DAC1L
Vol
VS
DAC1R
Vol
VS
DAC2L
Vol
VS
DAC2R
Vol
VS
DAC
1L
DAC
1R
DAC
2L
DAC
2R
G G
DAC Digital
Volume
Audio Interface 1
(AIF1) Output Mixing
Sample Rate
Conversion
= SRC not supported
DRC / Microphone signal activity detector (GPIO)
Ultrasonic (4FS)
AIF Output Modes
Digital Audio
Interface Control
+ + + +
DRC DRC
MBC
DRC
EQ
3D
MBC
DRC
EQ
3D
AIF1 Slot 0
AIF1 Slot 1 fs / 4fs Select
AIF1 Slot 0
AIF1 Slot 1
MBC
DRC
EQ
3D
DRC
Left / Right source select / Mono Mix control
Left / Right source select /
Mono Mix control
0R 0L 1R 1L 0R 0L 1R 1L
DIGITAL AUDIO
INTERFACE 1 (AIF1)
0R 0L 0R 0L
DIGITAL AUDIO
INTERFACE 2 (AIF2)
MONO PCM
INTERFACE
Multiband Compressor (MBC) /
Dynamic Range Control (DRC) /
Retune Mobile Parameter Equalizer (EQ) /
Stereo 3D Expansion /
Digital Volume and Filter Control
Note the Multi-band Compressor (MBC) cannot be enabled on AIF1 and AIF2 simultaneously.
Note the Dynamic Range Control (DRC) cannot be enabled in the input and output paths of any
Digital Audio Interface simultaneously.
Note -
AIF1 is referenced to the DBVDD1 power domain
AIF2 is referenced to DBVDD2
AIF3 / Mono PCM is referenced to DBVDD3
Figure 18 Digital Core Architecture w PP, August 2012, Rev 3.4
62
Pre-Production
DIGITAL MIXING
WM8958
This section describes the digital mixing functions of the WM8958.
Digital audio mixing is provided on four AIF1 output paths, two digital sidetone paths, and four Digital to Analogue converters (DACs).
Note that the two AIF2 output paths are connected to the DAC2L and DAC2R signal paths.
The digital mixing functions and associated control registers are illustrated in Figure 19. w
Figure 19 Digital Mixing Block Diagram
PP, August 2012, Rev 3.4
63
WM8958
Pre-Production
AUDIO INTERFACE 1 (AIF1) OUTPUT MIXING
There are four AIF1 digital mixers, one for each AIF1 audio channel (ie. Left/Right channels on
Timeslots 0/1). The inputs to each AIF1 mixer comprise signals from the ADC / Digital Microphone inputs and from AIF2.
Note that the Left/Right channels of AIF1 can be inverted or interchanged if required; see “Digital
Audio Interface Control”.
The AIF1 Left Timeslot 0 output channel is derived from the ADCL / DMIC1 (Left) and AIF2 (Left) inputs. The ADCL / DMIC1 (Left) path is enabled by ADC1L_TO_AIF1ADC1L, whilst the AIF2 (Left) path is enabled by AIF2DACL_TO_AIF1ADC1L.
The AIF1 Right Timeslot 0 output channel is derived from the ADCR / DMIC1 (Right) and AIF2 (Right) inputs. The ADCR / DMIC1 (Right) path is enabled by ADC1R_TO_AIF1ADC1R, whilst the AIF2
(Right) path is enabled by AIF2DACR_TO_AIF1ADC1R.
The AIF1 Left Timeslot 1 output channel is derived from the DMIC2 (Left) and AIF2 (Left) inputs. The
DMIC2 (Left) path is enabled by ADC2L_TO_AIF1ADC2L, whilst the AIF2 (Left) path is enabled by
AIF2DACL_TO_AIF1ADC2L. w
The AIF1 Right Timeslot 1 output channel is derived from the DMIC2 (Right) and AIF2 (Right) inputs.
The DMIC2 (Right) path is enabled by ADC2R_TO_AIF1ADC2R, whilst the AIF2 (Right) path is enabled by AIF2DACR_TO_AIF1ADC2R.
The AIF1 output mixer controls are defined in Table 16.
REGISTER
ADDRESS
R1542 (0606h)
AIF1 ADC1
Left Mixer
Routing
1 ADC1L_TO_AIF
1ADC1L
0
0 AIF2DACL_TO_
AIF1ADC1L
0
R1543 (0607h)
AIF1 ADC1
Right Mixer
Routing
R1544 (0608h)
AIF1 ADC2
Left Mixer
Routing
R1545 (0609h)
AIF1 ADC2
Right Mixer
Routing
1
0
1
0
1
0
ADC1R_TO_AIF
1ADC1R
AIF2DACR_TO_
AIF1ADC1R
ADC2L_TO_AIF
1ADC2L
AIF2DACL_TO_
AIF1ADC2L
ADC2R_TO_AIF
1ADC2R
AIF2DACR_TO_
AIF1ADC2R
0
0
0
0
0
0
DESCRIPTION
Enable ADCL / DMIC1 (Left) to
AIF1 (Timeslot 0, Left) output
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to AIF1 (Timeslot
0, Left) output
0 = Disabled
1 = Enabled
Enable ADCR / DMIC1 (Right) to
AIF1 (Timeslot 0, Right) output
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to AIF1
(Timeslot 0, Right) output
0 = Disabled
1 = Enabled
Enable DMIC2 (Left) to AIF1
(Timeslot 1, Left) output
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to AIF1 (Timeslot
1, Left) output
0 = Disabled
1 = Enabled
Enable DMIC2 (Right) to AIF1
(Timeslot 1, Right) output
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to AIF1
(Timeslot 1, Right) output
0 = Disabled
1 = Enabled
Table 16 AIF1 Output Mixing
PP, August 2012, Rev 3.4
64
Pre-Production
WM8958
DIGITAL SIDETONE MIXING
There are two digital sidetone signal paths, STL and STR. The sidetone sources are selectable for each path. The sidetone mixer outputs are inputs to the DAC signal mixers.
The following sources can be selected for sidetone path STL.
ADCL or DMICDAT1 (Left) channel
DMICDAT2 (Left) channel
The following sources can be selected for sidetone path STR.
ADCR or DMICDAT1 (Right) channel
DMICDAT2 (Right) channel
The sidetone signal sources are selected using STR_SEL and STL_SEL as described in Table 17.
Note that, when STR_SEL = 0 or STL_SEL = 0, and the respective ADC is enabled (for analogue inputs), then the ADC data will be selected for applicable sidetone path.
DESCRIPTION REGISTER
ADDRESS
R1569 (0621h)
Sidetone
1 STR_SEL 0
0 STL_SEL 0
Select source for sidetone STR path
0 = ADCR / DMICDAT1 (Right)
1 = DMICDAT2 (Right)
Select source for sidetone STL path
0 = ADCL / DMICDAT1 (Left)
1 = DMICDAT2 (Left)
Table 17 Digital Sidetone Mixing
DIGITAL SIDETONE VOLUME AND FILTER CONTROL
A digital volume control is provided for the digital sidetone paths. The associated register controls are described in Table 18.
A digital high-pass filter can be enabled in the sidetone paths to remove DC offsets. This filter is enabled using the ST_HPF register bit; the cut-off frequency is configured using ST_HPF_CUT.
When the filter is enabled, it is enabled in both digital sidetone paths.
Note that the sidetone filter cut-off frequency scales according to the sample rate of AIF1 or AIF2.
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then the ST_HPF cut-off frequency is scaled according to the AIF1_SR register. When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then the ST_HPF cut-off frequency is scaled according to the AIF2_SR register. See “Clocking and Sample Rates” for further details of the system clocks and control registers.
DESCRIPTION REGISTER
ADDRESS
R1536 (0600h)
DAC1 Mixer
Volumes
8:5 ADCR_DAC1_V
OL [3:0]
0000 Sidetone STR to DAC1L and
DAC1R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
(see Table 19 for gain range) w PP, August 2012, Rev 3.4
65
WM8958
REGISTER
ADDRESS
R1539 (0603h)
DAC2 Mixer
Volumes
R1569 (0621h)
Sidetone
Pre-Production
DESCRIPTION
3:0
8:5
3:0
9:7
6
ADCL_DAC1_V
OL [3:0]
ADCR_DAC2_V
OL [3:0]
ADCL_DAC2_V
OL [3:0]
ST_HPF_CUT
[2:0]
ST_HPF
0000
0000
0000
000
0
Sidetone STL to DAC1L and
DAC1R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
(see Table 19 for gain range)
Sidetone STR to DAC2L and
DAC2R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
(see Table 19 for gain range)
Sidetone STL to DAC2L and
DAC2R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
(see Table 19 for gain range)
Sidetone HPF cut-off frequency
(relative to 44.1kHz sample rate)
000 = 2.7kHz
001 = 1.35kHz
010 = 675Hz
011 = 370Hz
100 = 180Hz
101 = 90Hz
110 = 45Hz
111 = Reserved
Note - the cut-off frequencies scale with the Digital Mixing (SYSCLK) clocking rate. The quoted figures apply to 44.1kHz sample rate.
Digital Sidetone HPF Select
0 = Disabled
1 = Enabled
Table 18 Digital Sidetone Volume Control w PP, August 2012, Rev 3.4
66
Pre-Production
WM8958
ADCR_DAC1_VOL,
ADCL_DAC2_VOL,
ADCR_DAC1_VOL or
ADCL_DAC2_VOL
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
1100
1101
1110
1111
SIDETONE
GAIN (dB)
-12
-9
-6
-3
0
0
0
0
-24
-21
-18
-15
-36
-33
-30
-27
Table 19 Digital Sidetone Volume Range
DAC OUTPUT DIGITAL MIXING
There are four DAC digital mixers, one for each DAC. The inputs to each DAC mixer comprise signals from AIF1, AIF2 and the digital sidetone signals. w
Note that the Left/Right channels of the AIF1 and AIF2 inputs can be inverted or interchanged if required; see “Digital Audio Interface Control”.
DESCRIPTION REGISTER
ADDRESS
R1537 (0601h)
DAC1 Left
Mixer Routing
R1538 (0602h)
DAC1 Right
Mixer Routing
5
4
2
1
0
5
4
2
ADCR_TO_DAC
1L
ADCL_TO_DAC
1L
AIF2DACL_TO_
DAC1L
AIF1DAC2L_TO
_DAC1L
AIF1DAC1L_TO
_DAC1L
ADCR_TO_DAC
1R
ADCL_TO_DAC
1R
AIF2DACR_TO_
DAC1R
0
0
0
0
0
0
0
0
Enable Sidetone STR to DAC1L
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC1L
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to DAC1L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Left) to
DAC1L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Left) to
DAC1L
0 = Disabled
1 = Enabled
Enable Sidetone STR to DAC1R
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC1R
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to DAC1R
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
67
WM8958 w
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
R1540 (0604h)
DAC2 Left
Mixer Routing
R1541 (0605h)
DAC2 Right
Mixer Routing
1
0
5
4
2
1
0
5
4
2
1
0
AIF1DAC2R_TO
_DAC1R
AIF1DAC1R_TO
_DAC1R
ADCR_TO_DAC
2L
ADCL_TO_DAC
2L
AIF2DACL_TO_
DAC2L
AIF1DAC2L_TO
_DAC2L
AIF1DAC1L_TO
_DAC2L
ADCR_TO_DAC
2R
ADCL_TO_DAC
2R
AIF2DACR_TO_
DAC2R
AIF1DAC2R_TO
_DAC2R
AIF1DAC1R_TO
_DAC2R
0
0
0
0
0
0
0
0
0
0
0
0
Enable AIF1 (Timeslot 1, Right) to
DAC1R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Right) to
DAC1R
0 = Disabled
1 = Enabled
Enable Sidetone STR to DAC2L
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC2L
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to DAC2L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Left) to
DAC2L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Left) to
DAC2L
0 = Disabled
1 = Enabled
Enable Sidetone STR to DAC2R
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC2R
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to DAC2R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Right) to
DAC2R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Right) to
DAC2R
0 = Disabled
1 = Enabled
Table 20 DAC Output Digital Mixing
AUDIO INTERFACE 2 (AIF2) DIGITAL MIXING
There are two output channels on AIF2. The audio source for these two channels is the same as the selected source for DAC2L and DAC2R, as described in “DAC Output Digital Mixing”.
Note that the Left/Right channels of AIF2 can be inverted or interchanged if required; see “Digital
Audio Interface Control”.
PP, August 2012, Rev 3.4
68
Pre-Production
WM8958
ULTRASONIC (4FS) AIF OUTPUT MODE
The WM8958 provides an ultrasonic mode on the output paths of the AIF1 audio interface. The ultrasonic mode enables high frequency signals (such as ultrasonic microphone signals) to be output.
Ultrasonic mode is enabled on AIF1 using the AIF1ADC_4FS register bit. When the ultrasonic mode is selected, the AIF1 output sample rate is increased by a factor of 4. For example, a 48kHz sample rate will be output at 192kHz in ultrasonic mode.
Ultrasonic mode is only supported in AIF Master mode and uses the ADCLRCLK output (not the
LRCLK). When ultrasonic mode is enabled, the AIF1 must be configured in Master mode, as described in “Digital Audio Interface Control”. See “General Purpose Input/Output” to configure the
GPIO1 pin as ADCLRCLK1. The ADCLRCLK1 rate is controlled as described in “Digital Audio
Interface Control”.
When ultrasonic mode is enabled, the audio band filtering and digital volume controls (see “Digital
Volume and Filter Control”) are bypassed on the affected output paths.
The Dynamic Range Control (DRC) function is not available on the AIF1 output signal paths in ultrasonic mode. Note, however, that the DRC is still available on the AIF input paths in this case.
The ultrasonic (4FS) signal paths are illustrated in Figure 20. The AIF1ADC_4FS register bit is defined in Table 21. w
Figure 20 Ultrasonic (4FS) Signal Paths
REGISTER
ADDRESS
R1040 (0410h)
AIF1 ADC1
Filters
15 AIF1ADC_4FS
Table 21 Ultrasonic (4FS) Mode Control
0
DESCRIPTION
Enable AIF1ADC ultrasonic mode
(4FS) output, bypassing all AIF1 baseband output filtering
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
69
WM8958
MULTIBAND COMPRESSOR (MBC)
Pre-Production
The Multiband Compressor (MBC) is a DSP function which can be enabled in the digital playback path of the WM8958 audio interfaces. The normal function of the MBC is to maximise the loudness of the audio signal.
The MBC uses selective processing of the received digital audio signal to control the loudness; independent gain control algorithms are applied to different audio frequency bands. The effect of this is to increase the perceived loudness of the audio path, producing an enhanced audio signal without overdriving the output transducers (eg. loudspeakers).
The MBC provides two internal signal paths, allowing low frequencies and high frequencies to be processed separately. Each signal path incorporates a programmable compressor which can be used to dynamically control the level of each frequency band.
The most significant advantage of multiband compression is that a signal peak in one frequency band will not cause gain reduction in the other frequency band. Similarly, gain can be applied to one frequency band without boosting (and potentially distorting) the other. This provides a powerful capability to maximise the loudness of the signal path.
The two signal paths are re-combined at the output of the MBC. If necessary, any difference in tonal balance between the frequency bands can be restored by changing the levels of the two signal paths relative to each other.
The MBC incorporates a high-pass and a low-pass filter, which set the lower and upper frequency limits of the MBC signal path. The crossover frequency that divides the two frequency bands can be adjusted according to the system requirements. The attack and decay times of the compressors are separately programmable on each frequency band.
An RMS Limiter is included within the MBC function. This is a signal limiter that responds to the RMS output level of the digital playback path, allowing the maximum signal level to be matched to the application requirements and/or the power rating of the loudspeaker.
The WM8958 provides one stereo Multiband Compressor (MBC). The MBC can be enabled on the input path of AIF1 timeslot 0, AIF1 timeslot 1 or on the input path of AIF2. Note that the MBC cannot be enabled on more than one of these paths simultaneously.
A Dynamic Range Control (DRC) function is also available on the digital playback paths. Note that the
DRC and MBC functions should not be enabled simultaneously on the same playback path. The DRC is enabled using the registers described in Table 28.
The MBC signal paths and control parameters are illustrated in Figure 21. w
Figure 21 Multiband Compressor
PP, August 2012, Rev 3.4
70
Pre-Production
The MBC filter cut-off frequencies are shown in Table 22.
WM8958
Low Cut-Off Frequency
Crossover Frequency
High Cut-Off Frequency
20Hz 350Hz 1kHz
500Hz 2.5kHz 7.9kHz
11kHz 12kHz 16kHz
Table 22 Multiband Compressor Cut-Off Frequencies
RMS LIMITER
An RMS Limiter is included within the MBC function. This is a signal limiter that responds to the RMS output level of the digital playback path, allowing the maximum signal level to be matched to the application requirements and/or the power rating of the loudspeaker.
The Wolfson WISCE™ software must be used to derive the register settings for the RMS Limiter. The
WISCE™ software allows users to select the desired RMS voltage level of the analogue output.
Note that the selected RMS voltage level applies to each output pin. For differential (BTL) outputs, note that a limit of 1.0Vrms on each pin equates to 2.0Vrms across the load.
The MBC operates within the digital core of the WM8958, and the playback signal may be subject to boost or attenuation in the digital and/or analogue stages of the output signal path. Therefore, the
RMS Limiter configuration must take account of the applicable gain settings of the output signal path.
The WISCE™ software allows the user to input the amount of gain (dB) applicable to the relevant output signal path. (This is the total signal gain of the applicable DACs, output/boost mixers and PGA volume controls.) Note that the register settings for the RMS Limiter will only be valid for the specified gain level.
MBC CLOCKING CONTROL
Clocking for the MBC is derived from DSP2CLK. This clock is derived from the output of the
AIF1CLK_SRC or AIF2CLK_SRC multiplexers, according to the SYSCLK_SRC register. This is illustrated in Figure 22, and described further in the “Clocking and Sample Rates” section.
Figure 22 Audio Interface Clock Control
The MBC can enabled on the AIF1 or AIF2 input paths, regardless of the SYSCLK_SRC setting, provided that the minimum clocking requirement for the MBC is satisfied. w PP, August 2012, Rev 3.4
71
WM8958 w
Pre-Production
To support the MBC function, it is required that DSP2CLK ≥ 256 x fs (where fs is the sample rate of the AIF on which the MBC is enabled). When the MBC is enabled on either of the AIF1 input paths, it is required that DSP2CLK ≥ 256 x AIF1_SR; when the MBC is enabled on the AIF2 input path, it is required that DSP2CLK ≥ 256 x AIF2_SR.
The MBC is supported in 44.1kHz and 48kHz AIF sample rate modes only; note that these modes require clocking rates of AIFnCLK = 256 x fs. (See “Digital to Analogue Converter (DAC)” for details of the valid clocking rates.)
The DSP2CLK clock for the MBC is enabled when the DSP2CLK_ENA register is set. The DSP2CLK clock is required for running the MBC function, and also for accessing any of the MBC configuration registers. Note that the applicable source clock must also be present when using DSP2CLK.
See “Clocking and Sample Rates” for details of the WM8958 clocking control registers.
MBC CONTROL SEQUENCES
Specific control sequences must be followed when enabling or configuring the MBC function; these sequences are described in Table 23 to Table 26. The associated MBC control registers are described in Table 27.
Note that the WM8958 is provided with a working set of default MBC configuration parameters, allowing the MBC feature to be enabled easily in a default operating mode. For user-specific configuration of the MBC, the Wolfson WISCE™ software must be used to derive the configuration parameters (refer to WISCE TM for further information).
The control sequence for enabling the MBC is described in Table 23 (for default MBC settings) and
Table 24 (for user-specific MBC settings). It is recommended that the applicable DAC playback path is muted during this sequence, as described below.
STEP DESCRIPTION
1 Mute the applicable DAC output(s)
NOTES
The DAC Volume and Mute controls are described in Table 58.
2
3
Set DSP2CLK_ENA = 1
Set DSP2_ENA = 1
4
5
6
Write DSP2_RUNR = 1
Set MBC_SEL as required
Set MBC_ENA = 1
Un-mute the applicable DAC output(s)
For AIF1DAC1 path, set MBC_SEL = 00.
For AIF1DAC2 path, set MBC_SEL = 01.
For AIF2DAC path, set MBC_SEL = 10.
Table 23 MBC Enable Sequence (default MBC configuration)
6
7
8
9
STEP DESCRIPTION
1 Mute the applicable DAC output(s)
2
3
Set DSP2CLK_ENA = 1
Set DSP2_ENA = 1
NOTES
The DAC Volume and Mute controls are described in Table 58.
4
5
Set Register R2568 (0A08h) = 007Bh
Set Register R2569 (0A09h) = 0007h
Set Register R2570 (0A0Ah) = 0073h
Set the configuration parameters
Write DSP2_RUNR = 1
Set MBC_SEL as required
Refer to WISCE™ for register settings.
10
Set MBC_ENA = 1
Un-mute the applicable DAC output(s)
For AIF1DAC1 path, set MBC_SEL = 00.
For AIF1DAC2 path, set MBC_SEL = 01.
For AIF2DAC path, set MBC_SEL = 10.
Table 24 MBC Enable Sequence (user-specific MBC configuration)
PP, August 2012, Rev 3.4
72
Pre-Production
WM8958
The control sequence for disabling the MBC is described in Table 25. It is recommended that the applicable DAC playback path is muted during this sequence, as described below.
2
3
4
5
STEP DESCRIPTION
1 Mute the applicable DAC output(s)
NOTES
The DAC Volume and Mute controls are described in Table 58.
Set MBC_ENA = 0
Un-mute the applicable DAC output(s)
Set DSP2_ENA = 0
Set DSP2CLK_ENA = 0
Table 25 MBC Disable Sequence
The control sequence for updating the MBC configuration parameters is described in Table 26. It is recommended that the applicable DAC playback paths are muted during this sequence, as described below.
The same sequence is required when reading the MBC configuration parameters; note that readback of these registers is not possible when the MBC function is active.
STEP DESCRIPTION
1 Mute the applicable DAC output(s)
NOTES
The DAC Volume and Mute controls are described in Table 58. (If changing the MBC from one signal path to another, then both
DAC paths should be muted.)
2
3
4
Set MBC_ENA = 0
Write DSP2_STOP = 1
Refer to WISCE™ for register settings.
5
Readback the MBC configuration parameters
Set Register R2568 (0A08h) = 007Bh
Set Register R2569 (0A09h) = 0007h
Set Register R2570 (0A0Ah) = 0073h
Set the configuration parameters
Refer to WISCE™ for register settings.
Note that these actions are only required for user-specific MBC configuration.
6
7
8
Write DSP2_RUNR = 1
Set MBC_SEL
Set MBC_ENA = 1
Un-mute the applicable DAC output(s)
For AIF1DAC1 path, set MBC_SEL = 00.
For AIF1DAC2 path, set MBC_SEL = 01.
For AIF2DAC path, set MBC_SEL = 10.
Table 26 MBC Update / Readback Sequence w PP, August 2012, Rev 3.4
73
WM8958
R2305 (0901h)
DSP2_Config
R2573
(0A0Dh)
DSP2_ExecCo ntrol
Pre-Production
The MBC control registers are described in Table 27.
REGISTER
ADDRESS
R2304 (0900h)
DSP2_Progra m
0 DSP2_ENA 0
5:4
0
2
1
MBC_SEL [1:0]
MBC_ENA
DSP2_STOP
DSP2_RUNR
00
0
0
0
DESCRIPTION
DSP2 Audio Processor Enable.
0 = Disabled
1 = Enabled
This bit must be set before the MBC is enabled. It must remain set whenever the MBC is enabled.
MBC Signal Path select
00 = AIF1DAC1 input path (AIF1,
Timeslot 0)
01 = AIF1DAC2 input path (AIF1,
Timeslot 1)
10 = AIF2DAC input path
11 = Reserved
MBC Enable
0 = Disabled
1 = Enabled
Stop the DSP2 audio processor
Writing a 1 to this bit will cause the
DSP2 processor to stop processing audio data
Start the DSP2 audio processor
Writing a 1 to this bit will cause the
DSP2 processor to start processing audio data
Table 27 Multiband Compressor (MBC) Control w PP, August 2012, Rev 3.4
74
Pre-Production
WM8958
DYNAMIC RANGE CONTROL (DRC)
The Dynamic Range Control (DRC) is a circuit which can be enabled in the digital playback or digital record paths of the WM8958 audio interfaces. The function of the DRC is to adjust the signal gain in conditions where the input amplitude is unknown or varies over a wide range, e.g. when recording from microphones built into a handheld system.
The DRC can apply Compression and Automatic Level Control to the signal path. It incorporates ‘anticlip’ and ‘quick release’ features for handling transients in order to improve intelligibility in the presence of loud impulsive noises.
The DRC also incorporates a Noise Gate function, which provides additional attenuation of very lowlevel input signals. This means that the signal path is quiet when no signal is present, giving an improvement in background noise level under these conditions.
The WM8958 provides three stereo Dynamic Range Controllers (DRCs); these are associated with
AIF1 timeslot 0, AIF1 timeslot 1 and AIF2 respectively. Each DRC can be enabled either in the DAC playback (AIF input) path or in the ADC record (AIF output) path, as described in the “Digital Core
Architecture” section.
The DRCs are enabled in the DAC or ADCs audio signal paths using the register bits described in
Table 28. Note that enabling any DRC in the DAC and ADC paths simultaneously is an invalid selection.
A Multiband Compressor (MBC) function is also available on the digital playback paths. Note that the
DRC and MBC functions should not be enabled simultaneously on the same playback path. The MBC control registers are described in Table 27.
When the DRC is enabled in any of the ADC (digital record) paths, the associated High Pass Filter
(HPF) must be enabled also; this ensures that DC offsets are removed prior to the DRC processing.
The output path HPF control registers are described in Table 42 (for AIF1 output paths) and Table 50
(for AIF2 output paths). These are described in the “Digital Volume and Filter Control” section.
Note that, when ultrasonic (4FS) mode is selected on AIF1, then the DRC function is bypassed on the respective ADC (output) signal paths. The DRC may still be selected on the AIF1 DAC (input) signal paths.
DESCRIPTION REGISTER
ADDRESS
R1088 (0440h)
AIF1 DRC1 (1)
2 AIF1DAC1_DRC
_ENA
0
R1104 (0450h)
AIF1 DRC2 (1)
1
0
2
1
0
AIF1ADC1L_DR
C_ENA
AIF1ADC1R_DR
C_ENA
AIF1DAC2_DRC
_ENA
AIF1ADC2L_DR
C_ENA
AIF1ADC2R_DR
C_ENA
0
0
0
0
0
Enable DRC in AIF1DAC1 playback path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC1 (Left) record path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC1 (Right) record path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable DRC in AIF1DAC2 playback path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC2 (Left) record path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC2 (Right) record path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
75
WM8958
REGISTER
ADDRESS
R1344 (0540h)
AIF2 DRC (1)
Pre-Production
2
1
0
AIF2DAC_DRC_
ENA
AIF2ADCL_DRC
_ENA
AIF2ADCR_DRC
_ENA
0
0
0
DESCRIPTION
Enable DRC in AIF2DAC playback path
0 = Disabled
1 = Enabled
Enable DRC in AIF2ADC (Left) record path
0 = Disabled
1 = Enabled
Enable DRC in AIF2ADC (Right) record path
0 = Disabled
1 = Enabled
Table 28 DRC Enable
The following description of the DRC is applicable to all three DRCs. The associated register control fields are described in Table 30, Table 31 and Table 32 for the respective DRCs.
Note that, where the following description refers to register names, the generic prefix [DRC] is quoted:
For the DRC associated with AIF1 timeslot 0, [DRC] = AIF1DRC1.
For the DRC associated with AIF1 timeslot 1, [DRC] = AIF1DRC2.
For the DRC associated with AIF2, [DRC] = AIF2DRC.
DRC COMPRESSION / EXPANSION / LIMITING
The DRC supports two different compression regions, separated by a “Knee” at a specific input amplitude. In the region above the knee, the compression slope [DRC]_HI_COMP applies; in the region below the knee, the compression slope [DRC]_LO_COMP applies.
The DRC also supports a noise gate region, where low-level input signals are heavily attenuated. This function can be enabled or disabled according to the application requirements. The DRC response in this region is defined by the expansion slope [DRC]_NG_EXP.
For additional attenuation of signals in the noise gate region, an additional “knee” can be defined
(shown as “Knee2” in Figure 23). When this knee is enabled, this introduces an infinitely steep dropoff in the DRC response pattern between the [DRC]_LO_COMP and [DRC]_NG_EXP regions.
The overall DRC compression characteristic in “steady state” (i.e. where the input amplitude is nearconstant) is illustrated in Figure 23. w PP, August 2012, Rev 3.4
76
Pre-Production
DRC Output Amplitude (dB)
(Y0)
[DRC]_KNEE_OP
[DRC]_KNEE2_OP
Knee2
Knee1
[DRC
]_HI_
COM
P
[DR
C]_
LO
_C
OM
P
WM8958
[DRC]_KNEE2_IP [DRC]_KNEE_IP 0dB
DRC Input Amplitude (dB)
Figure 23 DRC Response Characteristic
The slope of the DRC response is determined by register fields [DRC]_HI_COMP and
[DRC]_LO_COMP. A slope of 1 indicates constant gain in this region. A slope less than 1 represents compression (i.e. a change in input amplitude produces only a smaller change in output amplitude). A slope of 0 indicates that the target output amplitude is the same across a range of input amplitudes; this is infinite compression.
When the noise gate is enabled, the DRC response in this region is determined by the
[DRC]_NG_EXP register. A slope of 1 indicates constant gain in this region. A slope greater than 1 represents expansion (ie. a change in input amplitude produces a larger change in output amplitude).
When the DRC_KNEE2_OP knee is enabled (“Knee2” in Figure 23), this introduces the vertical line in the response pattern illustrated, resulting in infinitely steep attenuation at this point in the response.
The DRC parameters are listed in Table 29.
4
5
6
7
REF PARAMETER
1 [DRC]_KNEE_IP
DESCRIPTION
Input level at Knee1 (dB)
2
3
[DRC]_KNEE_OP
[DRC]_HI_COMP
Output level at Knee2 (dB)
Compression ratio above Knee1
[DRC]_LO_COMP
[DRC]_KNEE2_IP
[DRC]_NG_EXP
[DRC]_KNEE2_OP
Compression ratio below Knee1
Input level at Knee2 (dB)
Expansion ratio below Knee2
Output level at Knee2 (dB)
Table 29 DRC Response Parameters
The noise gate is enabled when the [DRC]_NG_ENA register is set. When the noise gate is not enabled, parameters 5, 6, 7 above are ignored, and the [DRC]_LO_COMP slope applies to all input signal levels below Knee1.
The DRC_KNEE2_OP knee is enabled when the [DRC]_KNEE2_OP_ENA register is set. When this bit is not set, then parameter 7 above is ignored, and the Knee2 position always coincides with the low end of the [DRC]_LO_COMP region. w PP, August 2012, Rev 3.4
77
WM8958
Pre-Production
The “Knee1” point in Figure 23 is determined by register fields [DRC]_KNEE_IP and
[DRC]_KNEE_OP.
Parameter Y0, the output level for a 0dB input, is not specified directly, but can be calculated from the other parameters, using the equation:
GAIN LIMITS
The minimum and maximum gain applied by the DRC is set by register fields [DRC]_MINGAIN,
[DRC]_MAXGAIN and [DRC]_NG_MINGAIN. These limits can be used to alter the DRC response from that illustrated in Figure 23. If the range between maximum and minimum gain is reduced, then the extent of the dynamic range control is reduced.
The minimum gain in the Compression regions of the DRC response is set by [DRC]_MINGAIN. The mimimum gain in the Noise Gate region is set by [DRC]_NG_MINGAIN. The minimum gain limit prevents excessive attenuation of the signal path.
The maximum gain limit set by [DRC]_MAXGAIN prevents quiet signals (or silence) from being excessively amplified. w PP, August 2012, Rev 3.4
78
Pre-Production
WM8958
DYNAMIC CHARACTERISTICS
The dynamic behaviour determines how quickly the DRC responds to changing signal levels. Note that the DRC responds to the average (RMS) signal amplitude over a period of time.
The [DRC]_ATK determines how quickly the DRC gain decreases when the signal amplitude is high.
The [DRC]_DCY determines how quickly the DRC gain increases when the signal amplitude is low.
These register fields are described in Table 30, Table 31 and Table 32. Note that the register defaults are suitable for general purpose microphone use.
ANTI-CLIP CONTROL
The DRC includes an Anti-Clip feature to avoid signal clipping when the input amplitude rises very quickly. This feature uses a feed-forward technique for early detection of a rising signal level. Signal clipping is avoided by dynamically increasing the gain attack rate when required. The Anti-Clip feature is enabled using the [DRC]_ANTICLIP bit.
Note that the feed-forward processing increases the latency in the input signal path.
Note that the Anti-Clip feature operates entirely in the digital domain. It cannot be used to prevent signal clipping in the analogue domain nor in the source signal. Analogue clipping can only be prevented by reducing the analogue signal gain or by adjusting the source signal.
Note that the Anti-Clip feature should not be enabled at the same time as the Quick Release feature
(described below) on the same DRC.
QUICK RELEASE CONTROL
The DRC includes a Quick-Release feature to handle short transient peaks that are not related to the intended source signal. For example, in handheld microphone recording, transient signal peaks sometimes occur due to user handling, key presses or accidental tapping against the microphone.
The Quick Release feature ensures that these transients do not cause the intended signal to be masked by the longer time constants of [DRC]_DCY.
The Quick-Release feature is enabled by setting the [DRC]_QR bit. When this bit is enabled, the DRC measures the crest factor (peak to RMS ratio) of the input signal. A high crest factor is indicative of a transient peak that may not be related to the intended source signal. If the crest factor exceeds the level set by [DRC]_QR_THR, then the normal decay rate ([DRC]_DCY) is ignored and a faster decay rate ([DRC]_QR_DCY) is used instead.
Note that the Quick Release feature should not be enabled at the same time as the Anti-Clip feature
(described above) on the same DRC.
SIGNAL ACTIVITY DETECT
The DRC incorporates a configurable signal detect function, allowing the signal level at the DRC input to be monitored and to be used to trigger other events. This can be used to detect the presence of a microphone signal on an ADC or digital mic channel, or can be used to detect an audio signal received over the digital audio interface.
The Peak signal level or the RMS signal level of the DRC input can be selected as the detection threshold. When the threshold condition is exceeded, an interrupt or GPIO output can be generated.
See “General Purpose Input/Output” for a full description of the applicable control fields. w PP, August 2012, Rev 3.4
79
WM8958 w
REGISTER
ADDRESS
R1088 (0440h)
AIF1 DRC1 (1)
Pre-Production
DRC REGISTER CONTROLS
The AIF1DRC1 control registers are described in Table 30. The AIF1DRC2 control registers are described in Table 31. The AIF2DRC control registers are described in Table 32.
8 AIF1DRC1_NG_
ENA
0
DESCRIPTION
R1089 (0441h)
AIF1 DRC1 (2)
5
4
3
12:9
8:5
4:2
AIF1DRC1_KNE
E2_OP_ENA
AIF1DRC1_QR
AIF1DRC1_ANTI
CLIP
AIF1DRC1_ATK
[3:0]
AIF1DRC1_DCY
[3:0]
AIF1DRC1_MIN
GAIN [2:0]
0
1
1
0100
0010
001
AIF1 DRC1 Noise Gate Enable
0 = Disabled
1 = Enabled
AIF1 DRC1 KNEE2_OP Enable
0 = Disabled
1 = Enabled
AIF1 DRC1 Quick-release Enable
0 = Disabled
1 = Enabled
AIF1 DRC1 Anti-clip Enable
0 = Disabled
1 = Enabled
AIF1 DRC1 Gain attack rate
(seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF1 DRC1 Gain decay rate
(seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
AIF1 DRC1 Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved
PP, August 2012, Rev 3.4
80
Pre-Production
REGISTER
ADDRESS
1:0 AIF1DRC1_MAX
GAIN [1:0]
R1090 (0442h)
AIF1 DRC1 (3)
15:12 AIF1DRC1_NG_
MINGAIN [3:0]
11:10 AIF1DRC1_NG_
EXP [1:0]
9:8 AIF1DRC1_QR_
THR [1:0]
7:6 AIF1DRC1_QR_
DCY [1:0]
5:3 AIF1DRC1_HI_C
OMP [2:0]
01
0000
00
00
00
000
WM8958
DESCRIPTION
AIF1 DRC1 Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
AIF1 DRC1 Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF1 DRC1 Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF1 DRC1 Quick-release threshold
(crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC1 Quick-release decay rate (seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved
AIF1 DRC1 Compressor slope
(upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved w PP, August 2012, Rev 3.4
81
WM8958
REGISTER
ADDRESS
R1091 (0443h)
AIF1 DRC1 (4)
R1092 (0444h)
AIF1 DRC1 (5)
Pre-Production
DESCRIPTION
2:0
10:5
4:0
9:5
4:0
AIF1DRC1_LO_
COMP [2:0]
AIF1DRC1_KNE
E_IP [5:0]
AIF1DRC1_KNE
E_OP [4:0]
AIF1DRC1_KNE
E2_IP [4:0]
AIF1DRC1_KNE
E2_OP [4:0]
000
000000
00000
00000
00000
AIF1 DRC1 Compressor slope
(lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
AIF1 DRC1 Input signal level at the
Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
AIF1 DRC1 Output signal at the
Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
AIF1 DRC1 Input signal level at the
Noise Gate threshold ‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when
AIF1DRC1_NG_ENA = 1.
AIF1 DRC1 Output signal at the
Noise Gate threshold ‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when
AIF1DRC1_KNEE2_OP_ENA = 1.
Table 30 AIF1 Timeslot 0 DRC Controls w PP, August 2012, Rev 3.4
82
Pre-Production
REGISTER
ADDRESS
R1104 (0450h)
AIF1 DRC2 (1)
8
5
4
3
4:2
AIF1DRC2_NG_
ENA
AIF1DRC2_KNE
E2_OP_ENA
AIF1DRC2_QR
AIF1DRC2_ANTI
CLIP
R1105 (0451h)
AIF1 DRC2 (2)
12:9 AIF1DRC2_ATK
[3:0]
8:5 AIF1DRC2_DCY
[3:0]
AIF1DRC2_MIN
GAIN [2:0]
0
0
1
1
0100
0010
001
WM8958
DESCRIPTION
AIF1 DRC2 Noise Gate Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 KNEE2_OP Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 Quick-release Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 Anti-clip Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 Gain attack rate
(seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF1 DRC2 Gain decay rate
(seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
AIF1 DRC2 Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved w PP, August 2012, Rev 3.4
83
WM8958
REGISTER
ADDRESS
R1106 (0452h)
AIF1 DRC2 (3)
Pre-Production
DESCRIPTION
1:0
15:12
11:10
9:8
7:6
5:3
AIF1DRC2_MAX
GAIN [1:0]
AIF1DRC2_NG_
MINGAIN [3:0]
AIF1DRC2_NG_
EXP [1:0]
AIF1DRC2_QR_
THR [1:0]
AIF1DRC2_QR_
DCY [1:0]
AIF1DRC2_HI_C
OMP [2:0]
01
0000
00
00
00
000
AIF1 DRC2 Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
AIF1 DRC2 Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF1 DRC2 Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF1 DRC2 Quick-release threshold
(crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC2 Quick-release decay rate (seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved
AIF1 DRC2 Compressor slope
(upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved w PP, August 2012, Rev 3.4
84
Pre-Production
REGISTER
ADDRESS
2:0 AIF1DRC2_LO_
COMP [2:0]
R1107 (0453h)
AIF1 DRC2 (4)
10:5 AIF1DRC2_KNE
E_IP [5:0]
4:0 AIF1DRC2_KNE
E_OP [4:0]
R1108 (0454h)
AIF1 DRC2 (5)
9:5 AIF1DRC2_KNE
E2_IP [4:0]
4:0 AIF1DRC2_KNE
E2_OP [4:0]
000
000000
00000
00000
00000
WM8958
DESCRIPTION
AIF1 DRC2 Compressor slope
(lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
AIF1 DRC2 Input signal level at the
Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
AIF1 DRC2 Output signal at the
Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
AIF1 DRC2 Input signal level at the
Noise Gate threshold ‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when
AIF1DRC2_NG_ENA = 1.
AIF1 DRC2 Output signal at the
Noise Gate threshold ‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when
AIF1DRC2_KNEE2_OP_ENA = 1.
Table 31 AIF1 Timeslot 1 DRC Controls w PP, August 2012, Rev 3.4
85
WM8958
Pre-Production
REGISTER
ADDRESS
R1344 (0540h)
AIF2 DRC (1)
8
5
4
3 AIF2DRC_ANTI
CLIP
R1345 (0541h)
AIF2 DRC (2)
12:9 AIF2DRC_ATK
[3:0]
8:5
4:2
AIF2DRC_NG_E
NA
AIF2DRC_KNEE
2_OP_ENA
AIF2DRC_QR
AIF2DRC_DCY
[3:0]
AIF2DRC_MING
AIN [2:0]
0
0
1
1
0100
0010
001
DESCRIPTION
AIF2 DRC Noise Gate Enable
0 = Disabled
1 = Enabled
AIF2 DRC KNEE2_OP Enable
0 = Disabled
1 = Enabled
AIF2 DRC Quick-release Enable
0 = Disabled
1 = Enabled
AIF2 DRC Anti-clip Enable
0 = Disabled
1 = Enabled
AIF2 DRC Gain attack rate
(seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF2 DRC Gain decay rate
(seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
AIF2 DRC Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved w PP, August 2012, Rev 3.4
86
Pre-Production
REGISTER
ADDRESS
1:0 AIF2DRC_MAX
GAIN [1:0]
R1346 (0542h)
AIF2 DRC (3)
15:12 AIF2DRC_NG_
MINGAIN [3:0]
11:10 AIF2DRC_NG_E
XP [1:0]
9:8 AIF2DRC_QR_T
HR [1:0]
7:6 AIF2DRC_QR_D
CY [1:0]
5:3 AIF2DRC_HI_C
OMP [2:0]
01
0000
00
00
00
000
WM8958
DESCRIPTION
AIF2 DRC Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
AIF2 DRC Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF2 DRC Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF2 DRC Quick-release threshold
(crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF2 DRC Quick-release decay rate
(seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved
AIF2 DRC Compressor slope
(upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved w PP, August 2012, Rev 3.4
87
WM8958
REGISTER
ADDRESS
R1347 (0543h)
AIF2 DRC (4)
R1348 (0544h)
AIF2 DRC (5)
Pre-Production
DESCRIPTION
2:0
10:5
4:0
9:5
4:0
AIF2DRC_LO_C
OMP [2:0]
AIF2DRC_KNEE
_IP [5:0]
AIF2DRC_KNEE
_OP [4:0]
AIF2DRC_KNEE
2_IP [4:0]
AIF2DRC_KNEE
2_OP [4:0]
000
000000
00000
00000
00000
AIF2 DRC Compressor slope (lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
AIF2 DRC Input signal level at the
Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
AIF2 DRC Output signal at the
Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
AIF2 DRC Input signal level at the
Noise Gate threshold ‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when
AIF2DRC_NG_ENA = 1.
AIF2 DRC Output signal at the
Noise Gate threshold ‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when
AIF2DRC_KNEE2_OP_ENA = 1.
Table 32 AIF2 DRC Controls w PP, August 2012, Rev 3.4
88
Pre-Production
WM8958
RETUNE TM MOBILE PARAMETRIC EQUALIZER (EQ)
The ReTune TM Mobile Parametric EQ is a circuit which can be enabled in the digital playback path of the WM8958 audio interfaces. The function of the EQ is to adjust the frequency characteristic of the output in order to compensate for unwanted frequency characteristics in the loudspeaker (or other output transducer). It can also be used to tailor the response according to user preferences, for example to accentuate or attenuate specific frequency bands to emulate different sound profiles or environments e.g. concert hall, rock etc.
The WM8958 provides three stereo EQ circuits; these are associated with AIF1 timeslot 0, AIF1 timeslot 1 and AIF2 respectively. The EQ is enabled in these three signal paths using the register bits described in Table 33.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1152 (0480h)
AIF1 DAC1
EQ Gains (1)
R1184
(04A0h)
AIF1 DAC2
EQ Gains (1)
R1408 (0580h)
AIF2 EQ Gains
(1)
0
0
0
AIF1DAC1_EQ_E
NA
AIF1DAC2_EQ_E
NA
AIF2DAC_EQ_EN
A
0
0
0
Enable EQ in AIF1DAC1 playback path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable EQ in AIF1DAC2 playback path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable EQ in AIF2DAC playback path
0 = Disabled
1 = Enabled
Table 33 ReTune TM Mobile Parametric EQ Enable
The following description of the EQ is applicable to all three EQ circuits. The associated register control fields are described in Table 35, Table 36 and Table 37 for the respective EQs. The EQ provides selective control of 5 frequency bands as described below.
The low frequency band (Band 1) filter can be configured either as a peak filter or a shelving filter.
When configured as a shelving filter, is provides adjustable gain below the Band 1 cut-off frequency.
As a peak filter, it provides adjustable gain within a defined frequency band that is centred on the
Band 1 frequency.
The mid frequency bands (Band 2, Band 3, Band 4) filters are peak filters, which provide adjustable gain around the respective centre frequency.
The high frequency band (Band 5) filter is a shelving filter, which provides adjustable gain above the
Band 5 cut-off frequency.
The EQ can be configured to operate in two modes - “Default” mode or “ReTune TM Mobile” mode. w PP, August 2012, Rev 3.4
89
WM8958
Pre-Production
DEFAULT MODE (5-BAND PARAMETRIC EQ)
In default mode, the cut-off / centre frequencies are fixed as per Table 34. The filter bandwidths are also fixed in default mode. The gain of the individual bands (-12dB to +12dB) can be controlled as described in Table 35.
The cut-off / centre frequencies noted in Table 34 are applicable to a sample rate of 48kHz. When using other sample rates, these frequencies will be scaled in proportion to the selected sample rate for the associated Audio Interface (AIF1 or AIF2).
If AIF1 and AIF2 are operating at different sample rates, then the cut-off / centre frequencies will be different for the two interfaces. Note that the frequencies can be set to other values by using the features described in “ReTune TM Mobile Mode”.
EQ BAND CUT-OFF/CENTRE FREQUENCY
Table 34 EQ Band Cut-off / Centre Frequencies
REGISTER
ADDRESS
R1152 (0480h)
AIF1 DAC1
EQ Gains (1)
15:11 AIF1DAC1_EQ
_B1_GAIN
[4:0]
01100
(0dB)
10:6 AIF1DAC1_EQ
_B2_GAIN
[4:0]
01100
(0dB)
01100
(0dB)
5:1 AIF1DAC1_EQ
_B3_GAIN
[4:0]
R1153 (0481h)
AIF1 DAC1
EQ Gains (2)
15:11 AIF1DAC1_EQ
_B4_GAIN
[4:0]
01100
(0dB)
10:6 AIF1DAC1_EQ
_B5_GAIN
[4:0]
0 AIF1DAC1_EQ
_MODE
01100
(0dB)
0
DESCRIPTION
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 1 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 2 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 3 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 4 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 5 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC1 (AIF1, Timeslot 0) EQ
Band 1 Mode
0 = Shelving filter
1 = Peak filter
Table 35 AIF1 Timeslot 0 EQ Band Gain Control w PP, August 2012, Rev 3.4
90
Pre-Production
REGISTER
ADDRESS
R1184
(04A0h)
AIF1 DAC2
EQ Gains (1)
R1185
(04A1h)
AIF1 DAC2
EQ Gains (2)
15:11 AIF1DAC2_EQ
_B1_GAIN
[4:0]
10:6 AIF1DAC2_EQ
_B2_GAIN
[4:0]
5:1 AIF1DAC2_EQ
_B3_GAIN
[4:0]
15:11 AIF1DAC2_EQ
_B4_GAIN
[4:0]
10:6 AIF1DAC2_EQ
_B5_GAIN
[4:0]
0 AIF1DAC2_EQ
_MODE
01100
(0dB)
01100
(0dB)
01100
(0dB)
01100
(0dB)
01100
(0dB)
0
WM8958
DESCRIPTION
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 1 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 2 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 3 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 4 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 5 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF1DAC2 (AIF1, Timeslot 1) EQ
Band 1 Mode
0 = Shelving filter
1 = Peak filter
Table 36 AIF1 Timeslot 1 EQ Band Gain Control
REGISTER
ADDRESS
R1408 (0580h)
AIF2 EQ Gains
(1)
R1409 (0581h)
AIF2 EQ Gains
(2)
15:11
10:6
5:1 AIF2DAC_EQ_
B3_GAIN
[4:0]
15:11 AIF2DAC_EQ_
B4_GAIN
[4:0]
10:6
0
AIF2DAC_EQ_
B1_GAIN
[4:0]
AIF2DAC_EQ_
B2_GAIN
[4:0]
AIF2DAC_EQ_
B5_GAIN
[4:0]
AIF2DAC_EQ_
MODE
01100
(0dB)
01100
(0dB)
01100
(0dB)
01100
(0dB)
01100
(0dB)
0
DESCRIPTION
AIF2 EQ Band 1 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF2 EQ Band 2 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF2 EQ Band 3 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF2 EQ Band 4 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF2 EQ Band 5 Gain
-12dB to +12dB in 1dB steps
(see Table 38 for gain range)
AIF2 EQ Band 1 Mode
0 = Shelving filter
1 = Peak filter
Table 37 AIF2 EQ Band Gain Control w PP, August 2012, Rev 3.4
91
WM8958
Pre-Production
EQ GAIN SETTING GAIN (dB)
00000 -12
00001 -11
00010 -10
00011 -9
00100 -8
00101 -7
00110 -6
00111 -5
01000 -4
01001 -3
01010 -2
01011 -1
01100 0
01101 +1
01110 +2
01111 +3
10000 +4
10001 +5
10010 +6
10011 +7
10100 +8
10101 +9
10110 +10
10111 +11
11000 +12
11001 to 11111 Reserved
Table 38 EQ Gain Control Range
RETUNE
TM
MOBILE MODE
ReTune TM Mobile mode provides a comprehensive facility for the user to define the cut-off/centre frequencies and filter bandwidth for each EQ band, in addition to the gain controls already described.
This enables the EQ to be accurately customised for a specific transducer characteristic or desired sound profile.
The EQ enable and EQ gain controls are the same as defined for the default mode. The additional coefficients used in ReTune TM Mobile mode are held in registers R1154 to R1172 for AIF1DAC1, registers R1186 to R1204 for AIF1DAC2 and registers R1410 to R1428 for AIF2. These coefficients are derived using tools provided in Wolfson’s WISCE™ evaluation board control software.
Please contact your local Wolfson representative for more details.
Note that the WM8958 audio interfaces may operate at different sample rates concurrently. The EQ settings for each interface must be programmed relative to the applicable sample rate of the corresponding audio interface. If the audio interface sample rate is changed, then different EQ register settings will be required to achieve a given EQ response. w PP, August 2012, Rev 3.4
92
Pre-Production
WM8958
EQ FILTER CHARACTERISTICS
The filter characteristics for each frequency band are shown in Figure 24 to Figure 28. These figures show the frequency response for all available gain settings, using default cut-off/centre frequencies and bandwidth. Note that EQ Band 1 can also be configured as a Peak Filter if required.
15
10
15
10
5 5
0
-5
0
-5
-10 -10
-15
1 10 100 1000
Frequency (Hz)
10000 100000
Figure 24 EQ Band 1 – Low Freq Shelf Filter Response
-15
1 10 100 1000
Frequency (Hz)
10000
Figure 25 EQ Band 2 – Peak Filter Response
100000
-5
-10
5
0
15
10
-15
1 10 100 1000
Frequency (Hz)
10000
Figure 26 EQ Band 3 – Peak Filter Response
15
10
5
0
-5
-10
-15
1 10 10000 100000 100 1000
Frequency (Hz)
Figure 28 EQ Band 5 – High Freq Shelf Filter Response w
15
10
5
0
-5
-10
100000
-15
1 10 100 1000
Frequency (Hz)
10000
Figure 27 EQ Band 4 – Peak Filter Response
100000
PP, August 2012, Rev 3.4
93
WM8958
3D STEREO EXPANSION
Pre-Production
The 3D Stereo Expansion is an audio enhancement feature which can be enabled in the digital playback path of the WM8958 audio interfaces. This feature uses configurable cross-talk mechanisms to adjust the depth or width of the stereo audio.
The WM8958 provides three 3D Stereo Expansion circuits; these are associated with AIF1 timeslot 0,
AIF1 timeslot 1 and AIF2 respectively. The 3D Stereo Expansion is enabled and controlled in these signal paths using the register bits described in Table 39.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1057 (0421h)
AIF1 DAC1
Filters (2)
R1059 (0423h)
AIF1 DAC2
Filters (2)
R1313 (0521h)
AIF2 DAC
Filters (2)
13:9
8
13:9
8
13:9
8
AIF1DAC1_3D_G
AIN
AIF1DAC1_3D_E
NA
AIF1DAC2_3D_G
AIN
AIF1DAC2_3D_E
NA
AIF2DAC_3D_GA
IN
AIF2DAC_3D_EN
A
00000
0
00000
0
00000
0
AIF1DAC1 playback path (AIF1,
Timeslot 0) 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
Enable 3D Stereo in AIF1DAC1 playback path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
AIF1DAC2 playback path (AIF1,
Timeslot 1) 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
Enable 3D Stereo in AIF1DAC2 playback path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
AIF2DAC playback path 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
Enable 3D Stereo in AIF2DAC playback path
0 = Disabled
1 = Enabled
Table 39 3D Stereo Expansion Control w PP, August 2012, Rev 3.4
94
Pre-Production
WM8958
DIGITAL VOLUME AND FILTER CONTROL
This section describes the digital volume and filter controls of the WM8958 AIF paths.
Digital volume control and High Pass Filter (HPF) control is provided on four AIF1 output (digital record) paths and two AIF2 output (digital record) paths.
Note that, when ultrasonic (4FS) mode is selected on AIF1, then the digital volume control and high pass filter (HPF) control are bypassed on the respective ADC (output) signal paths.
Digital volume, soft-mute and mono mix control is provided on four AIF1 input (digital playback) paths and two AIF2 input (digital playback) paths. A configurable noise gate function is available on each of the digital playback paths.
AIF1 - OUTPUT PATH VOLUME CONTROL
The AIF1 interface supports up to four output channels. A digital volume control is provided on each of these output signal paths, allowing attenuation in the range -71.625dB to +17.625dB in 0.375dB steps. The level of attenuation for an eight-bit code X is given by:
0.375 (X-192) dB for 1 X 239; MUTE for X = 0 +17.625dB for 239 X 255
The AIF1ADC1_VU and AIF1ADC2_VU bits control the loading of digital volume control data. When the volume update bit is set to 0, the associated volume control data will be loaded into the respective control register, but will not actually change the digital gain setting.
The AIF1ADC1L and AIF1ADC1R gain settings are updated when a 1 is written to AIF1ADC1_VU.
The AIF1ADC2L and AIF1ADC2R gain settings are updated when a 1 is written to AIF1ADC2_VU.
This makes it possible to update the gain of left and right channels simultaneously.
DESCRIPTION REGISTER
ADDRESS
R1024
(0400h)
AIF1 ADC1
Left Volume
R1025
(0401h)
AIF1 ADC1
Right Volume
R1028
(0404h)
AIF1 ADC2
Left Volume
8
7:0
8
7:0
8
AIF1ADC1_
VU
AIF1ADC1L
_VOL [7:0]
AIF1ADC1_
VU
AIF1ADC1R
_VOL [7:0]
AIF1ADC2_
VU
N/A
C0h
(0dB)
N/A
C0h
(0dB)
N/A
AIF1ADC1 output path (AIF1, Timeslot 0)
Volume Update
Writing a 1 to this bit will cause the
AIF1ADC1L and AIF1ADC1R volume to be updated simultaneously
AIF1ADC1 (Left) output path (AIF1, Timeslot
0) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
AIF1ADC1 output path (AIF1, Timeslot 0)
Volume Update
Writing a 1 to this bit will cause the
AIF1ADC1L and AIF1ADC1R volume to be updated simultaneously
AIF1ADC1 (Right) output path (AIF1, Timeslot
0) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
AIF1ADC2 output path (AIF1, Timeslot 1)
Volume Update
Writing a 1 to this bit will cause the
AIF1ADC2L and AIF1ADC2R volume to be updated simultaneously w PP, August 2012, Rev 3.4
95
WM8958
REGISTER
ADDRESS
R1029
(0405h)
AIF1 ADC2
Right Volume
7:0 AIF1ADC2L
_VOL [7:0]
8
Pre-Production
DESCRIPTION
7:0
AIF1ADC2_
VU
AIF1ADC2R
_VOL [7:0]
C0h
(0dB)
N/A
C0h
(0dB)
AIF1ADC2 (Left) output path (AIF1, Timeslot
1) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
AIF1ADC2 output path (AIF1, Timeslot 1)
Volume Update
Writing a 1 to this bit will cause the
AIF1ADC2L and AIF1ADC2R volume to be updated simultaneously
AIF1ADC2 (Right) output path (AIF1, Timeslot
1) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
Table 40 AIF1 Output Path Volume Control w PP, August 2012, Rev 3.4
96
Pre-Production w
WM8958
6Dh
6Eh
6Fh
70h
71h
65h
66h
67h
68h
69h
6Ah
6Bh
6Ch
60h
61h
62h
63h
64h
58h
59h
5Ah
5Bh
5Ch
5Dh
5Eh
5Fh
7Ah
7Bh
7Ch
7Dh
7Eh
7Fh
72h
73h
74h
75h
76h
77h
78h
79h
50h
51h
52h
53h
54h
55h
56h
57h
4Bh
4Ch
4Dh
4Eh
4Fh
AIF1/AIF2
Output Volume
40h
41h
42h
43h
44h
45h
46h
47h
48h
49h
4Ah
-58.125
-57.750
-57.375
-57.000
-56.625
-56.250
-55.875
-55.500
-55.125
-54.750
-54.375
-54.000
-53.625
-63.000
-62.625
-62.250
-61.875
-61.500
-61.125
-60.750
-60.375
-60.000
-59.625
-59.250
-58.875
-58.500
-53.250
-52.875
-52.500
-52.125
-51.750
-51.375
-51.000
-50.625
-50.250
-49.875
-49.500
-49.125
-48.750
-48.375
-67.875
-67.500
-67.125
-66.750
-66.375
-66.000
-65.625
-65.250
-64.875
-64.500
-64.125
-63.750
-63.375
Volume
(dB)
MUTE
-71.625
-71.250
-70.875
-70.500
-70.125
-69.750
-69.375
-69.000
-68.625
-68.250
2Dh
2Eh
2Fh
30h
31h
25h
26h
27h
28h
29h
2Ah
2Bh
2Ch
20h
21h
22h
23h
24h
18h
19h
1Ah
1Bh
1Ch
1Dh
1Eh
1Fh
3Ah
3Bh
3Ch
3Dh
3Eh
3Fh
32h
33h
34h
35h
36h
37h
38h
39h
10h
11h
12h
13h
14h
15h
16h
17h
Bh
Ch
Dh
Eh
Fh
AIF1/AIF2
Output Volume
0h
1h
2h
6h
7h
8h
3h
4h
5h
9h
Ah
A5h
A6h
A7h
A8h
A9h
AAh
ABh
ACh
ADh
AEh
AFh
B0h
B1h
A0h
A1h
A2h
A3h
A4h
98h
99h
9Ah
9Bh
9Ch
9Dh
9Eh
9Fh
BAh
BBh
BCh
BDh
BEh
BFh
B2h
B3h
B4h
B5h
B6h
B7h
B8h
B9h
90h
91h
92h
93h
94h
95h
96h
97h
8Bh
8Ch
8Dh
8Eh
8Fh
AIF1/AIF2
Output Volume
80h
81h
82h
83h
84h
85h
86h
87h
88h
89h
8Ah
-34.125
-33.750
-33.375
-33.000
-32.625
-32.250
-31.875
-31.500
-31.125
-30.750
-30.375
-30.000
-29.625
-39.000
-38.625
-38.250
-37.875
-37.500
-37.125
-36.750
-36.375
-36.000
-35.625
-35.250
-34.875
-34.500
-29.250
-28.875
-28.500
-28.125
-27.750
-27.375
-27.000
-26.625
-26.250
-25.875
-25.500
-25.125
-24.750
-24.375
-43.875
-43.500
-43.125
-42.750
-42.375
-42.000
-41.625
-41.250
-40.875
-40.500
-40.125
-39.750
-39.375
Volume
(dB)
-48.000
-47.625
-47.250
-46.875
-46.500
-46.125
-45.750
-45.375
-45.000
-44.625
-44.250
E5h
E6h
E7h
E8h
E9h
EAh
EBh
ECh
EDh
EEh
EFh
F0h
F1h
E0h
E1h
E2h
E3h
E4h
D8h
D9h
DAh
DBh
DCh
DDh
DEh
DFh
FAh
FBh
FCh
FDh
FEh
FFh
F2h
F3h
F4h
F5h
F6h
F7h
F8h
F9h
CBh
CCh
CDh
CEh
CFh
D0h
D1h
D2h
D3h
D4h
D5h
D6h
D7h
AIF1/AIF2
Output Volume
C0h
C1h
C2h
C3h
C4h
C5h
C6h
C7h
C8h
C9h
CAh
-10.125
-9.750
-9.375
-9.000
-8.625
-8.250
-7.875
-7.500
-7.125
-6.750
-6.375
-6.000
-5.625
-15.000
-14.625
-14.250
-13.875
-13.500
-13.125
-12.750
-12.375
-12.000
-11.625
-11.250
-10.875
-10.500
-5.250
-4.875
-4.500
-4.125
-3.750
-3.375
-3.000
-2.625
-2.250
-1.875
-1.500
-1.125
-0.750
-0.375
-19.875
-19.500
-19.125
-18.750
-18.375
-18.000
-17.625
-17.250
-16.875
-16.500
-16.125
-15.750
-15.375
Volume
(dB)
-24.000
-23.625
-23.250
-22.875
-22.500
-22.125
-21.750
-21.375
-21.000
-20.625
-20.250
13.875
14.250
14.625
15.000
15.375
15.750
16.125
16.500
16.875
17.250
17.625
17.625
17.625
9.000
9.375
9.750
10.125
10.500
10.875
11.250
11.625
12.000
12.375
12.750
13.125
13.500
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
17.625
4.125
4.500
4.875
5.250
5.625
6.000
6.375
6.750
7.125
7.500
7.875
8.250
8.625
Volume
(dB)
0.000
0.375
0.750
1.125
1.500
1.875
2.250
2.625
3.000
3.375
3.750
Table 41 AIF1 Output Path Digital Volume Range
AIF1 - OUTPUT PATH HIGH PASS FILTER
A digital high-pass filter can be enabled in the AIF1 output paths to remove DC offsets. This filter is enabled independently in the four AIF1 output channels using the register bits described in Table 42.
The HPF cut-off frequency for the AIF1 Timeslot 0 channels is set using AIF1ADC1_HPF_CUT. The
HPF cut-off frequency for the AIF1 Timeslot 1 channels is set using AIF1ADC2_HPF_CUT.
PP, August 2012, Rev 3.4
97
WM8958 w
Pre-Production
In hi-fi mode, the high pass filter is optimised for removing DC offsets without degrading the bass response and has a cut-off frequency of 3.7Hz when the sample rate (fs) = 44.1kHz.
In voice modes, the high pass filter is optimised for voice communication; it is recommended to set the cut-off frequency below 300Hz.
Note that the cut-off frequencies scale with the AIF1 sample rate. (The AIF1 sample rate is set using the AIF1_SR register, as described in the “Clocking and Sample Rates” section.) See Table 43 for the
HPF cut-off frequencies at all supported sample rates.
DESCRIPTION REGISTER
ADDRESS
R1040 (0410h)
AIF1 ADC1
Filters
14:13 AIF1ADC1_
HPF_CUT
[1:0]
R1041 (0411h)
AIF1 ADC2
Filters
12
11
AIF1ADC1L_
HPF
AIF1ADC1R
_HPF
14:13 AIF1ADC2_
12
11
HPF_CUT
[1:0]
AIF1ADC2L_
HPF
AIF1ADC2R
_HPF
00
0
0
00
0
0
AIF1ADC1 output path (AIF1, Timeslot 0)
Digital HPF cut-off frequency (fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 64Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
AIF1ADC1 (Left) output path (AIF1, Timeslot
0) Digital HPF Enable
0 = Disabled
1 = Enabled
AIF1ADC1 (Right) output path (AIF1, Timeslot
0) Digital HPF Enable
0 = Disabled
1 = Enabled
AIF1ADC2 output path (AIF1, Timeslot 1)
Digital HPF cut-off frequency (fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 64Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
AIF1ADC2 (Left) output path (AIF1, Timeslot
1) Digital HPF Enable
0 = Disabled
1 = Enabled
AIF1ADC2 (Right) output path (AIF1, Timeslot
1) Digital HPF Enable
0 = Disabled
1 = Enabled
Table 42 AIF1 Output Path High Pass Filter
Sample
Frequency
(kHz)
Cut-Off Frequency (Hz) for given value of AIF n ADC n _HPF_CUT
00 01 10 11
8.000 0.7 64 130 267
11.025 0.9 88 178 367
16.000 1.3 127 258 532
22.050 1.9 175 354 733
24.000 2.0 190 386 798
32.000 2.7 253 514 1063
44.100 3.7 348 707 1464
48.000 4.0 379 770 1594
88.200 7.4 696 1414 2928
96.000 8.0 758 1540 3188
Table 43 AIF1 Output Path High Pass Filter Cut-Off Frequencies
PP, August 2012, Rev 3.4
98
Pre-Production
WM8958
AIF1 - INPUT PATH VOLUME CONTROL
The AIF1 interface supports up to four input channels. A digital volume control is provided on each of these input signal paths, allowing attenuation in the range -71.625dB to 0dB in 0.375dB steps. The level of attenuation for an eight-bit code X is given by:
0.375 (X-192) dB for 1 X 192; MUTE for X = 0 0dB for 192 X 255
The AIF1DAC1_VU and AIF1DAC2_VU bits control the loading of digital volume control data. When the volume update bit is set to 0, the associated volume control data will be loaded into the respective control register, but will not actually change the digital gain setting.
The AIF1DAC1L and AIF1DAC1R gain settings are updated when a 1 is written to AIF1DAC1_VU.
The AIF1DAC2L and AIF1DAC2R gain settings are updated when a 1 is written to AIF1DAC2_VU.
This makes it possible to update the gain of left and right channels simultaneously.
Note that a digital gain function is also available at the audio interface input, to boost the DAC volume when a small signal is received on DACDAT1. See “Digital Audio Interface Control” for further details.
Digital volume control is also possible at the DAC stage of the signal path, after the audio signal has passed through the DAC digital mixers. See “Digital to Analogue Converter (DAC)” for further details.
DESCRIPTION REGISTER
ADDRESS
R1026
(0402h)
AIF1 DAC1
Left Volume
R1027
(0403h)
AIF1 DAC1
Right Volume
R1030
(0406h)
AIF1 DAC2
Left Volume
8
7:0
8
7:0
8
AIF1DAC1_
VU
AIF1DAC1L
_VOL [7:0]
AIF1DAC1_
VU
AIF1DAC1R
_VOL [7:0]
AIF1DAC2_
VU
N/A
C0h
(0dB)
N/A
C0h
(0dB)
N/A
AIF1DAC1 input path (AIF1, Timeslot 0)
Volume Update
Writing a 1 to this bit will cause the
AIF1DAC1L and AIF1DAC1R volume to be updated simultaneously
AIF1DAC1 (Left) input path (AIF1, Timeslot 0)
Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
AIF1DAC1 input path (AIF1, Timeslot 0)
Volume Update
Writing a 1 to this bit will cause the
AIF1DAC1L and AIF1DAC1R volume to be updated simultaneously
AIF1DAC1 (Right) input path (AIF1, Timeslot
0) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
AIF1DAC2 input path (AIF1, Timeslot 1)
Volume Update
Writing a 1 to this bit will cause the
AIF1DAC2L and AIF1DAC2R volume to be updated simultaneously w PP, August 2012, Rev 3.4
99
WM8958
REGISTER
ADDRESS
R1031
(0407h)
AIF1 DAC2
Right Volume
7:0 AIF1DAC2L
_VOL [7:0]
8
Pre-Production
DESCRIPTION
7:0
AIF1DAC2_
VU
AIF1DAC2R
_VOL [7:0]
C0h
(0dB)
N/A
C0h
(0dB)
AIF1DAC2 (Left) input path (AIF1, Timeslot 1)
Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
AIF1DAC2 input path (AIF1, Timeslot 1)
Volume Update
Writing a 1 to this bit will cause the
AIF1DAC2L and AIF1DAC2R volume to be updated simultaneously
AIF1DAC2 (Right) input path (AIF1, Timeslot
1) Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
Table 44 AIF1 Input Path Volume Control w PP, August 2012, Rev 3.4
100
Pre-Production
WM8958
Volume
(dB)
MUTE
-58.125
-57.750
-57.375
-57.000
-56.625
-56.250
-55.875
-55.500
-55.125
-54.750
-54.375
-54.000
-53.625
-53.250
-52.875
-52.500
-52.125
-51.750
-51.375
-63.750
-63.375
-63.000
-62.625
-62.250
-61.875
-61.500
-61.125
-60.750
-60.375
-60.000
-59.625
-59.250
-58.875
-58.500
-51.000
-50.625
-50.250
-49.875
-49.500
-49.125
-48.750
-48.375
-71.625
-71.250
-70.875
-70.500
-70.125
-69.750
-69.375
-69.000
-68.625
-68.250
-67.875
-67.500
-67.125
-66.750
-66.375
-66.000
-65.625
-65.250
-64.875
-64.500
-64.125
AIF1/AIF2 Input
Volume
0h
25h
26h
27h
28h
29h
2Ah
2Bh
2Ch
2Dh
2Eh
2Fh
30h
31h
32h
33h
34h
35h
36h
37h
38h
39h
3Ah
3Bh
3Ch
3Dh
3Eh
3Fh
1Bh
1Ch
1Dh
1Eh
1Fh
20h
21h
22h
23h
24h
16h
17h
18h
19h
1Ah
Ah
Bh
Ch
6h
7h
8h
9h
1h
2h
3h
4h
5h
Dh
Eh
Fh
10h
11h
12h
13h
14h
15h
AIF1/AIF2 Input
Volume
40h
65h
66h
67h
68h
69h
6Ah
6Bh
6Ch
6Dh
6Eh
6Fh
70h
71h
72h
73h
74h
75h
76h
77h
78h
79h
7Ah
7Bh
7Ch
7Dh
7Eh
7Fh
5Bh
5Ch
5Dh
5Eh
5Fh
60h
61h
62h
63h
64h
56h
57h
58h
59h
5Ah
46h
47h
48h
49h
4Ah
4Bh
4Ch
41h
42h
43h
44h
45h
4Dh
4Eh
4Fh
50h
51h
52h
53h
54h
55h
Volume
(dB)
-24.000
-10.125
-9.750
-9.375
-9.000
-8.625
-8.250
-7.875
-7.500
-7.125
-6.750
-6.375
-6.000
-5.625
-5.250
-4.875
-4.500
-4.125
-3.750
-3.375
-3.000
-2.625
-2.250
-1.875
-1.500
-1.125
-0.750
-0.375
-15.750
-15.375
-15.000
-14.625
-14.250
-13.875
-13.500
-13.125
-12.750
-12.375
-12.000
-11.625
-11.250
-10.875
-10.500
-23.625
-23.250
-22.875
-22.500
-22.125
-21.750
-21.375
-21.000
-20.625
-20.250
-19.875
-19.500
-19.125
-18.750
-18.375
-18.000
-17.625
-17.250
-16.875
-16.500
-16.125
AIF1/AIF2 Input
Volume
80h
A5h
A6h
A7h
A8h
A9h
AAh
ABh
ACh
ADh
AEh
AFh
B0h
B1h
B2h
B3h
B4h
B5h
B6h
B7h
B8h
B9h
BAh
BBh
BCh
BDh
BEh
BFh
9Bh
9Ch
9Dh
9Eh
9Fh
A0h
A1h
A2h
A3h
A4h
96h
97h
98h
99h
9Ah
86h
87h
88h
89h
8Ah
8Bh
8Ch
81h
82h
83h
84h
85h
8Dh
8Eh
8Fh
90h
91h
92h
93h
94h
95h
Volume
(dB)
-48.000
-34.125
-33.750
-33.375
-33.000
-32.625
-32.250
-31.875
-31.500
-31.125
-30.750
-30.375
-30.000
-29.625
-29.250
-28.875
-28.500
-28.125
-27.750
-27.375
-39.750
-39.375
-39.000
-38.625
-38.250
-37.875
-37.500
-37.125
-36.750
-36.375
-36.000
-35.625
-35.250
-34.875
-34.500
-27.000
-26.625
-26.250
-25.875
-25.500
-25.125
-24.750
-24.375
-47.625
-47.250
-46.875
-46.500
-46.125
-45.750
-45.375
-45.000
-44.625
-44.250
-43.875
-43.500
-43.125
-42.750
-42.375
-42.000
-41.625
-41.250
-40.875
-40.500
-40.125
Volume
(dB)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
AIF1/AIF2 Input
Volume
C0h
E5h
E6h
E7h
E8h
E9h
EAh
EBh
ECh
EDh
EEh
EFh
F0h
F1h
F2h
F3h
F4h
F5h
F6h
F7h
F8h
F9h
FAh
FBh
FCh
FDh
FEh
FFh
D6h
D7h
D8h
D9h
DAh
DBh
DCh
DDh
DEh
DFh
E0h
E1h
E2h
E3h
E4h
C6h
C7h
C8h
C9h
CAh
CBh
CCh
C1h
C2h
C3h
C4h
C5h
CDh
CEh
CFh
D0h
D1h
D2h
D3h
D4h
D5h
Table 45 AIF1 Input Path Digital Volume Range w PP, August 2012, Rev 3.4
101
WM8958
Pre-Production
AIF1 - INPUT PATH SOFT MUTE CONTROL
The WM8958 provides a soft mute function for each of the AIF1 interface input paths. When the softmute function is selected, the WM8958 gradually attenuates the associated signal paths until the path is entirely muted.
When the soft-mute function is de-selected, the gain will either return instantly to the digital gain setting, or will gradually ramp back to the digital gain setting, depending on the applicable
_UNMUTE_RAMP register field.
The mute and un-mute ramp rate is selectable between two different rates.
The AIF1 input paths are soft-muted by default. To play back an audio signal, the soft-mute must first be de-selected by setting the applicable Mute bit to 0.
The soft un-mute would typically be used during playback of audio data so that when the Mute is subsequently disabled, a smooth transition is scheduled to the previous volume level and pop noise is avoided. This is desirable when resuming playback after pausing during a track. w
The soft un-mute would typically not be required when un-muting at the start of a music file, in order that the first part of the music track is not attenuated. The instant un-mute behaviour is desirable in this case, when starting playback of a new track. See “DAC Soft Mute and Soft Un-Mute” (Figure 29) for an illustration of the soft mute function.
DESCRIPTION REGISTER
ADDRESS
R1056 (0420h)
AIF1 DAC1
Filters (1)
R1058 (0422h)
AIF1 DAC2
Filters (1)
9
5
4
9
5
4
AIF1DAC1_
MUTE
AIF1DAC1_
MUTERAT
E
AIF1DAC1_
UNMUTE_
RAMP
AIF1DAC2_
MUTE
AIF1DAC2_
MUTERAT
E
AIF1DAC2_
UNMUTE_
RAMP
1
0
0
1
0
0
AIF1DAC1 input path (AIF1, Timeslot 0) Soft
Mute Control
0 = Un-mute
1 = Mute
AIF1DAC1 input path (AIF1, Timeslot 0) Soft
Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is
10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is
171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF1DAC1 input path (AIF1, Timeslot 0)
Unmute Ramp select
0 = Disabling soft-mute (AIF1DAC1_MUTE=0) will cause the volume to change immediately to
AIF1DAC1L_VOL and AIF1DAC1R_VOL settings
1 = Disabling soft-mute (AIF1DAC1_MUTE=0) will cause the DAC volume to ramp up gradually to the AIF1DAC1L_VOL and
AIF1DAC1R_VOL settings
AIF1DAC2 input path (AIF1, Timeslot 1) Soft
Mute Control
0 = Un-mute
1 = Mute
AIF1DAC2 input path (AIF1, Timeslot 1) Soft
Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is
10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is
171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF1DAC2 input path (AIF1, Timeslot 1)
Unmute Ramp select
0 = Disabling soft-mute (AIF1DAC2_MUTE=0) will cause the volume to change immediately to
AIF1DAC2L_VOL and AIF1DAC2R_VOL settings
1 = Disabling soft-mute (AIF1DAC2_MUTE=0)
PP, August 2012, Rev 3.4
102
Pre-Production w
WM8958
REGISTER
ADDRESS
DESCRIPTION will cause the DAC volume to ramp up gradually to the AIF1DAC2L_VOL and
AIF1DAC2R_VOL settings
Table 46 AIF1 Input Path Soft Mute Control
AIF1 - INPUT PATH NOISE GATE CONTROL
The WM8958 provides a digital noise gate function for the AIF1 input paths. The noise gate ensures best noise performance when the signal path is idle. When the noise gate is enabled, and the signal level is below the noise gate threshold, then the noise gate is activated, causing the signal path to be muted.
The AIF1 Timeslot 0 input path noise gate is enabled using the AIF1DAC1_NG_ENA register. The
AIF1 Timeslot 1 input path noise gate is enabled using the AIF1DAC2_NG_ENA register.
The noise gate threshold (the signal level below which the noise gate is activated) is set using
AIF1DAC1_NG_THR or AIF1DAC2_NG_THR.
To prevent erroneous triggering, a time delay is applied before the gate is activated; the signal path is only muted when the signal level stays below the threshold for longer than ‘hold time’, determined by the AIF1DAC1_NG_HLD or AIF1DAC2_NG_HLD registers.
When the noise gate is activated, the WM8958 gradually attenuates the associated AIF1 input signal paths until each is entirely muted. When the signal level increases, and the noise gate is de-activated, the gain will return to the AIF1DACnL_VOL and AIF1DACnR_VOL digital gain settings (where n = 1 for AIF1 Timeslot 0, and n = 2 for AIF1 Timeslot 1). The un-mute behaviour can be immediate or gradual; this is determined by the AIF1DACn_MUTERATE and AIF1DACn_UNMUTE_RAMP registers described in Table 46.
DESCRIPTION REGISTER
ADDRESS
R1072 (0430h)
AIF1 DAC1
Noise Gate
6:5 AIF1DAC1_
NG_HLD
[1:0]
11
R1073 (0431h)
AIF1 DAC2
Noise Gate
3:1
0
6:5
AIF1DAC1_
NG_THR
[2:0]
AIF1DAC1_
NG_ENA
AIF1DAC2_
NG_HLD
[1:0]
100
0
11
AIF1DAC1 input path (AIF1, Timeslot 0) Noise
Gate Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
AIF1DAC1 input path (AIF1, Timeslot 0) Noise
Gate Threshold
000 = -60dB
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF1DAC1 input path (AIF1, Timeslot 0) Noise
Gate Enable
0 = Disabled
1 = Enabled
AIF1DAC2 input path (AIF1, Timeslot 1) Noise
Gate Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
PP, August 2012, Rev 3.4
103
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
3:1 AIF1DAC2_
NG_THR
[2:0]
0 AIF1DAC2_
NG_ENA
100
0
AIF1DAC2 input path (AIF1, Timeslot 1) Noise
Gate Threshold
000 = -60dB
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF1DAC2 input path (AIF1, Timeslot 1) Noise
Gate Enable
0 = Disabled
1 = Enabled
Table 47 AIF1 Input Path Noise Gate Control
AIF1 - INPUT PATH MONO MIX CONTROL
A digital mono mix can be selected on one or both pairs of AIF1 input channels. The mono mix is generated as the sum of the Left and Right AIF channel data. When the mono mix function is enabled, the combined mono signal is applied to the Left channel and the Right channel of the respective AIF1 signal processing and digital mixing paths. To prevent clipping, 6dB attenuation is applied to the mono mix.
REGISTER
ADDRESS
R1056 (0420h)
AIF1 DAC1
Filters (1)
R1058 (0422h)
AIF1 DAC2
Filters (1)
7
7
AIF1DAC1_
MONO
AIF1DAC2_
MONO
0
0
DESCRIPTION
AIF1DAC1 input path (AIF1, Timeslot 0) Mono
Mix Control
0 = Disabled
1 = Enabled
AIF1DAC2 input path (AIF1, Timeslot 1) Mono
Mix Control
0 = Disabled
1 = Enabled
Table 48 AIF1 Input Path Mono Mix Control
AIF2 - OUTPUT PATH VOLUME CONTROL
The AIF2 interface supports two output channels. A digital volume control is provided on each output signal path, allowing attenuation in the range -71.625dB to +17.625dB in 0.375dB steps. The level of attenuation for an eight-bit code X is given by:
0.375 (X-192) dB for 1 X 239; MUTE for X = 0 +17.625dB for 239 X 255
The AIF2ADC_VU bit controls the loading of digital volume control data. When AIF2ADC_VU bit is set to 0, the AIF2ADCL_VOL and AIF2ADCR_VOL control data will be loaded into the respective control register, but will not actually change the digital gain setting. Both left and right gain settings are updated when a 1 is written to AIF2ADC_VU. This makes it possible to update the gain of left and right channels simultaneously. w PP, August 2012, Rev 3.4
104
Pre-Production w
WM8958
REGISTER
ADDRESS
R1280
(0500h)
AIF2 ADC
Left Volume
R1281
(0501h)
AIF2 ADC
Right Volume
8
8
AIF2ADC_V
U
7:0 AIF2ADCL_
VOL [7:0]
AIF2ADC_V
U
7:0 AIF2ADCR_
VOL [7:0]
N/A
C0h
(0dB)
N/A
C0h
(0dB)
DESCRIPTION
AIF2ADC output path Volume Update
Writing a 1 to this bit will cause the AIF2ADCL and AIF2ADCR volume to be updated simultaneously
AIF2ADC (Left) output path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
AIF2ADC output path Volume Update
Writing a 1 to this bit will cause the AIF2ADCL and AIF2ADCR volume to be updated simultaneously
AIF2ADC (Right) output path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
(See Table 41 for volume range)
Table 49 AIF2 Output Path Volume Control
AIF2 - OUTPUT PATH HIGH PASS FILTER
A digital high-pass filter can be enabled in the AIF2 output paths to remove DC offsets. This filter is enabled independently in the two AIF2 output channels using the register bits described in Table 50.
The HPF cut-off frequency for the AIF2 channels is set using AIF2ADC_HPF_CUT.
In hi-fi mode, the high pass filter is optimised for removing DC offsets without degrading the bass response and has a cut-off frequency of 3.7Hz when the sample rate (fs) = 44.1kHz.
In voice modes, the high pass filter is optimised for voice communication; it is recommended to set the cut-off frequency below 300Hz.
Note that the cut-off frequencies scale with the AIF2 sample rate. (The AIF2 sample rate is set using the AIF2_SR register, as described in the “Clocking and Sample Rates” section.) See Table 43 for the
HPF cut-off frequencies at all supported sample rates.
DESCRIPTION REGISTER
ADDRESS
R1296 (0510h)
AIF2 ADC
Filters
14:13 AIF2ADC_H
PF_CUT
[1:0]
12
11
AIF2ADCL_
HPF
AIF2ADCR_
HPF
00
0
0
AIF2ADC output path Digital HPF Cut-Off
Frequency (fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 127Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
AIF2ADC (Left) output path Digital HPF
Enable
0 = Disabled
1 = Enabled
AIF2ADC (Right) output path Digital HPF
Enable
0 = Disabled
1 = Enabled
Table 50 AIF2 Output Path High Pass Filter
PP, August 2012, Rev 3.4
105
WM8958 w
Pre-Production
AIF2 - INPUT PATH VOLUME CONTROL
The AIF2 interface supports two input channels. A digital volume control is provided on each input signal path, allowing attenuation in the range -71.625dB to 0dB in 0.375dB steps. The level of attenuation for an eight-bit code X is given by:
0.375 (X-192) dB for 1 X 192; MUTE for X = 0 0dB for 192 X 255
The AIF2DAC_VU bit controls the loading of digital volume control data. When AIF2DAC_VU bit is set to 0, the AIF2DACL_VOL and AIF2DACR_VOL control data will be loaded into the respective control register, but will not actually change the digital gain setting. Both left and right gain settings are updated when a 1 is written to AIF2DAC_VU. This makes it possible to update the gain of left and right channels simultaneously.
Note that a digital gain function is also available at the audio interface input, to boost the DAC volume when a small signal is received on DACDAT2. See “Digital Audio Interface Control” for further details.
Digital volume control is also possible at the DAC stage of the signal path, after the audio signal has passed through the DAC digital mixers. See “Digital to Analogue Converter (DAC)” for further details.
DESCRIPTION REGISTER
ADDRESS
R1282
(0502h)
AIF2 DAC
Left Volume
R1283
(0503h)
AIF2 DAC
Right Volume
8
7:0
8
7:0
AIF2DAC_V
U
AIF2DACL_
VOL [7:0]
AIF2DAC_V
U
AIF2DACR_
VOL [7:0]
N/A
C0h
(0dB)
N/A
C0h
(0dB)
AIF2DAC input path Volume Update
Writing a 1 to this bit will cause the AIF2DACL and AIF2DACR volume to be updated simultaneously
AIF2DAC (Left) input path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
AIF2DAC input path Volume Update
Writing a 1 to this bit will cause the AIF2DACL and AIF2DACR volume to be updated simultaneously
AIF2DAC (Right) input path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
(See Table 45 for volume range)
Table 51 AIF2 Input Path Volume Control
AIF2 - INPUT PATH SOFT MUTE CONTROL
The WM8958 provides a soft mute function for each of the AIF2 interface input paths. When the softmute function is selected, the WM8958 gradually attenuates the associated signal paths until the path is entirely muted.
When the soft-mute function is de-selected, the gain will either return instantly to the digital gain setting, or will gradually ramp back to the digital gain setting, depending on the
AIF2DAC_UNMUTE_RAMP register field.
The mute and un-mute ramp rate is selectable between two different rates.
The AIF2 input paths are soft-muted by default. To play back an audio signal, the soft-mute must first be de-selected by setting AIF2DAC_MUTE = 0.
PP, August 2012, Rev 3.4
106
Pre-Production w
WM8958
The soft un-mute would typically be used during playback of audio data so that when the Mute is subsequently disabled, a smooth transition is scheduled to the previous volume level and pop noise is avoided. This is desirable when resuming playback after pausing during a track.
The soft un-mute would typically not be required when un-muting at the start of a music file, in order that the first part of the music track is not attenuated. The instant un-mute behaviour is desirable in this case, when starting playback of a new track. See “DAC Soft Mute and Soft Un-Mute” (Figure 29) for an illustration of the soft mute function.
DESCRIPTION REGISTER
ADDRESS
R1312 (0520h)
AIF2 DAC
Filters (1)
9
5
4
AIF2DAC_M
UTE
AIF2DAC_M
UTERATE
AIF2DAC_U
NMUTE_RA
MP
1
0
0
AIF2DAC input path Soft Mute Control
0 = Un-mute
1 = Mute
AIF2DAC input path Soft Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is
10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is
171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF2DAC input path Unmute Ramp select
0 = Disabling soft-mute (AIF2DAC_MUTE=0) will cause the volume to change immediately to AIF2DACL_VOL and AIF2DACR_VOL settings
1 = Disabling soft-mute (AIF2DAC_MUTE=0) will cause the DAC volume to ramp up gradually to the AIF2DACL_VOL and
AIF2DACR_VOL settings
Table 52 AIF2 Input Path Soft Mute Control
AIF2 - INPUT PATH NOISE GATE CONTROL
The WM8958 provides a digital noise gate function for the AIF2 input paths. The noise gate ensures best noise performance when the signal path is idle. When the noise gate is enabled, and the signal level is below the noise gate threshold, then the noise gate is activated, causing the signal path to be muted.
The AIF2 input path noise gate is enabled using the AIF2DAC_NG_ENA register.
The noise gate threshold (the signal level below which the noise gate is activated) is set using
AIF2DAC_NG_THR.
To prevent erroneous triggering, a time delay is applied before the gate is activated; the signal path is only muted when the signal level stays below the threshold for longer than ‘hold time’, determined by the AIF2DAC_NG_HLD register.
When the noise gate is activated, the WM8958 gradually attenuates the AIF2 input signal paths until each is entirely muted. When the signal level increases, and the noise gate is de-activated, the gain will return to the AIF2DACL_VOL and AIF2DACR_VOL digital gain settings. The un-mute behaviour can be immediate or gradual; this is determined by the AIF2DAC_MUTERATE and
AIF2DAC_UNMUTE_RAMP registers described in Table 52.
DESCRIPTION REGISTER
ADDRESS
R1328 (0530h)
AIF2 DAC
Noise Gate
6:5
3:1
AIF2DAC_
NG_HLD
[1:0]
AIF2DAC_
NG_THR
11
100
AIF2DAC input path Noise Gate Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
AIF2DAC input path Noise Gate Threshold
000 = -60dB
PP, August 2012, Rev 3.4
107
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
0
[2:0]
AIF2DAC_
NG_ENA
0
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF2DAC input path Noise Gate Enable
0 = Disabled
1 = Enabled
Table 53 AIF2 Input Path Noise Gate Control
AIF2 - INPUT PATH MONO MIX CONTROL
A digital mono mix can be selected on the AIF2 input channels. The mono mix is generated as the sum of the Left and Right AIF channel data. When the mono mix function is enabled, the combined mono signal is applied to the Left channel and the Right channel of the AIF2 signal processing and digital mixing paths. To prevent clipping, 6dB attenuation is applied to the mono mix.
REGISTER
ADDRESS
R1312 (0520h)
AIF2 DAC
Filters (1)
7 AIF2DAC_M
ONO
Table 54 AIF2 Input Path Mono Mix Control
0
DESCRIPTION
AIF2DAC input path Mono Mix Control
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
108
Pre-Production
WM8958
DIGITAL TO ANALOGUE CONVERTER (DAC)
The WM8958 DACs receive digital input data from the DAC mixers - see “Digital Mixing”. The digital audio data is converted to oversampled bit streams in the on-chip, true 24-bit digital interpolation filters. The bitstream data enters four multi-bit, sigma-delta DACs, which convert them to high quality analogue audio signals. The multi-bit DAC architecture reduces high frequency noise and sensitivity to clock jitter. It also uses a Dynamic Element Matching technique for high linearity and low distortion.
A high performance mode of DAC operation can be selected by setting the DAC_OSR128 bit - see
“Clocking and Sample Rates” for details.
The analogue outputs from the DACs can be mixed with analogue line/mic inputs using the line output mixers MIXOUTL / MIXOUTR and the speaker output mixers SPKMIXL / SPKMIXR.
The DACs are enabled using the register bits defined in Table 55.
Note that the DAC clock must be enabled whenever the DACs are enabled.
REGISTER
ADDRESS
R5 (0005h)
Power
Management (5)
BIT LABEL DEFAULT
3
2
1
0
DAC2L_EN
A
DAC2R_EN
A
DAC1L_EN
A
DAC1R_EN
A
0
0
0
0
DESCRIPTION
Left DAC2 Enable
0 = Disabled
1 = Enabled
Right DAC2 Enable
0 = Disabled
1 = Enabled
Left DAC1 Enable
0 = Disabled
1 = Enabled
Right DAC1 Enable
0 = Disabled
1 = Enabled
Table 55 DAC Enable Control
DAC CLOCKING CONTROL
Clocking for the DACs is derived from SYSCLK. The required clock is enabled when the
SYSDSPCLK_ENA register is set.
The DAC clock rate is configured automatically, according to the AIFn_SR, AIFnCLK_RATE and
DAC_OSR128 registers. (See “Clocking and Sample Rates” for further details of the system clocks and control registers.)
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then the DAC clocking is controlled by the AIF1_SR and AIF1CLK_RATE registers.
When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then the DAC clocking is controlled by the AIF2_SR and AIF2CLK_RATE registers.
The supported DAC clocking configurations are described in Table 56 (for DAC_OSR128=0) and
Table 57 (for DAC_OSR128=1). Under default conditions, the DAC_OSR128 bit is not set. w PP, August 2012, Rev 3.4
109
WM8958 w
Pre-Production
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
22.05
24
32
44.1
48
88.2
Note 1
Note 1
Note 1
Note 1
Note 1 Note 1
Note 1 Note 1
Note 1 Note 1
Note 1
Note 1
96
When DAC_OSR128=0, DAC operation is only supported for the configurations indicated above
Table 56 DAC Clocking - DAC_OSR128 = 0 (Default)
SAMPLE
RATE (kHz)
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
8
11.025
12
16
22.05
24
32
44.1
Note 1 Note 1
Note 1 Note 1
Note 1 Note 1
48
88.2
96
When DAC_OSR128=1, DAC operation is only supported for the configurations indicated above
Table 57 DAC Clocking - DAC_OSR128 = 1
Note 1 - These clocking rates are only supported for ‘simple’ DAC-only playback modes, under the following conditions:
AIF input is enabled on a single interface (AIF1 or AIF2) only, or is enabled on AIF1 and
AIF2 simultaneously provided AIF1 and AIF2 are synchronised (ie. AIF1CLK_SRC =
AIF2CLK_SRC)
All AIF output paths are disabled
All DSP functions (ReTune™ Mobile Parametric Equaliser, 3D stereo expansion and
Dynamic Range Control) are disabled
The clocking requirements in Table 56 and Table 57 are only applicable to the AIFnCLK that is selected as the SYSCLK source. Note that both clocks (AIF1CLK and AIF2CLK) must satisfy the requirements noted in the “Clocking and Sample Rates” section.
The applicable clocks (SYSCLK, and AIF1CLK or AIF2CLK) must be present and enabled when using the Digital to Analogue Converters (DACs).
Note that the presence of a suitable clock is automatically detected by the WM8958; if the clock signal is absent, then any speaker or earpiece output driver(s) associated with the DAC signal paths will be disabled. (This is applicable to the SPKOUTL, SPKOUTR and HPOUT2 outputs only, whenever one or more DAC is routed to these output drivers.)
PP, August 2012, Rev 3.4
110
Pre-Production w
WM8958
DAC DIGITAL VOLUME
The output level of each DAC can be controlled digitally over a range from -71.625dB to +12dB in
0.375dB steps. The level of attenuation for an eight-bit code X is given by:
0.375 (X-192) dB for 1 X 224; MUTE for X = 0; 12dB to 224 X 255
Each of the DACs can be muted using the soft mute control bits described in Table 58. The WM8958 always applies a soft mute, where the volume is decreased gradually. The un-mute behaviour is configurable, as described in the “DAC Soft Mute and Soft Un-Mute” section.
The DAC1_VU and DAC2_VU bits control the loading of digital volume control data. When DAC1_VU is set to 0, the DAC1L_VOL or DAC1R_VOL control data will be loaded into the respective control register, but will not actually change the digital gain setting. Both left and right gain settings are updated when a 1 is written to DAC1_VU. This makes it possible to update the gain of both channels simultaneously. A similar function for DAC2L and DAC2R is controlled by the DAC2_VU register bit.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1552 (0610h)
DAC1 Left
Volume
9 DAC1L_MU
TE
1
R1553 (0611h)
DAC1 Right
Volume
R1554 (0612h)
DAC2 Left
Volume
8
7:0
9
8
7:0
9
8
DAC1_VU
DAC1L_VO
L [7:0]
DAC1R_MU
TE
DAC1_VU
DAC1R_VO
L [7:0]
DAC2L_MU
TE
DAC2_VU
N/A
C0h
(0dB)
1
N/A
C0h
(0dB)
1
N/A
DAC1L Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
DAC1L and DAC1R Volume Update
Writing a 1 to this bit will cause the
DAC1L and DAC1R volume to be updated simultaneously
DAC1L Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
(See Table 59 for volume range)
DAC1R Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
DAC1L and DAC1R Volume Update
Writing a 1 to this bit will cause the
DAC1L and DAC1R volume to be updated simultaneously
DAC1R Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
(See Table 59 for volume range)
DAC2L Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
DAC2L and DAC2R Volume Update
Writing a 1 to this bit will cause the
DAC2L and DAC2R volume to be updated simultaneously
PP, August 2012, Rev 3.4
111
WM8958
REGISTER
ADDRESS
R1555 (0613h)
DAC2 Right
Volume
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
7:0 DAC2L_VO
L [7:0]
9
8
7:0
DAC2R_MU
TE
DAC2_VU
DAC2R_VO
L [7:0]
C0h
(0dB)
1
N/A
C0h
(0dB)
DAC2L Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
(See Table 59 for volume range)
DAC2R Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
DAC2R and DAC2R Volume Update
Writing a 1 to this bit will cause the
DAC2R and DAC2R volume to be updated simultaneously
DAC2R Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
(See Table 59 for volume range)
Table 58 DAC Digital Volume Control w PP, August 2012, Rev 3.4
112
Pre-Production
DAC Volume
0h
18h
19h
1Ah
1Bh
1Ch
1Dh
1Eh
1Fh
20h
21h
22h
23h
24h
25h
26h
27h
Ch
Dh
Eh
Fh
10h
8h
9h
Ah
Bh
1h
2h
3h
4h
5h
6h
7h
11h
12h
13h
14h
15h
16h
17h
38h
39h
3Ah
3Bh
3Ch
3Dh
3Eh
3Fh
31h
32h
33h
34h
35h
36h
37h
28h
29h
2Ah
2Bh
2Ch
2Dh
2Eh
2Fh
30h
Volume
(dB)
MUTE
-63.000
-62.625
-62.250
-61.875
-61.500
-61.125
-60.750
-60.375
-60.000
-59.625
-59.250
-58.875
-58.500
-58.125
-57.750
-57.375
-71.625
-71.250
-70.875
-70.500
-70.125
-69.750
-69.375
-69.000
-68.625
-68.250
-67.875
-67.500
-67.125
-66.750
-66.375
-66.000
-65.625
-65.250
-64.875
-64.500
-64.125
-63.750
-63.375
-53.625
-53.250
-52.875
-52.500
-52.125
-51.750
-51.375
-51.000
-50.625
-50.250
-49.875
-49.500
-49.125
-48.750
-48.375
-57.000
-56.625
-56.250
-55.875
-55.500
-55.125
-54.750
-54.375
-54.000
DAC Volume
40h
58h
59h
5Ah
5Bh
5Ch
5Dh
5Eh
5Fh
60h
61h
62h
63h
64h
65h
66h
67h
48h
49h
4Ah
4Bh
4Ch
4Dh
4Eh
4Fh
50h
41h
42h
43h
44h
45h
46h
47h
51h
52h
53h
54h
55h
56h
57h
78h
79h
7Ah
7Bh
7Ch
7Dh
7Eh
7Fh
71h
72h
73h
74h
75h
76h
77h
68h
69h
6Ah
6Bh
6Ch
6Dh
6Eh
6Fh
70h
Volume
(dB)
-48.000
-39.000
-38.625
-38.250
-37.875
-37.500
-37.125
-36.750
-36.375
-36.000
-35.625
-35.250
-34.875
-34.500
-34.125
-33.750
-33.375
-47.625
-47.250
-46.875
-46.500
-46.125
-45.750
-45.375
-45.000
-44.625
-44.250
-43.875
-43.500
-43.125
-42.750
-42.375
-42.000
-41.625
-41.250
-40.875
-40.500
-40.125
-39.750
-39.375
-29.625
-29.250
-28.875
-28.500
-28.125
-27.750
-27.375
-27.000
-26.625
-26.250
-25.875
-25.500
-25.125
-24.750
-24.375
-33.000
-32.625
-32.250
-31.875
-31.500
-31.125
-30.750
-30.375
-30.000
Table 59 DAC Digital Volume Range
DAC Volume
80h
98h
99h
9Ah
9Bh
9Ch
9Dh
9Eh
9Fh
A0h
A1h
A2h
A3h
A4h
A5h
A6h
A7h
88h
89h
8Ah
8Bh
8Ch
8Dh
8Eh
8Fh
90h
81h
82h
83h
84h
85h
86h
87h
91h
92h
93h
94h
95h
96h
97h
B8h
B9h
BAh
BBh
BCh
BDh
BEh
BFh
B1h
B2h
B3h
B4h
B5h
B6h
B7h
A8h
A9h
AAh
ABh
ACh
ADh
AEh
AFh
B0h w
DAC Volume
C0h
D8h
D9h
DAh
DBh
DCh
DDh
DEh
DFh
E0h
E1h
E2h
E3h
E4h
E5h
E6h
E7h
C8h
C9h
CAh
CBh
CCh
CDh
CEh
CFh
D0h
C1h
C2h
C3h
C4h
C5h
C6h
C7h
D1h
D2h
D3h
D4h
D5h
D6h
D7h
F8h
F9h
FAh
FBh
FCh
FDh
FEh
FFh
F1h
F2h
F3h
F4h
F5h
F6h
F7h
E8h
E9h
EAh
EBh
ECh
EDh
EEh
EFh
F0h
Volume
(dB)
-24.000
-15.000
-14.625
-14.250
-13.875
-13.500
-13.125
-12.750
-12.375
-12.000
-11.625
-11.250
-10.875
-10.500
-10.125
-9.750
-9.375
-23.625
-23.250
-22.875
-22.500
-22.125
-21.750
-21.375
-21.000
-20.625
-20.250
-19.875
-19.500
-19.125
-18.750
-18.375
-18.000
-17.625
-17.250
-16.875
-16.500
-16.125
-15.750
-15.375
-5.625
-5.250
-4.875
-4.500
-4.125
-3.750
-3.375
-3.000
-2.625
-2.250
-1.875
-1.500
-1.125
-0.750
-0.375
-9.000
-8.625
-8.250
-7.875
-7.500
-7.125
-6.750
-6.375
-6.000
WM8958
7.875
8.250
8.625
9.000
9.375
9.750
10.125
10.500
10.875
11.250
11.625
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
3.000
3.375
3.750
4.125
4.500
4.875
5.250
5.625
6.000
6.375
6.750
7.125
7.500
Volume
(dB)
0.000
0.375
0.750
1.125
1.500
1.875
2.250
2.625
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
12.000
PP, August 2012, Rev 3.4
113
WM8958
Pre-Production
DAC SOFT MUTE AND SOFT UN-MUTE
The WM8958 has a soft mute function which ensures that a gradual attenuation is applied to the DAC outputs when the mute is asserted. The soft mute rate can be selected using the DAC_MUTERATE bit.
When a mute bit is disabled, the gain will either gradually ramp back up to the digital gain setting, or return instantly to the digital gain setting, depending on the DAC_SOFTMUTEMODE register bit. If the gradual un-mute ramp is selected (DAC_SOFTMUTEMODE = 1), then the un-mute rate is determined by the DAC_MUTERATE bit.
Note that each DAC is soft-muted by default. To play back an audio signal, the mute must first be disabled by setting the applicable mute control to 0 (see Table 58).
Soft Mute Mode would typically be enabled (DAC_SOFTMUTEMODE = 1) when using mute during playback of audio data so that when the mute is subsequently disabled, the volume increase will not create pop noise by jumping immediately to the previous volume level (e.g. resuming playback after pausing during a track).
Soft Mute Mode would typically be disabled (DAC_SOFTMUTEMODE = 0) when un-muting at the start of a music file, in order that the first part of the track is not attenuated (e.g. when starting playback of a new track, or resuming playback after pausing between tracks).
The DAC soft-mute function is illustrated in Figure 29 for DAC1L and DAC1R. The same function is applicable to DAC2L and DAC2R also.
DAC1L_VOL or DAC1R_VOL = [non-zero] = 00000000 = [non-zero]
DAC muting and un-muting using volume control bits
DAC1L_VOL and DAC1R_VOL
DAC1L_MUTE = 0
DAC1R_MUTE = 0
DAC_SOFTMUTEMODE = 0
DAC1L_MUTE = 1
DAC1R_MUTE = 1
DAC1L_MUTE = 0
DAC1R_MUTE = 0
DAC muting and un-muting using soft mute bits
DAC1L_MUTE or DAC1R_MUTE
Soft mute mode not enabled (DAC_SOFTMUTEMODE = 0).
DAC muting and un-muting using soft mute bit DAC_MUTE.
Soft mute mode enabled (DAC_SOFTMUTEMODE = 1).
DAC1L_MUTE = 0
DAC1R_MUTE = 0
DAC_SOFTMUTEMODE = 1
DAC1L_MUTE = 1
DAC1R_MUTE = 1
DAC1L_MUTE = 0
DAC1R_MUTE = 0
Figure 29 DAC Soft Mute Control
The DAC Soft Mute register controls are defined in Table 60.
The volume ramp rate during soft mute and un-mute is controlled by the DAC_MUTERATE bit. Ramp rates of fs/32 and fs/2 are selectable. The ramp rate determines the rate at which the volume will be increased or decreased. Note that the actual ramp time depends on the extent of the difference between the muted and un-muted volume settings. w PP, August 2012, Rev 3.4
114
Pre-Production
REGISTER
ADDRESS
R1556 (0614h)
DAC Softmute
BIT LABEL DEFAULT
1
0
DAC_SOFT
MUTEMODE
DAC_MUTE
RATE
0
0
WM8958
DESCRIPTION
DAC Unmute Ramp select
0 = Disabling soft-mute
(DAC[1/2][L/R]_MUTE=0) will cause the DAC volume to change immediately to DAC[1/2][L/R]_VOL settings
1 = Disabling soft-mute
(DAC[1/2][L/R]_MUTE=0) will cause the DAC volume to ramp up gradually to the DAC[1/2][L/R]_VOL settings
DAC Soft Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is 10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is 171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
Table 60 DAC Soft-Mute Control w PP, August 2012, Rev 3.4
115
WM8958
Pre-Production
ANALOGUE OUTPUT SIGNAL PATH
The WM8958 output routing and mixers provide a high degree of flexibility, allowing operation of many simultaneous signal paths through the device to a variety of analogue outputs. The outputs include a ground referenced headphone driver, two switchable class D/AB loudspeaker drivers, an ear speaker driver and four highly flexible line drivers. See “Analogue Outputs” for further details of these outputs.
The WM8958 output signal paths and control registers are illustrated in Figure 30. w
Figure 30 Control Registers for Output Signal Path
PP, August 2012, Rev 3.4
116
Pre-Production w
WM8958
OUTPUT SIGNAL PATHS ENABLE
The output mixers and drivers can be independently enabled and disabled as described in Table 61.
The supply rails for headphone outputs HPOUT1L and HPOUT1R are generated using an integrated dual-mode Charge Pump, which must be enabled whenever the headphone outputs are used. See the “Charge Pump” section for details on enabling and configuring this circuit.
Note that the headphone outputs HPOUT1L and HPOUT1R have dedicated output PGAs and volume controls. As a result, a low power consumption DAC playback path can be supported without needing to enable the output mixers MIXOUTL / MIXOUTR or the mixer output PGAs MIXOUTLVOL /
MIXOUTRVOL.
Note that the Headphone Outputs are also controlled by fields located within Register R96, which provide suppression of pops & clicks when enabling and disabling the HPOUT1L and HPOUT1R signal paths. These registers are described in the following “Headphone Signal Paths Enable” section.
Under recommended usage conditions, the Headphone Pop Suppression control bits will be configured by scheduling the default Start-Up and Shutdown sequences as described in the “Control
Write Sequencer” section. In these cases, the user does not need to set the register fields in R1 and
R96 directly.
For normal operation of the output signal paths, the reference voltage VMID and the bias current must also be enabled. See “Reference Voltages and Master Bias” for details of the associated controls
VMID_SEL and BIAS_ENA.
DESCRIPTION REGISTER
ADDRESS
R1 (0001h)
Power
Management
(1)
R3 (0003h)
Power
Management
(3)
13
12
11
9
8
13
12
SPKOUTR_ENA
SPKOUTL_ENA
HPOUT2_ENA
HPOUT1L_ENA
HPOUT1R_ENA
LINEOUT1N_ENA
LINEOUT1P_ENA
0
0
0
0
0
0
0
SPKMIXR Mixer, SPKRVOL PGA and SPKOUTR Output Enable
0 = Disabled
1 = Enabled
SPKMIXL Mixer, SPKLVOL PGA and SPKOUTL Output Enable
0 = Disabled
1 = Enabled
HPOUT2 Output Stage Enable
0 = Disabled
1 = Enabled
Enables HPOUT1L input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the
HPOUT1L Enable sequence.
Enables HPOUT1R input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the
HPOUT1R Enable sequence.
LINEOUT1N Line Out and
LINEOUT1NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT1P Line Out and
LINEOUT1PMIX Enable
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
117
WM8958
REGISTER
ADDRESS
R56 (0038h)
AntiPOP (1)
Pre-Production
DESCRIPTION
11 LINEOUT2N_ENA
10
9
8
7
6
5
4
6
LINEOUT2P_ENA
SPKRVOL_ENA
SPKLVOL_ENA
MIXOUTLVOL_ENA
MIXOUTRVOL_ENA
MIXOUTL_ENA
MIXOUTR_ENA
HPOUT2_IN_ENA
0
0
0
0
0
0
0
0
0
LINEOUT2N Line Out and
LINEOUT2NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT2P Line Out and
LINEOUT2PMIX Enable
0 = Disabled
1 = Enabled
SPKMIXR Mixer and SPKRVOL
PGA Enable
0 = Disabled
1 = Enabled
Note that SPKMIXR and SPKRVOL are also enabled when
SPKOUTR_ENA is set.
SPKMIXL Mixer and SPKLVOL
PGA Enable
0 = Disabled
1 = Enabled
Note that SPKMIXL and SPKLVOL are also enabled when
SPKOUTL_ENA is set.
MIXOUTL Left Volume Control
Enable
0 = Disabled
1 = Enabled
MIXOUTR Right Volume Control
Enable
0 = Disabled
1 = Enabled
MIXOUTL Left Output Mixer Enable
0 = Disabled
1 = Enabled
MIXOUTR Right Output Mixer
Enable
0 = Disabled
1 = Enabled
HPOUT2MIX Mixer and Input Stage
Enable
0 = Disabled
1 = Enabled
Table 61 Output Signal Paths Enable w PP, August 2012, Rev 3.4
118
Pre-Production
WM8958
HEADPHONE SIGNAL PATHS ENABLE
The HPOUT1L and HPOUT1R output paths can be actively discharged to AGND through internal resistors if desired. This is desirable at start-up in order to achieve a known output stage condition prior to enabling the VMID reference voltage. This is also desirable in shutdown to prevent the external connections from being affected by the internal circuits. The HPOUT1L and HPOUT1R outputs are shorted to AGND by default; the short circuit is removed on each of these paths by setting the applicable fields HPOUT1L_RMV_SHORT or HPOUT1R_RMV_SHORT.
The ground-referenced Headphone output drivers are designed to suppress pops and clicks when enabled or disabled. However, it is necessary to control the drivers in accordance with a defined sequence in start-up and shutdown to achieve the pop suppression. It is also necessary to schedule the DC Servo offset correction at the appropriate point in the sequence (see “DC Servo”). Table 62 and Table 63 describe the recommended sequences for enabling and disabling these output drivers.
Step 1
Step 2
Step 3
HPOUT1L_ENA = 1
HPOUT1R_ENA = 1
20 s delay
HPOUT1L_DLY = 1
HPOUT1R_DLY = 1
Step 4
DC offset correction
Step 5 HPOUT1L_OUTP = 1
HPOUT1L_RMV_SHORT = 1
HPOUT1R_OUTP = 1
HPOUT1R_RMV_SHORT = 1
Table 62 Headphone Output Enable Sequence
Step 1
Step 2
HPOUT1L_RMV_SHORT = 0
HPOUT1L_DLY = 0
HPOUT1L_OUTP = 0
HPOUT1R_RMV_SHORT = 0
HPOUT1R_DLY = 0
HPOUT1R_OUTP = 0
HPOUT1L_ENA = 0
HPOUT1R_ENA = 0
Table 63 Headphone Output Disable Sequence
The register bits relating to pop suppression control are defined in Table 64. w PP, August 2012, Rev 3.4
119
WM8958 w
REGISTER
ADDRESS
R1 (0001h)
Power
Management
(1)
R96 (0060h)
Analogue HP
(1)
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
9
8
7
6
5
3
2
1
HPOUT1L_ENA
HPOUT1R_ENA
HPOUT1L_RMV_
SHORT
HPOUT1L_OUTP
HPOUT1L_DLY
HPOUT1R_RMV_
SHORT
HPOUT1R_OUTP
HPOUT1R_DLY
0
0
0
0
0
0
0
0
Enables HPOUT1L input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the
HPOUT1L Enable sequence.
Enables HPOUT1R input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the
HPOUT1R Enable sequence.
Removes HPOUT1L short
0 = HPOUT1L short enabled
1 = HPOUT1L short removed
For normal operation, this bit should be set as the final step of the
HPOUT1L Enable sequence.
Enables HPOUT1L output stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the DC offset cancellation has been scheduled.
Enables HPOUT1L intermediate stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the output signal path has been configured, and before DC offset cancellation is scheduled. This bit should be set with at least 20us delay after HPOUT1L_ENA.
Removes HPOUT1R short
0 = HPOUT1R short enabled
1 = HPOUT1R short removed
For normal operation, this bit should be set as the final step of the
HPOUT1R Enable sequence.
Enables HPOUT1R output stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the DC offset cancellation has been scheduled.
Enables HPOUT1R intermediate stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the output signal path has been configured, and before DC offset cancellation is scheduled. This bit should be set with at least 20us delay after HPOUT1R_ENA.
Table 64 Headphone Output Signal Paths Control
PP, August 2012, Rev 3.4
120
Pre-Production
WM8958
OUTPUT MIXER CONTROL
The Output Mixer path select and volume controls are described in Table 65 for the Left Channel
(MIXOUTL) and Table 66 for the Right Channel (MIXOUTR). The gain of each of input path may be controlled independently in the range 0dB to -9dB.
Note that the DAC input levels may also be controlled by the DAC digital volume controls (see “Digital to Analogue Converter (DAC)”) and the Audio Interface digital volume controls (see “Digital Volume and Filter Control”).
When using the IN2LP, IN2LN, IN2RP or IN2RN signal paths to the output mixers, the buffered VMID reference must be enabled, using the VMID_BUF_ENA register, as described in “Reference Voltages and Master Bias”.
DESCRIPTION REGISTER
ADDRESS
R45 (002Dh)
Output Mixer
(1)
R49 (0031h)
Output Mixer
(5)
R45 (002Dh)
Output Mixer
(1)
R47 (002Fh)
Output Mixer
(3)
R45 (002Dh)
Output Mixer
(1)
R47 (002Fh)
Output Mixer
(3)
R45 (002Dh)
Output Mixer
(1)
5
8:6
4
8:6
2
2:0
3
IN2RN_TO_MIXOUTL
IN2RN_MIXOUTL_VOL
[2:0]
IN2LN_TO_MIXOUTL
IN2LN_MIXOUTL_VOL
[2:0]
IN1L_TO_MIXOUTL
IN1L_MIXOUTL_VOL
[2:0]
IN1R_TO_MIXOUTL
0
000
0
000
0
000
0
IN2RN to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2RN input to
MIXOUTL.
IN2RN to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2LN to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2LN input to
MIXOUTL.
IN2LN to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1L PGA Output to MIXOUTL
Mute
0 = Mute
1 = Un-mute
IN1L PGA Output to MIXOUTL
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1R PGA Output to MIXOUTL
Mute
0 = Mute
1 = Un-mute w PP, August 2012, Rev 3.4
121
WM8958 w
Pre-Production
REGISTER
ADDRESS
R47 (002Fh)
Output Mixer
(3)
5:3 IN1R_MIXOUTL_VOL
[2:0]
R45 (002Dh)
Output Mixer
(1)
R47 (002Fh)
Output Mixer
(3)
R45 (002Dh)
Output Mixer
(1)
R49 (0031h)
Output Mixer
(5)
R45 (002Dh)
Output Mixer
(1)
R49 (0031h)
Output Mixer
(5)
R45 (002Dh)
Output Mixer
(1)
R49 (0031h)
Output Mixer
(5)
1
11:9
7
5:3
6
2:0
0
IN2LP_TO_MIXOUTL
IN2LP_MIXOUTL_VOL
[2:0]
MIXINR_TO_MIXOUTL
MIXINR_MIXOUTL_VO
L [2:0]
MIXINL_TO_MIXOUTL
MIXINL_MIXOUTL_VOL
[2:0]
DAC1L_TO_MIXOUTL
11:9 DAC1L_MIXOUTL_VOL
[2:0]
000
0
000
0
000
0
000
0
000
DESCRIPTION
IN1R PGA Output to MIXOUTL
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2LP to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2LP input to
MIXOUTL.
IN2LP to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINR Output (Right ADC bypass) to MIXOUTL Mute
0 = Mute
1 = Un-mute
MIXINR Output (Right ADC bypass) to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINL Output (Left ADC bypass) to MIXOUTL Mute
0 = Mute
1 = Un-mute
MIXINL Output (Left ADC bypass) to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Left DAC1 to MIXOUTL Mute
0 = Mute
1 = Un-mute
Left DAC1 to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Table 65 Left Output Mixer (MIXOUTL) Control
PP, August 2012, Rev 3.4
122
Pre-Production w
REGISTER
ADDRESS
R46 (002Eh)
Output Mixer
(2)
R50 (0032h)
Output Mixer
(6)
8:6 IN2LN_MIXOUTR_VOL
[2:0]
000
R46 (002Eh)
Output Mixer
(2)
R46 (002Eh)
Output Mixer
(2)
R48 (0030h)
Output Mixer
(4)
3 IN1L_TO_MIXOUTR
5:3 IN1L_MIXOUTR_VOL
[2:0]
R46 (002Eh)
Output Mixer
(2)
R48 (0030h)
Output Mixer
(4)
2 IN1R_TO_MIXOUTR
2:0 IN1R_MIXOUTR_VOL
[2:0]
R46 (002Eh)
Output Mixer
(2)
5 IN2LN_TO_MIXOUTR
4 IN2RN_TO_MIXOUTR
1 IN2RP_TO_MIXOUTR
0
0
R48 (0030h)
Output Mixer
(4)
8:6 IN2RN_MIXOUTR_VOL
[2:0]
000
0
000
0
000
0
WM8958
DESCRIPTION
IN2LN to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2LN input to
MIXOUTR.
IN2LN to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2RN to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2RN input to
MIXOUTR.
IN2RN to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1L PGA Output to MIXOUTR Mute
0 = Mute
1 = Un-mute
IN1L PGA Output to MIXOUTR
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1R PGA Output to MIXOUTR
Mute
0 = Mute
1 = Un-mute
IN1R PGA Output to MIXOUTR
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2RP to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN2RP input to
MIXOUTR.
PP, August 2012, Rev 3.4
123
WM8958
REGISTER
ADDRESS
R48 (0030h)
Output Mixer
(4)
11:9 IN2RP_MIXOUTR_VOL
[2:0]
R46 (002Eh)
Output Mixer
(2)
R50 (0032h)
Output Mixer
(6)
R46 (002Eh)
Output Mixer
(2)
R50 (0032h)
Output Mixer
(6)
R46 (002Eh)
Output Mixer
(2)
R50 (0032h)
Output Mixer
(6)
7
5:3
6
2:0
0
Pre-Production
11:9
MIXINL_TO_MIXOUTR
MIXINL_MIXOUTR_VO
L[2:0]
MIXINR_TO_MIXOUTR
MIXINR_MIXOUTR_VO
L [2:0]
DAC1R_TO_MIXOUTR
DAC1R_MIXOUTR_VO
L [2:0]
000
0
000
0
000
0
000
DESCRIPTION
IN2RP to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINL Output (Left ADC bypass) to
MIXOUTR Mute
0 = Mute
1 = Un-mute
MIXINL Output (Left ADC bypass) to
MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINR Output (Right ADC bypass) to MIXOUTR Mute
0 = Mute
1 = Un-mute
MIXINR Output (Right ADC bypass) to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Right DAC1 to MIXOUTR Mute
0 = Mute
1 = Un-mute
Right DAC1 to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Table 66 Right Output Mixer (MIXOUTR) Control w PP, August 2012, Rev 3.4
124
Pre-Production
WM8958
SPEAKER MIXER CONTROL
The Speaker Mixer path select and volume controls are described in Table 67 for the Left Channel
(SPKMIXL) and Table 68 for the Right Channel (SPKMIXR).
Care should be taken when enabling more than one path to a speaker mixer in order to avoid clipping. The gain of each input path is adjustable using a selectable -3dB control in each path to facilitate this. Each Speaker Mixer output is also controlled by an additional independent volume control.
Note that the DAC input levels may also be controlled by the DAC digital volume controls (see “Digital to Analogue Converter (DAC)”) and the Audio Interface digital volume controls (see “Digital Volume and Filter Control”).
When using the IN1LP or IN1RP signal paths to the speaker mixers, the buffered VMID reference must be enabled, using the VMID_BUF_ENA register, as described in “Reference Voltages and
Master Bias”. w PP, August 2012, Rev 3.4
125
WM8958
REGISTER
ADDRESS
R54 (0034h)
Speaker Mixer
9
R34 (0022h)
SPKMIXL
Attenuation
7
5
3
1
6
5
4
3
2
Pre-Production
DESCRIPTION
1:0
DAC2L_TO_SPKMIXL
MIXINL_TO_SPKMIXL
IN1LP_TO_SPKMIXL
MIXOUTL_TO_SPKMIX
L
DAC1L_TO_SPKMIXL
DAC2L_SPKMIXL_VOL
MIXINL_SPKMIXL_VOL
IN1LP_SPKMIXL_VOL
MIXOUTL_SPKMIXL_V
OL
DAC1L_SPKMIXL_VOL
SPKMIXL_VOL [1:0]
0
0
0
0
0
0
0
0
0
0
11
Left DAC2 to SPKMIXL Mute
0 = Mute
1 = Un-mute
MIXINL (Left ADC bypass) to
SPKMIXL Mute
0 = Mute
1 = Un-mute
IN1LP to SPKMIXL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN1LP input to SPKMIXL.
Left Mixer Output to SPKMIXL
Mute
0 = Mute
1 = Un-mute
Left DAC1 to SPKMIXL Mute
0 = Mute
1 = Un-mute
Left DAC2 to SPKMIXL Fine
Volume Control
0 = 0dB
1 = -3dB
MIXINL (Left ADC bypass) to
SPKMIXL Fine Volume Control
0 = 0dB
1 = -3dB
IN1LP to SPKMIXL Fine Volume
Control
0 = 0dB
1 = -3dB
Left Mixer Output to SPKMIXL Fine
Volume Control
0 = 0dB
1 = -3dB
Left DAC1 to SPKMIXL Fine
Volume Control
0 = 0dB
1 = -3dB
Left Speaker Mixer Volume Control
00 = 0dB
01 = -6dB
10 = -12dB
11 = Mute
Table 67 Left Speaker Mixer (SPKMIXL) Control w PP, August 2012, Rev 3.4
126
Pre-Production
REGISTER
ADDRESS
R54 (0034h)
Speaker
Mixer
BIT LABEL DEFAULT
8 DAC2R_TO_SPKMIXR 0
6 MIXINR_TO_SPKMIXR 0
R35 (0023h)
SPKMIXR
Attenuation
4
2
0
6
5
4
3
2
1:0
IN1RP_TO_SPKMIXR
MIXOUTR_TO_SPKMIX
R
DAC1R_TO_SPKMIXR
DAC2R_SPKMIXR_VOL
MIXINR_SPKMIXR_VOL
IN1RP_SPKMIXR_VOL
MIXOUTR_SPKMIXR_V
OL
DAC1R_SPKMIXR_VOL
SPKMIXR_VOL [1:0]
0
0
0
0
0
0
0
0
11
WM8958
DESCRIPTION
Right DAC2 to SPKMIXR Mute
0 = Mute
1 = Un-mute
MIXINR (Right ADC bypass) to
SPKMIXR Mute
0 = Mute
1 = Un-mute
IN1RP to SPKMIXR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the IN1RP input to SPKMIXR.
Right Mixer Output to SPKMIXR
Mute
0 = Mute
1 = Un-mute
Right DAC1 to SPKMIXR Mute
0 = Mute
1 = Un-mute
Right DAC2 to SPKMIXR Fine
Volume Control
0 = 0dB
1 = -3dB
MIXINR (Right ADC bypass) to
SPKMIXR Fine Volume Control
0 = 0dB
1 = -3dB
IN1RP to SPKMIXR Fine Volume
Control
0 = 0dB
1 = -3dB
Right Mixer Output to SPKMIXR
Fine Volume Control
0 = 0dB
1 = -3dB
Right DAC1 to SPKMIXR Fine
Volume Control
0 = 0dB
1 = -3dB
Right Speaker Mixer Volume
Control
00 = 0dB
01 = -6dB
10 = -12dB
11 = Mute
Table 68 Right Speaker Mixer (SPKMIXR) Control w PP, August 2012, Rev 3.4
127
WM8958 w
Pre-Production
OUTPUT SIGNAL PATH VOLUME CONTROL
There are six output PGAs - MIXOUTLVOL, MIXOUTRVOL, HPOUT1LVOL, HPOUT1RVOL,
SPKLVOL and SPKRVOL. Each can be independently controlled, with MIXOUTLVOL and
MIXOUTRVOL providing volume control to both the earpiece and line drivers, HPOUT1LVOL and
HPOUT1RVOL to the headphone driver, and SPKLVOL and SPKRVOL to the speaker drivers.
The volume control of each of these output PGAs can be adjusted over a wide range of values. To minimise pop noise, it is recommended that only the MIXOUTLVOL, MIXOUTRVOL, HPOUT1LVOL,
HPOUT1RVOL, SPKLVOL and SPKRVOL are modified while the output signal path is active. Other gain controls are provided in the signal paths to provide scaling of signals from different sources, and to prevent clipping when multiple signals are mixed. However, to prevent pop noise, it is recommended that those other gain controls should not be modified while the signal path is active.
To prevent "zipper noise", a zero-cross function is provided on the output PGAs. When this feature is enabled, volume updates will not take place until a zero-crossing is detected. In the case of a long period without zero-crossings, a timeout function is provided. When the zero-cross function is enabled, the volume will update after the timeout period if no earlier zero-cross has occurred. The timeout clock is enabled using TOCLK_ENA; the timeout period is set by TOCLK_DIV. See “Clocking and Sample Rates” for more information on these fields.
The mixer output PGA controls are shown in Table 69. The MIXOUT_VU bits control the loading of the output mixer PGA volume data. When MIXOUT_VU is set to 0, the volume control data will be loaded into the respective control register, but will not actually change the gain setting. The output mixer PGA volume settings are both updated when a 1 is written to either MIXOUT_VU bit. This makes it possible to update the gain of both output paths simultaneously.
DESCRIPTION REGISTER
ADDRESS
R32 (0020h)
Left OPGA
Volume
8 MIXOUT_VU N/A
R33 (0021h)
Right OPGA
Volume
7
6
5:0
8
7
6
MIXOUTL_ZC
MIXOUTL_MUTE_N
MIXOUTL_VOL [5:0]
MIXOUT_VU
MIXOUTR_ZC
MIXOUTR_MUTE_N
0
1
39h
(0dB)
N/A
0
1
Mixer Output PGA Volume Update
Writing a 1 to this bit will update
MIXOUTLVOL and MIXOUTRVOL volumes simultaneously.
MIXOUTLVOL (Left Mixer Output
PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
MIXOUTLVOL (Left Mixer Output
PGA) Mute
0 = Mute
1 = Un-mute
MIXOUTLVOL (Left Mixer Output
PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
Mixer Output PGA Volume Update
Writing a 1 to this bit will update
MIXOUTLVOL and MIXOUTRVOL volumes simultaneously.
MIXOUTRVOL (Right Mixer Output
PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
MIXOUTLVOL (Right Mixer Output
PGA) Mute
0 = Mute
1 = Un-mute
PP, August 2012, Rev 3.4
128
Pre-Production w
WM8958
REGISTER
ADDRESS
DESCRIPTION
5:0 MIXOUTR_VOL [5:0] 39h
(0dB)
MIXOUTRVOL (Right Mixer Output
PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
Table 69 Mixer Output PGA (MIXOUTLVOL, MIXOUTRVOL) Control
The headphone output PGA is configurable between two input sources. The default input to each headphone output PGA is the respective output mixer (MIXOUTL or MIXOUTR). A direct path from the DAC1L or DAC1R can be selected using the DAC1L_TO_HPOUT1L and DAC1R_TO_HPOUT1R register bits. When these bits are selected, a DAC to Headphone playback path is possible without using the output mixers; this offers reduced power consumption by allowing the output mixers to be disabled in this typical usage case.
The headphone output PGA controls are shown in Table 70. The HPOUT1_VU bits control the loading of the headphone PGA volume data. When HPOUT1_VU is set to 0, the volume control data will be loaded into the respective control register, but will not actually change the gain setting. The headphone PGA volume settings are both updated when a 1 is written to either HPOUT1_VU bit. This makes it possible to update the gain of both output paths simultaneously.
DESCRIPTION REGISTER
ADDRESS
R28 (001Ch)
Left Output
Volume
8 HPOUT1_VU N/A
R45 (002Dh)
Output Mixer
(1)
R29 (001Dh)
Right Output
Volume
7
6
5:0
8
8
HPOUT1L_ZC
HPOUT1L_MUTE_N
HPOUT1L_VOL [5:0]
DAC1L_TO_HPOUT1
L
HPOUT1_VU
0
1
2Dh
(-12dB)
0
N/A
Headphone Output PGA Volume
Update
Writing a 1 to this bit will update
HPOUT1LVOL and HPOUT1RVOL volumes simultaneously.
HPOUT1LVOL (Left Headphone
Output PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
HPOUT1LVOL (Left Headphone
Output PGA) Mute
0 = Mute
1 = Un-mute
HPOUT1LVOL (Left Headphone
Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
HPOUT1LVOL (Left Headphone
Output PGA) Input Select
0 = MIXOUTL
1 = DAC1L
Headphone Output PGA Volume
Update
Writing a 1 to this bit will update
HPOUT1LVOL and HPOUT1RVOL volumes simultaneously.
PP, August 2012, Rev 3.4
129
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
R46 (002Eh)
Output Mixer
(2)
7
6
5:0
8
HPOUT1R_ZC
HPOUT1R_MUTE_N
HPOUT1R_VOL [5:0]
DAC1R_TO_HPOUT1
R
0
1
2Dh
(-12dB)
0
HPOUT1RVOL (Right Headphone
Output PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
HPOUT1RVOL (Right Headphone
Output PGA) Mute
0 = Mute
1 = Un-mute
HPOUT1RVOL (Right Headphone
Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
HPOUT1RVOL (Right Headphone
Output PGA) Input Select
0 = MIXOUTR
1 = DAC1R
Table 70 Headphone Output PGA (HPOUT1LVOL, HPOUT1RVOL) Control w PP, August 2012, Rev 3.4
130
Pre-Production
WM8958
The speaker output PGA controls are shown in Table 71.The SPKOUT_VU bits control the loading of the speaker PGA volume data. When SPKOUT_VU is set to 0, the volume control data will be loaded into the respective control register, but will not actually change the gain setting. The speaker PGA volume settings are both updated when a 1 is written to either SPKOUT_VU bit. This makes it possible to update the gain of both output paths simultaneously.
REGISTER
ADDRESS
R38 (0026h)
Speaker
Volume Left
R39 (0027h)
Speaker
Volume Right
8
7
6
5:0
8
7
6
5:0
SPKOUT_VU
SPKOUTL_ZC
SPKOUTL_MUTE_N
SPKOUTL_VOL [5:0]
SPKOUT_VU
SPKOUTR_ZC
SPKOUTR_MUTE_N
SPKOUTR_VOL [5:0]
N/A
0
1
39h
(0dB)
N/A
0
1
39h
(0dB)
DESCRIPTION
Speaker Output PGA Volume
Update
Writing a 1 to this bit will update
SPKLVOL and SPKRVOL volumes simultaneously.
SPKLVOL (Left Speaker Output
PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
SPKLVOL (Left Speaker Output
PGA) Mute
0 = Mute
1 = Un-mute
SPKLVOL (Left Speaker Output
PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
Speaker PGA Volume Update
Writing a 1 to this bit will update
SPKLVOL and SPKRVOL volumes simultaneously.
SPKRVOL (Right Speaker Output
PGA) Zero Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
SPKRVOL (Right Speaker Output
PGA) Mute
0 = Mute
1 = Un-mute
SPKRVOL (Right Speaker Output
PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
(See Table 72 for output PGA volume control range)
Table 71 Speaker Output PGA (SPKLVOL, SPKRVOL) Control w PP, August 2012, Rev 3.4
131
WM8958
Pre-Production
PGA GAIN SETTING VOLUME (dB) PGA GAIN SETTING VOLUME (dB)
00h -57 20h -25
01h -56 21h -24
02h -55 22h -23
03h -54 23h -22
04h -53 24h -21
05h -52 25h -20
06h -51 26h -19
07h -50 27h -18
08h -49 28h -17
09h -48 29h -16
0Ah -47 2Ah -15
0Bh -46 2Bh -14
0Ch -45 2Ch -13
0Dh -44 2Dh -12
0Eh -43 2Eh -11
0Fh -42 2Fh -10
10h -41 30h -9
11h -40 31h -8
12h -39 32h -7
13h -38 33h -6
14h -37 34h -5
15h -36 35h -4
16h -35 36h -3
17h -34 37h -2
18h -33 38h -1
19h -32 39h 0
1Ah -31 3Ah +1
1Bh -30 3Bh +2
1Ch -29 3Ch +3
1Dh -28 3Dh +4
1Eh -27 3Eh +5
1Fh -26 3Fh +6
Table 72 Output PGA Volume Range w PP, August 2012, Rev 3.4
132
Pre-Production
WM8958
SPEAKER BOOST MIXER
Each class D/AB speaker driver has its own boost mixer which performs a dual role. It allows the output from the left speaker mixer (via SPKLVOL), right speaker mixer (via SPKRVOL), or the ‘Direct
Voice’ path to be routed to either speaker driver. The speaker boost mixers are controlled using the registers defined in Table 73 below.
The ‘Direct Voice’ path is the differential input, VRXN-VRXP, routed directly to the output drivers, providing a low power differential path from baseband voice to loudspeakers. Note that a phase inversion exists between VRXP and SPKOUTxP. The ‘Direct Voice’ path output therefore represents
V
VRXN
- V
VRXP
.
The second function of the speaker boost mixers is that they provide an additional AC gain (boost) function to shift signal levels between the AVDD1 and SPKVDD voltage domains for maximum output power. The AC gain (boost) function is described in the “Analogue Outputs” section.
REGISTER
ADDRESS
R36 (0024h)
SPKOUT
Mixers
5
4
3
2
1
0
IN2LRP_TO_SPKOUT
L
SPKMIXL_TO_SPKOU
TL
SPKMIXR_TO_SPKO
UTL
IN2LRP_TO_SPKOUT
R
SPKMIXL_TO_SPKOU
TR
SPKMIXR_TO_SPKO
UTR
0
1
0
0
0
1
DESCRIPTION
Direct Voice (VRXN-VRXP) to Left
Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXL Left Speaker Mixer to
Left Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXR Right Speaker Mixer to
Left Speaker Mute
0 = Mute
1 = Un-mute
Direct Voice (VRXN-VRXP) to Right
Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXL Left Speaker Mixer to
Right Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXR Right Speaker Mixer to
Right Speaker Mute
0 = Mute
1 = Un-mute
Table 73 Speaker Boost Mixer (SPKOUTLBOOST, SPKOUTRBOOST) Control
EARPIECE DRIVER MIXER
The earpiece driver has a dedicated mixer, HPOUT2MIX, which is controlled using the registers defined in Table 74. The earpiece driver is configurable to select output from the left output mixer (via
MIXOUTLVOL), the right output mixer (via MIXOUTRVOL), or the ‘Direct Voice’ path.
The ‘Direct Voice’ path is the differential input, VRXN-VRXP, routed directly to the output drivers, providing a low power differential path from baseband voice to earpiece. Note that a phase inversion exists between VRXP and HPOUT2P. The ‘Direct Voice’ path output therefore represents V
VRXN
-
V
VRXP
.
Care should be taken to avoid clipping when enabling more than one path to the earpiece driver. The
HPOUT2VOL volume control can be used to avoid clipping when more than one full scale signal is input to the mixer. w PP, August 2012, Rev 3.4
133
WM8958 w
REGISTER
ADDRESS
R31 (001Fh)
HPOUT2
Volume
R51 (0033h)
HPOUT2
Mixer
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
5
4
5
4
3
HPOUT2_MUTE
HPOUT2_VOL
IN2LRP_TO_HPOUT2
MIXOUTLVOL_TO_HP
OUT2
MIXOUTRVOL_TO_HP
OUT2
1
0
0
0
0
HPOUT2 (Earpiece Driver) Mute
0 = Un-mute
1 = Mute
HPOUT2 (Earpiece Driver) Volume
0 = 0dB
1 = -6dB
Direct Voice (VRXN-VRXP) to
Earpiece Driver
0 = Mute
1 = Un-mute
MIXOUTLVOL (Left Output Mixer
PGA) to Earpiece Driver
0 = Mute
1 = Un-mute
MIXOUTRVOL (Right Output Mixer
PGA) to Earpiece Driver
0 = Mute
1 = Un-mute
Table 74 Earpiece Driver Mixer (HPOUT2MIX) Control
LINE OUTPUT MIXERS
The WM8958 provides two pairs of line outputs, both with highly configurable output mixers. The outputs LINEOUT1N and LINEOUT1P can be configured as two single-ended outputs or as a differential output. In the same manner, LINEOUT2N and LINEOUT2P can be configured either as two single-ended outputs or as a differential output. The respective line output mixers can be configured in single-ended mode or differential mode; each mode supports multiple signal path configurations.
LINEOUT1 single-ended mode is selected by setting LINEOUT1_MODE = 1. In single-ended mode, any of three possible signal paths may be enabled:
MIXOUTL (left output mixer) to LINEOUT1P
MIXOUTR (right output mixer) to LINEOUT1N
MIXOUTL (left output mixer) to LINEOUT1N
LINEOUT1 differential mode is selected by setting LINEOUT1_MODE = 0. In differential mode, any of three possible signal paths may be enabled:
MIXOUTL (left output mixer) to LINEOUT1N and LINEOUT1P
IN1L (input PGA) to LINEOUT1N and LINEOUT1P
IN1R (input PGA) to LINEOUT1N and LINEOUT1P
The LINEOUT1 output mixers are controlled as described in Table 75. Care should be taken to avoid clipping when enabling more than one path to the line output mixers. The LINEOUT1_VOL control can be used to provide -6dB attenuation when more than one full scale signal is applied.
When using the LINEOUT1 mixers in single-ended mode, a buffered VMID must be enabled. This is achieved by setting LINEOUT_VMID_BUF_ENA, as described in the “Analogue Outputs” section.
PP, August 2012, Rev 3.4
134
Pre-Production
REGISTER
ADDRESS
R30 (001Eh)
Line Outputs
Volume
R52 (0034h)
Line Mixer (1)
BIT LABEL DEFAULT
6 LINEOUT1N_MUTE 1
5
4
6
5
4
2
1
0
LINEOUT1P_MUTE
LINEOUT1_VOL
MIXOUTL_TO_LINEO
UT1N
MIXOUTR_TO_LINE
OUT1N
LINEOUT1_MODE
IN1R_TO_LINEOUT1
P
IN1L_TO_LINEOUT1
P
MIXOUTL_TO_LINEO
UT1P
1
0
0
0
0
0
0
0
WM8958
DESCRIPTION
LINEOUT1N Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT1P Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT1 Line Output Volume
0 = 0dB
1 = -6dB
Applies to both LINEOUT1N and
LINEOUT1P
MIXOUTL to Single-Ended Line
Output on LINEOUT1N
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 1)
MIXOUTR to Single-Ended Line
Output on LINEOUT1N
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 1)
LINEOUT1 Mode Select
0 = Differential
1 = Single-Ended
IN1R Input PGA to Differential Line
Output on LINEOUT1
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 0)
IN1L Input PGA to Differential Line
Output on LINEOUT1
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 0)
Differential Mode
(LINEOUT1_MODE = 0):
MIXOUTL to Differential Output on
LINEOUT1
0 = Mute
1 = Un-mute
Single Ended Mode
(LINEOUT1_MODE = 1):
MIXOUTL to Single-Ended Line
Output on LINEOUT1P
0 = Mute
1 = Un-mute
Table 75 LINEOUT1N and LINEOUT1P Control w PP, August 2012, Rev 3.4
135
WM8958
Pre-Production
LINEOUT2 single-ended mode is selected by setting LINEOUT2_MODE = 1. In single-ended mode, any of three possible signal paths may be enabled:
MIXOUTR (right output mixer) to LINEOUT2P
MIXOUTL (left output mixer) to LINEOUT2N
MIXOUTR (right output mixer) to LINEOUT2N
LINEOUT2 differential mode is selected by setting LINEOUT2_MODE = 0. In differential mode, any of three possible signal paths may be enabled:
MIXOUTR (right output mixer) to LINEOUT2N and LINEOUT2P
IN1L (input PGA) to LINEOUT2P and LINEOUT2P
IN1R (input PGA) to LINEOUT2N and LINEOUT2P
The LINEOUT2 output mixers are controlled as described in Table 76. Care should be taken to avoid clipping when enabling more than one path to the line output mixers. The LINEOUT2_VOL control can be used to provide -6dB attenuation when more than one full scale signal is applied.
When using the LINEOUT2 mixers in single-ended mode, a buffered VMID must be enabled. This is achieved by setting LINEOUT_VMID_BUF_ENA, as described in the “Analogue Outputs” section. w PP, August 2012, Rev 3.4
136
Pre-Production
REGISTER
ADDRESS
R30 (001Eh)
Line Outputs
Volume
BIT LABEL DEFAULT
2 LINEOUT2N_MUTE 1
1 LINEOUT2P_MUTE 1
R53 (0035h)
Line Mixer (2)
0
6
5
4
2
1
0
LINEOUT2_VOL
MIXOUTR_TO_LINEO
UT2N
MIXOUTL_TO_LINEO
UT2N
LINEOUT2_MODE
IN1L_TO_LINEOUT2P
IN1R_TO_LINEOUT2P
MIXOUTR_TO_LINEO
UT2P
0
0
0
0
0
0
0
WM8958
DESCRIPTION
LINEOUT2N Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT2P Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT2 Line Output Volume
0 = 0dB
1 = -6dB
Applies to both LINEOUT2N and
LINEOUT2P
MIXOUTR to Single-Ended Line
Output on LINEOUT2N
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 1)
MIXOUTL to Single-Ended Line
Output on LINEOUT2N
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 1)
LINEOUT2 Mode Select
0 = Differential
1 = Single-Ended
IN1L Input PGA to Differential Line
Output on LINEOUT2
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 0)
IN1R Input PGA to Differential Line
Output on LINEOUT2
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 0)
Differential Mode
(LINEOUT2_MODE = 0):
MIXOUTR to Differential Output on
LINEOUT2
0 = Mute
1 = Un-mute
Single-Ended Mode
(LINEOUT2_MODE = 0):
MIXOUTR to Single-Ended Line
Output on LINEOUT2P
0 = Mute
1 = Un-mute
Table 76 LINEOUT2N and LINEOUT2P Control w PP, August 2012, Rev 3.4
137
WM8958
CHARGE PUMP
Pre-Production
The WM8958 incorporates a dual-mode Charge Pump which generates the supply rails for the headphone output drivers, HPOUT1L and HPOUT1R.
The Charge Pump has a single supply input, CPVDD, and generates split rails CPVOUTP and
CPVOUTN according to the selected mode of operation.
The Charge Pump connections are illustrated in Figure 31 (see “Applications Information” for external component values). An input decoupling capacitor may also be required at CPVDD, depending upon the system configuration.
CPCA CPCB
CPVDD
Charge Pump
CPVOUTP
CPVOUTN w
CPGND
Figure 31 Charge Pump External Connections
The Charge Pump is enabled by setting the CP_ENA bit. When enabled, the charge pump adjusts the output voltages (CPVOUTP and CPVOUTN) as well as the switching frequency in order to optimise the power consumption according to the operating conditions. This can take two forms, which are selected using the CP_DYN_PWR register bit.
Register control (CP_DYN_PWR = 0)
Dynamic control (CP_DYN_PWR = 1)
Under Register control, the HPOUT1L_VOL and HPOUT1R_VOL register settings are used to control the charge pump mode of operation.
Under Dynamic control, the audio signal level in the digital audio interface is used to control the charge pump mode of operation. The CP_DYN_SRC_SEL register determines which of the digital signal paths is used for this function - this may be AIF1 Timeslot 0, AIF Timeslot 1 or AIF2. The
CP_DYN_SRC_SEL should be set according to the active source for the HPOUT1L and HPOUT1R outputs.
The Dynamic Charge Pump Control mode is the Wolfson ‘Class W’ mode, which allows the power consumption to be optimised in real time, but can only be used if a single AIF source is the only signal source. The Class W mode should not be used if any of the bypass paths are used to feed analogue inputs into the output signal path, or if more than one AIF source is used to feed the headphone output via the Digital Mixers.
Under the recommended usage conditions of the WM8958, the Charge Pump will be enabled by running the default headphone Start-Up sequence as described in the “Control Write Sequencer” section. (Similarly, it will be disabled by running the Shut-Down sequence.) In these cases, the user does not need to write to the CP_ENA bit. The Charge Pump operating mode defaults to Register control; Dynamic control may be selected by setting the CP_DYN_PWR register bit, if appropriate.
Note that the charge pump clock is derived from internal clock SYSCLK; either MCLK or the FLL output selectable using the SYSCLK_SRC bit. Under normal circumstances an external clock signal must be present for the charge pump to function. However, the FLL has a free-running mode that does not require an external clock but will generate an internal clock suitable for running the charge pump. The clock division from SYSCLK is handled transparently by the WM8958 without user
PP, August 2012, Rev 3.4
138
Pre-Production
WM8958 intervention, as long as SYSCLK and sample rates are set correctly. Refer to the “Clocking and
Sample Rates” section for more detail on the FLL and clocking configuration.
When the Charge Pump is disabled, the output can be left floating or can be actively discharged, depending on the CP_DISCH control bit.
If the headphone output drivers (HPOUT1L and HPOUT1R) are not used, then the Charge Pump and the associated external components are not required. The Charge Pump and Headphone drivers should not be enabled in this case (CP_ENA=0, HPOUT1L_ENA=0, HPOUT1R_ENA=0).
If the Charge Pump is not used, and the associated external components are omitted, then the CPCA and CPCB pins can be left floating; the CPVOUTP and CPVOUTN pins should be grounded as illustrated in Figure 32.
Note that, when the Charge Pump is disabled, it is still recommended that the CPVDD pin is kept within its recommended operating conditions. w
Figure 32 External Configuration when Charge Pump not used
The Charge Pump control fields are described in Table 77.
REGISTER
ADDRESS
R76 (004Ch)
Charge Pump
(1)
R77 (004Dh)
Charge Pump
(2)
15
15
CP_ENA
CP_DISCH
0
1
R81 (0051h)
Class W (1)
9:8
0
CP_DYN_SRC_SEL
CP_DYN_PWR
00
0
Enable charge-pump digits
0 = Disable
1 = Enable
Charge Pump Discharge Select
0 = Charge Pump outputs floating when disabled
1 = Charge Pump outputs discharged when disabled
Selects the digital audio source for envelope tracking
00 = AIF1, DAC Timeslot 0
01 = AIF1, DAC Timeslot 1
10 = AIF2, DAC data
11 = Reserved
Enable dynamic charge pump power control
0 = charge pump controlled by volume register settings (Class G)
1 = charge pump controlled by real-time audio level (Class W)
Table 77 Charge Pump Control
PP, August 2012, Rev 3.4
139
WM8958
DC SERVO w
Pre-Production
The WM8958 provides a DC servo circuit on the headphone outputs HPOUT1L and HPOUT1R in order to remove DC offset from these ground-referenced outputs. When enabled, the DC servo ensures that the DC level of these outputs remains within 1mV of ground. Removal of the DC offset is important because any deviation from GND at the output pin will cause current to flow through the load under quiescent conditions, resulting in increased power consumption. Additionally, the presence of DC offsets can result in audible pops and clicks at power up and power down.
The recommended usage of the DC Servo is initialised by running the default Start-Up sequence as described in the “Control Write Sequencer” section. The default Start-Up sequence executes a series of DC offset corrections, after which the measured offset correction is maintained on the headphone output channels. If a different usage is required, eg. if a periodic DC offset correction is required, then the default Start-Up sequence may be modified according to specific requirements. The relevant control fields are described in the following paragraphs and are defined in Table 78.
DC SERVO ENABLE AND START-UP
The DC Servo circuit is enabled on HPOUT1L and HPOUT1R by setting DCS_ENA_CHAN_0 and
DCS_ENA_CHAN_1 respectively. When the DC Servo is enabled, the DC offset correction can be commanded in a number of different ways, including single-shot and periodically recurring events.
Writing a logic 1 to DCS_TRIG_STARTUP_n initiates a series of DC offset measurements and applies the necessary correction to the associated output; (‘n’ = 0 for Left channel, 1 for Right channel). On completion, the headphone output will be within 1mV of AGND. This is the DC Servo mode selected by the default Start-Up sequence. Completion of the DC offset correction triggered in this way is indicated by the DCS_STARTUP_COMPLETE field, as described in Table 78. Typically, this operation takes 86ms per channel.
For correct operation of the DC Servo Start-Up mode, it is important that there is no active audio signal present on the signal path while the mode is running. The DC Servo Start-Up mode should be scheduled at the correct position within the Headphone Output Enable sequence, as described in the
“Analogue Output Signal Path” section. All other stages of the analogue signal path should be fully enabled prior to commanding the Start-Up mode; the DAC Digital Mute function should be used, where appropriate, to ensure there is no active audio signal present during the DC Servo measurements.
Writing a logic 1 to DCS_TRIG_DAC_WR_n causes the DC offset correction to be set to the value contained in the DCS_DAC_WR_VAL_n fields in Register R87. This mode is useful if the required offset correction has already been determined and stored; it is faster than the
DCS_TRIG_STARTUP_n mode, but relies on the accuracy of the stored settings. Completion of the
DC offset correction triggered in this way is indicated by the DCS_DAC_WR_COMPLETE field, as described in Table 78. Typically, this operation takes 2ms per channel.
For pop-free operation of the DC Servo DAC Write mode, it is important that the mode is scheduled at the correct position within the Headphone Output Enable sequence, as described in the “Analogue
Output Signal Path” section.
The current DC offset value for each Headphone output channel can be read from the
DCS_DAC_WR_VAL_n fields. These values may form the basis of settings that are subsequently used by the DC Servo in DAC Write mode. Note that these fields have a different definition for Read and Write, as described in Table 78.
When using either of the DC Servo options above, the status of the DC offset correction process is indicated by the DCS_CAL_COMPLETE field; this is the logical OR of the
DCS_STARTUP_COMPLETE and DCS_DAC_WR_COMPLETE fields.
The DCS_DAC_WR_COMPLETE bits can be used as inputs to the Interrupt control circuit or used to generate an external logic signal on a GPIO pin. See “Interrupts” and “General Purpose Input/Output” for further details.
The DC Servo control fields associated with start-up operation are described in Table 78. It is important to note that, to minimise audible pops/clicks, the Start-Up and DAC Write modes of DC
Servo operation should be commanded as part of a control sequence which includes muting and shorting of the headphone outputs; a suitable sequence is defined in the default Start-Up sequence.
PP, August 2012, Rev 3.4
140
Pre-Production w
REGISTER
ADDRESS
R84 (0054h)
DC Servo (1)
R87 (0057h)
DC Servo (4)
R88 (0058h)
DC Servo
Readback
5
4
3
2
15:8
7:0
9:8
DCS_TRIG_START
UP_1
DCS_TRIG_START
UP_0
DCS_TRIG_DAC_W
R_1
DCS_TRIG_DAC_W
R_0
1 DCS_ENA_CHAN_1
0 DCS_ENA_CHAN_0
DCS_DAC_WR_VA
L_1 [7:0]
DCS_DAC_WR_VA
L_0 [7:0]
DCS_CAL_COMPL
ETE [1:0]
WM8958
0
0
0
0
0
0
00h
00h
00
Writing 1 to this bit selects Start-
Up DC Servo mode for
HPOUT1R.
In readback, a value of 1 indicates that the DC Servo
Start-Up correction is in progress.
Writing 1 to this bit selects Start-
Up DC Servo mode for
HPOUT1L.
In readback, a value of 1 indicates that the DC Servo
Start-Up correction is in progress.
Writing 1 to this bit selects DAC
Write DC Servo mode for
HPOUT1R.
In readback, a value of 1 indicates that the DC Servo DAC
Write correction is in progress.
Writing 1 to this bit selects DAC
Write DC Servo mode for
HPOUT1L.
In readback, a value of 1 indicates that the DC Servo DAC
Write correction is in progress.
DC Servo enable for HPOUT1R
0 = Disabled
1 = Enabled
DC Servo enable for HPOUT1L
0 = Disabled
1 = Enabled
Writing to this field sets the DC
Offset value for HPOUT1R in
DAC Write DC Servo mode.
Reading this field gives the current DC Offset value for
HPOUT1R.
Two’s complement format.
LSB is 0.25mV.
Range is -32mV to +31.75mV
Writing to this field sets the DC
Offset value for HPOUT1L in
DAC Write DC Servo mode.
Reading this field gives the current DC Offset value for
HPOUT1L.
Two’s complement format.
LSB is 0.25mV.
Range is -32mV to +31.75mV
DC Servo Complete status
0 = DAC Write or Start-Up DC
Servo mode not completed.
1 = DAC Write or Start-Up DC
Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
PP, August 2012, Rev 3.4
141
WM8958
Pre-Production
REGISTER
ADDRESS
5:4
1:0
DCS_DAC_WR_CO
MPLETE [1:0]
DCS_STARTUP_C
OMPLETE [1:0]
00
00
DC Servo DAC Write status
0 = DAC Write DC Servo mode not completed.
1 = DAC Write DC Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
DC Servo Start-Up status
0 = Start-Up DC Servo mode not completed.
1 = Start-Up DC Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
Table 78 DC Servo Enable and Start-Up Modes
DC SERVO ACTIVE MODES
The DC Servo modes described above are suitable for initialising the DC offset correction circuit on the Headphone outputs as part of a controlled start-up sequence which is executed before the signal path is fully enabled. Additional modes are available for use whilst the signal path is active; these modes may be of benefit following a large change in signal gain, which can lead to a change in DC offset level. Periodic updates may also be desirable to remove slow drifts in DC offset caused by changes in parameters such as device temperature.
The DC Servo circuit is enabled on HPOUT1L and HPOUT1R by setting DCS_ENA_CHAN_0 and
DCS_ENA_CHAN_1 respectively, as described earlier in Table 78.
Writing a logic 1 to DCS_TRIG_SINGLE_n initiates a single DC offset measurement and adjustment to the associated output; (‘n’ = 0 for Left channel, 1 for Right channel). This will adjust the DC offset correction on the selected channel by no more than 1LSB (0.25mV).
Setting DCS_TIMER_PERIOD_01 to a non-zero value will cause a single DC offset measurement and adjustment to be scheduled on a periodic basis. Periodic rates ranging from every 0.52s to in excess of 2 hours can be selected.
Writing a logic 1 to DCS_TRIG_SERIES_n initiates a series of DC offset measurements and applies the necessary correction to the associated output. The number of DC Servo operations performed is determined by DCS_SERIES_NO_01. A maximum of 128 operations may be selected, though a much lower value will be sufficient in most applications.
The DC Servo control fields associated with active modes (suitable for use on a signal path that is in active use) are described in Table 79. w PP, August 2012, Rev 3.4
142
Pre-Production w
WM8958
REGISTER
ADDRESS
R84 (0054h)
DC Servo (1)
R85 (0055h)
DC Servo (2)
13
12
9
8
DCS_TRIG_SINGLE
_1
DCS_TRIG_SINGLE
_0
DCS_TRIG_SERIES
_1
DCS_TRIG_SERIES
_0
11:5 DCS_SERIES_NO_
01 [6:0]
3:0 DCS_TIMER_PERI
OD_01 [3:0]
0
0
0
0
Writing 1 to this bit selects a single DC offset correction for
HPOUT1R.
In readback, a value of 1 indicates that the DC Servo single correction is in progress.
Writing 1 to this bit selects a single DC offset correction for
HPOUT1L.
In readback, a value of 1 indicates that the DC Servo single correction is in progress.
Writing 1 to this bit selects a series of DC offset corrections for HPOUT1R.
In readback, a value of 1 indicates that the DC Servo DAC
Write correction is in progress.
Writing 1 to this bit selects a series of DC offset corrections for HPOUT1L.
In readback, a value of 1 indicates that the DC Servo DAC
Write correction is in progress.
010 1010 Number of DC Servo updates to perform in a series event.
0 = 1 update
1 = 2 updates
...
127 = 128 updates
1010 Time between periodic updates.
Time is calculated as
0.251s x (2^PERIOD), where PERIOD =
DCS_TIMER_PERIOD_01.
0000 = Off
0001 = 0.502s
….
1010 = 257s (4min 17s)
1111 = 8225s (2hr 17min)
Table 79 DC Servo Active Modes
GPIO / INTERRUPT OUTPUTS FROM DC SERVO
When using the DC Servo Start-Up or DAC Write modes, the DCS_CAL_COMPLETE register provides readback of the status of the DC offset correction. This can be read from register R88 as described in Table 78.
The DCS_CAL_COMPLETE bits can also be used as inputs to the Interrupt control circuit and used to trigger an Interrupt event - see “Interrupts”.
The DCS_CAL_COMPLETE bits can also be used as inputs to the GPIO function and used to generate external logic signals indicating the DC Servo status. See “General Purpose Input/Output” for details of how to configure a GPIO pin to output the DC Servo status.
PP, August 2012, Rev 3.4
143
WM8958
Pre-Production
ANALOGUE OUTPUTS
The speaker, headphone, earpiece and line outputs are highly configurable and may be used in many different ways.
SPEAKER OUTPUT CONFIGURATIONS
The speaker outputs SPKOUTL and SPKOUTR can be driven by either of the speaker mixers,
SPKMIXL or SPKMIXR, or by the low power, differential Direct Voice path from IN2LP/VRXN and
IN2RP/VRXP. Fine volume control is available on the speaker mixer paths using the SPKLVOL and
SPKRVOL PGAs. A boost function is available on both the speaker mixer paths and the Direct Voice path. For information on the speaker mixing options, refer to the “Analogue Output Signal Path” section.
The speaker outputs SPKOUTL and SPKOUTR operate in a BTL configuration in Class AB or Class
D amplifier modes. The default mode is class D but class AB mode can be selected by setting the
SPKOUT_CLASSAB register bit, as defined in Table 81.
The speaker outputs can be configured as a pair of stereo outputs, or as a single mono output. Note that, for applications requiring only a single speaker output, it is possible to improve the THD performance by configuring the speaker outputs in mono mode. See “Typical Performance” for further details.
The mono configuration is selected by applying a logic high input to the SPKMODE pin (A4), as described in Table 80. For Stereo mode this pin should be connected to GND. Note that SPKMODE is referenced to DBVDD1.
An internal pull-up resistor is enabled by default on the SPKMODE pin; this can be configured using the SPKMODE_PU register bit described in Table 81.
SPEAKER CONFIGURATION
Stereo Mode
Mono Mode
Table 80 SPKMODE Pin Function
SPKMODE PIN (A4)
GND
DBVDD1
In the mono configuration, the P channels, SPKOUTLP and SPKOUTRP should be connected together on the PCB, and similarly with the N channels, SPKOUTLN and SPKOUTRN, as illustrated in Figure 33. In this configuration both left and right speaker drivers should be enabled
(SPKOUTL_ENA=1 and SPKOUTR_ENA=1), but path selection and volume controls are available on left channel only (SPKMIXL, SPKLVOL and SPKOUTLBOOST).
Note that the minimum speaker load resistance and the maximum power output has a dependency on the SPKMODE output configuration, and also on the Class D/AB mode selection. See “Electrical
Characteristics” for further details. w
Stereo Mono
Figure 33 Stereo / Mono Speaker Output Configurations
PP, August 2012, Rev 3.4
144
Pre-Production
WM8958
Eight levels of AC signal boost are provided in order to deliver maximum output power for many commonly-used SPKVDD/AVDD1 combinations. (Note that SPKVDD1 powers the Left Speaker driver, and SPKVDD2 powers the Right Speaker driver; it is assumed that SPKVDD1 = SPKVDD2 =
SPKVDD.)
The signal boost options are available in both Class AB and Class D modes. The AC boost levels from 0dB to +12dB are selected using register bits SPKOUTL_BOOST and SPKOUTR_BOOST. To prevent pop noise, SPKOUTL_BOOST and SPKOUTR_BOOST should not be modified while the speaker outputs are enabled. Figure 34 illustrates the speaker outputs and the mixing and gain/boost options available.
Ultra-low leakage and high PSRR allow the speaker supply SPKVDD to be directly connected to a lithium battery. Note that an appropriate SPKVDD supply voltage must be provided to prevent waveform clipping when speaker boost is used.
DC gain is applied automatically in both class AB and class D modes with a shift from VMID to
SPKVDD/2. This provides optimum signal swing for maximum output power. In class AB mode, an ultra-high PSRR mode is available, in which the DC reference for the speaker driver is fixed at VMID.
This mode is selected by enabling the SPKAB_REF_SEL bit (see Table 81). In this mode, the output power is limited but the driver will still be capable of driving more than 500mW in 8 while maintaining excellent suppression of noise on SPKVDD (for example, TDMA noise in a GSM phone application).
The AC and DC gain functions are illustrated in Figure 34. w
Figure 34 Speaker Output Configuration and AC Boost Operation
PP, August 2012, Rev 3.4
145
WM8958
REGISTER
ADDRESS
R35 (0023h)
SPKMIXR
Attenuation
R37 (0025h)
ClassD
R34 (0022h)
SPKMIXL
Attenuation
Pre-Production
8
5:3
2:0
8
SPKOUT_CLASSAB
SPKOUTL_BOOST
[2:0]
SPKOUTR_BOOST
[2:0]
SPKAB_REF_SEL
0
000
(1.0x)
000
(1.0x)
0
DESCRIPTION
Speaker Class AB Mode Enable
0 = Class D mode
1 = Class AB mode
Left Speaker Gain Boost
000 = 1.00x boost (+0dB)
001 = 1.19x boost (+1.5dB)
010 = 1.41x boost (+3.0dB)
011 = 1.68x boost (+4.5dB)
100 = 2.00x boost (+6.0dB)
101 = 2.37x boost (+7.5dB)
110 = 2.81x boost (+9.0dB)
111 = 3.98x boost (+12.0dB)
Right Speaker Gain Boost
000 = 1.00x boost (+0dB)
001 = 1.19x boost (+1.5dB)
010 = 1.41x boost (+3.0dB)
011 = 1.68x boost (+4.5dB)
100 = 2.00x boost (+6.0dB)
101 = 2.37x boost (+7.5dB)
110 = 2.81x boost (+9.0dB)
111 = 3.98x boost (+12.0dB)
Selects Reference for Speaker in
Class AB mode
0 = SPKVDD/2
1 = VMID
SPKMODE Pull-up enable
0 = Disabled
1 = Enabled
R1825
(0721h)
Pull Control
(2)
1 SPKMODE_PU
Table 81 Speaker Mode and Boost Control
1
Clocking of the Class D output driver is derived from SYSCLK. The clocking frequency division is configured automatically, according to the AIFn_SR and AIFnCLK_RATE registers. (See “Clocking and Sample Rates” for further details of the system clocks and control registers.)
The Class D switching clock is enabled whenever SPKOUTL_ENA or SPKOUTR_ENA is set, provided also that SPKOUT_CLASSAB = 0. The frequency is as described in Table 82.
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then the Class D clock frequency is controlled by the AIF1_SR and AIF1CLK_RATE registers.
When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then the Class D clock frequency is controlled by the AIF2_SR and AIF2CLK_RATE registers.
Note that the applicable clocks (SYSCLK, AIF1CLK or AIF2CLK) must be present and enabled when using the speaker outputs in Class D mode. w PP, August 2012, Rev 3.4
146
Pre-Production
WM8958
SAMPLE
RATE (kHz)
8
11.025
12
16
22.05
24
32
44.1
48
88.2
96
SYSCLK RATE (AIFnCLK / fs ratio)
128 192 256 384 512 768 1024 1536
256 256 341.3 256 341.3 256 341.3 256
352.8 352.8 352.8 352.8 352.8 352.8 352.8
384 384 384 384 384 384 384
341.3 384 341.3 384 341.3 384
352.8 352.8 352.8 352.8 352.8
384 384 384 384 384
341.3 384 341.3 384
352.8 352.8 352.8
384 384 384
352.8
384
Table 82 Class D Switching Frequency (kHz)
HEADPHONE OUTPUT CONFIGURATIONS
The headphone outputs HPOUT1L and HPOUT1R are driven by the headphone output PGAs
HPOUT1LVOL and HPOUT1RVOL. Each PGA has its own dedicated volume control, as described in the “Analogue Output Signal Path” section. The input to these PGAs can be either the output mixers
MIXOUTL and MIXOUTR or the direct DAC1 outputs DAC1L and DAC1R.
The headphone output driver is capable of driving up to 30mW into a 16Ω load or 25mW into a 32Ω load such as a stereo headset or headphones. The outputs are ground-referenced, eliminating any requirement for AC coupling capacitors. This is achieved by having separate positive and negative supply rails powered by an on-chip charge pump. A DC Servo circuit removes any DC offset from the headphone outputs, suppressing ‘pop’ noise and minimising power consumption. The Charge Pump and DC Servo are described separately (see “Charge Pump” and “DC Servo” respectively).
It is recommended to connect a zobel network to the headphone output pins HPOUT1L and
HPOUT1R for best audio performance in all applications. The components of the zobel network have the effect of dampening high frequency oscillations or instabilities that can arise outside the audio band under certain conditions. Possible sources of these instabilities include the inductive load of a headphone coil or an active load in the form of an external line amplifier. The capacitance of lengthy cables or PCB tracks can also lead to amplifier instability. The zobel network should comprise of a
20 resistor and 100nF capacitor in series with each other, as illustrated in Figure 35.
If any ground-referenced headphone output is not used, then the zobel network components can be omitted from the corresponding output pin, and the pin can be left floating. The respective headphone driver(s) should not be enabled in this case. w
Figure 35 Zobel Network Components for HPOUT1L and HPOUT1R
PP, August 2012, Rev 3.4
147
WM8958 w
Pre-Production
The headphone output incorporates a common mode, or ground loop, feedback path which provides rejection of system-related ground noise. The return path is via HPOUT1FB. This pin must be connected to ground for normal operation of the headphone output. No register configuration is required.
Note that the HPOUT1FB pin should be connected to GND close to the headphone jack, as illustrated in Figure 35.
EARPIECE DRIVER OUTPUT CONFIGURATIONS
The earpiece driver outputs HPOUT2P and HPOUT2N are driven by the HPOUT2MIX output mixer, which can take inputs from the mixer output PGAs MIXOUTLVOL and MIXOUTRVOL, or from the low power, differential Direct Voice path IN2LP/VRXN and IN2RP/VRXP. Fine volume control is available on the output mixer paths using MIXOUTLVOL and MIXOUTRVOL. A selectable -6dB attenuation is available on the HPOUT2MIX output, as described in Table 74 (refer to the “Analogue Output Signal
Path” section).
The earpiece outputs are designed to operate in a BTL configuration, driving 50mW into a typical 16 ear speaker.
For suppression of pop noise there are two separate enables for the earpiece driver; HPOUT2_ENA enables the output stage and HPOUT2_IN_ENA enables the mixer and input stage.
HPOUT2_IN_ENA should be enabled a minimum of 50 s before HPOUT2_ENA – see “Control Write
Sequencer” section for an example power sequence.
LINE OUTPUT CONFIGURATIONS
The four line outputs LINEOUT1P, LINEOUT1N, LINEOUT2P and LINEOUT2N provide a highly flexible combination of differential and single-ended configurations, each driven by a dedicated output mixer. There is a selectable -6dB gain option in each mixer to avoid clipping when mixing more than one signal into a line output. Additional volume control is available at other locations within each of the supported signal paths. For more information about the line output mixing options, refer to the
“Analogue Output Signal Path” section.
Typical applications for the line outputs (single-ended or differential) are:
Handset or headset microphone output to external voice CODEC
Stereo line output
Output to external speaker driver(s) to support additional loudspeakers
When single-ended mode is selected for either LINEOUT1 or LINEOUT2, a buffered VMID must be enabled as a reference for the outputs. This is enabled by setting the LINEOUT_VMID_BUF_ENA bit as defined in Table 83.
DESCRIPTION REGISTER
ADDRESS
R56 (0038h)
AntiPOP (1)
7 LINEOUT_VMID_BUF_E
NA
0 Enables VMID reference for line outputs in single-ended mode
0 = Disabled
1 = Enabled
Table 83 LINEOUT VMID Buffer for Single-Ended Operation
Some example line output configurations are listed and illustrated below.
Differential line output from Mic/Line input on IN1L PGA
Differential line output from Mic/Line input on IN1R PGA
Stereo differential line output from output mixers MIXOUTL and MIXOUTR
Stereo single-ended line output from output mixer to either LINEOUT1 or LINEOUT2
Mono single-ended line output from output mixer
PP, August 2012, Rev 3.4
148
Pre-Production
WM8958
IN1L
IN1R
LINEOUT1NMIX
MIXOUTLVOL
MIXOUTRVOL
IN1R
IN1L
+
0dB or -6dB
LINEOUT1PMIX
MIXOUTLVOL
IN1L
IN1R
+
0dB or -6dB
Ground Loop
Noise Rejection
Ground Loop
Noise Rejection
LINEOUT1N
LINEOUT1P
Min = -57dB
Max = +6dB
Step = 1dB
MIXOUTLVOL
Min = -57dB
Max = +6dB
Step = 1dB
MIXOUTRVOL
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=0
LINEOUT2_MODE=0
IN1L_TO_LINEOUT1P=1
IN1R_TO_LINEOUT2P=1
Figure 36 Differential Line Out from input PGA
IN1L (to LINEOUT1) and IN1R (to LINEOUT2)
IN1L
IN1R
LINEOUT2NMIX
MIXOUTLVOL
MIXOUTRVOL
IN1R
+
IN1L
0dB or -6dB
LINEOUT2PMIX
MIXOUTRVOL
IN1L
IN1R
+
0dB or -6dB
Ground Loop
Noise Rejection
LINEOUT2N
LINEOUT2P
Ground Loop
Noise Rejection
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=0
LINEOUT2_MODE=0
IN1R_TO_LINEOUT1P=1
IN1L_TO_LINEOUT2P=1
Figure 37 Differential Line Out from input PGA
IN1R (to LINEOUT1) and IN1L (to LINEOUT2)
IN1L
IN1R
LINEOUT1NMIX
MIXOUTLVOL
MIXOUTRVOL
IN1R
+
IN1L
0dB or -6dB
LINEOUT1PMIX
MIXOUTLVOL
IN1L
IN1R
+
0dB or -6dB
Ground Loop
Noise Rejection
Ground Loop
Noise Rejection
LINEOUT1N
LINEOUT1P
Min = -57dB
Max = +6dB
Step = 1dB
MIXOUTLVOL
Min = -57dB
Max = +6dB
Step = 1dB
MIXOUTRVOL
IN1L
IN1R
LINEOUT2NMIX
MIXOUTLVOL
MIXOUTRVOL
IN1R
IN1L
+
0dB or -6dB
LINEOUT2PMIX
MIXOUTRVOL
IN1L
IN1R
+
0dB or -6dB
Ground Loop
Noise Rejection
Ground Loop
Noise Rejection
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=1
MIXOUTL_TO_LINEOUT1P=1
MIXOUTR_TO_LINEOUT1N=1
LINEOUT_VMID_BUF_ENA=1
LINEOUT2N
LINEOUT2P
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=0
LINEOUT2_MODE=0
MIXOUTL_TO_LINEOUT1P=1
MIXOUTR_TO_LINEOUT2P=1
Figure 38 Stereo Differential Line Out from
MIXOUTL and MIXOUTR
Figure 39 Stereo Single-Ended Line Out from
MIXOUTL and MIXOUTR to LINEOUT1 w PP, August 2012, Rev 3.4
149
WM8958
Pre-Production
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=1
MIXOUTL_TO_LINEOUT2N=1
MIXOUTR_TO_LINEOUT2P=1
LINEOUT_VMID_BUF_ENA=1
LINEOUT1N_MUTE=0, LINEOUT1P_MUTE=0
LINEOUT2N_MUTE=0, LINEOUT2P_MUTE=0
LINEOUT1_MODE=1
LINEOUT2_MODE=1
MIXOUTL_TO_LINEOUT1N=1 and/or
MIXOUTL_TO_LINEOUT1P=1
MIXOUTR_TO_LINEOUT2N=1 and/or
MIXOUTR_TO_LINEOUT2P=1
LINEOUT_VMID_BUF_ENA=1
Figure 40 Stereo Single-Ended Line Out from
MIXOUTL and MIXOUTR to LINEOUT2
Figure 41 Mono Line Out to LINEOUT1N,
LINEOUT1P, LINEOUT2N, LINEOUT2P
The line outputs incorporate a common mode, or ground loop, feedback path which provides rejection of system-related ground noise. The return path, via LINEOUTFB, is enabled separately for
LINEOUT1 and LINEOUT2 using the LINEOUT1_FB and LINEOUT2_FB bits as defined in Table 84.
Ground loop feedback is a benefit to single-ended line outputs only; it is not applicable to differential outputs, which already inherently offer common mode noise rejection.
DESCRIPTION REGISTER
ADDRESS
R55 (0037h)
Additional
Control
7
6
LINEOUT1_FB
LINEOUT2_FB
0
0
Enable ground loop noise feedback on LINEOUT1
0 = Disabled
1 = Enabled
Enable ground loop noise feedback on LINEOUT2
0 = Disabled
1 = Enabled
Table 84 Line Output Ground Loop Feedback Enable w PP, August 2012, Rev 3.4
150
Pre-Production
WM8958
EXTERNAL ACCESSORY DETECTION
The WM8958 accessory detection circuit measures the impedance of an external load connected to the MICDET pin. This feature can be used to detect the insertion or removal of a microphone, and the status of the associated hookswitch. It can also be used to detect push-button status or the connection of other external accessories.
The microphone detection circuit measures the impedance connected to MICDET, and reports whether the measured impedance lies within one of 9 pre-defined levels (including the ‘no accessory detected’ level). This means it can detect the presence of a typical microphone and up to 7 pushbuttons. One of the impedance levels is specifically designed to detect a video accessory (typical
75Ω) load if required.
The microphone detection circuit uses the MICBIAS2 output as a reference. The WM8958 will automatically enable MICBIAS2 when required in order to perform the detection function; this allows the detection function to be supported in low-power standby operating conditions.
Microphone detection is enabled by setting the MICD_ENA register. When microphone detection is enabled, the WM8958 performs a number of measurements in order to determine the MICDET impedance. The measurement process is repeated at a cyclic rate controlled by MICD_RATE. (The
MICD_RATE register selects the delay between completion of one measurement and the start of the next.)
For best accuracy, the measured impedance is only deemed valid after more than one successive measurement has produced the same result. The MICD_DBTIME register provides control of the debounce period; this can be either 2 measurements or 4 measurements.
When the microphone detection result has settled (ie. after the applicable de-bounce period), the
WM8958 indicates valid data by setting the MICD_VALID bit. The measured impedance is indicated using the MICD_LVL and MICD_STS register bits, as described in Table 85.
The MICD_VALID bit, when set, remains asserted for as long as the microphone detection function is enabled (ie. while MICD_ENA = 1). If the detected impedance changes, then the MICD_LVL and
MICD_STS fields will change, but the MICD_VALID bit will remain set, indicating valid data at all times.
Note that the impedance levels quoted in the MICD_LVL description assume that a microphone
(475Ω to 30kΩ impedance) is also present on the MICDET pin. The limits quoted in the “Electrical
Characteristics” refer to the combined effective impedance on the MICDET pin. Typical external components are described in the “Applications Information” section.
The microphone detection reports a measurement result in one of the pre-defined impedance levels.
Each measurement level can be enabled or disabled independently; this provides flexibility according to the required thresholds, and offers a faster measurement time in some applications. The
MICD_LVL_SEL register is described in detail later in this section.
Clocking for the microphone detection function is derived from SYSCLK (defined in the “Clocking and
Sample Rates” section).
When AIF1CLK is selected as the SYSCLK source (SYSCLK_SRC = 0), then AIF1CLK must be present and enabled when using the accessory detect function. The AIF1_SR and AIF1CLK_RATE registers must be set to values that are consistent with the available AIF1CLK frequency.
When AIF2CLK is selected as the SYSCLK source (SYSCLK_SRC = 1), then AIF2CLK must be present and enabled when using the accessory detect function. The AIF2_SR and AIF2CLK_RATE registers must be set to values that are consistent with the available AIF2CLK frequency.
The Frequency Locked Loop (FLL) free-running mode provides flexibility to clock the microphone detection function without any external reference clock, eg. in low-power standby operating conditions. See “Clocking and Sample Rates” for details of the WM8958 clocking options and FLL.
The accessory detection function can also be supported using a low frequency (eg. 32kHz) clock, as described later in this section, see “Accessory Detection with Low Frequency SYSCLK”.
The microphone detection function is an input to the Interrupt control circuit and can be used to trigger an Interrupt event every time an accessory insertion, removal or impedance change is detected. See
“Interrupts” for further details. w PP, August 2012, Rev 3.4
151
WM8958
Pre-Production
The microphone detection function can also generate a GPIO output, providing an external indication of the microphone detection. This GPIO output is pulsed every time an accessory insertion, removal or impedance change is detected. See “General Purpose Input/Output” for details of how to configure a GPIO pin to output the microphone detection signal.
The register fields associated with Microphone Detection (or other accessories) are described in
Table 85. The external circuit configuration is illustrated in Figure 42.
DESCRIPTION REGISTER
ADDRESS
R208
(00D0h)
Mic Detect 1
15:12
11:8
1
0
MICD_BIAS_STARTTI
ME [3:0]
MICD_RATE [3:0]
MICD_DBTIME
MICD_ENA
0101
0110
0
0
Mic Detect Bias Startup Delay
(If MICBIAS2 is not enabled already, this field selects the delay time allowed for MICBIAS2 to startup prior to performing the MICDET function.)
0000 = 0ms (continuous)
0001 = 0.25ms
0010 = 0.5ms
0011 = 1ms
0100 = 2ms
0101 = 4ms
0110 = 8ms
0111 = 16ms
1000 = 32ms
1001 = 64ms
1010 = 128ms
1011 = 256ms
1100 to 1111 = 512ms
Mic Detect Rate
(Selects the delay between successive Mic Detect measurements.)
0000 = 0ms (continuous)
0001 = 0.25ms
0010 = 0.5ms
0011 = 1ms
0100 = 2ms
0101 = 4ms
0110 = 8ms
0111 = 16ms
1000 = 32ms
1001 = 64ms
1010 = 128ms
1011 = 256ms
1100 to 1111 = 512ms
Mic Detect De-bounce
0 = 2 measurements
1 = 4 measurements
Mic Detect Enable
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
152
Pre-Production w
WM8958
REGISTER
ADDRESS
R209
(00D1h)
Mic Detect 2
R210
(00D2h)
Mic Detect 3
7:0
10:2
1
0
MICD_LVL_SEL [7:0]
MICD_LVL [8:0]
MICD_VALID
MICD_STS
DESCRIPTION
0111_
1111
0_0000_
0000
0
0
Mic Detect Level Select
(enables Mic Detection in specific impedance ranges)
[7] = Not used - must be set to 0
[6] = Enable >475 ohm detection
[5] = Enable 326 ohm detection
[4] = Enable 152 ohm detection
[3] = Enable 77 ohm detection
[2] = Enable 47.6 ohm detection
[1] = Enable 29.4 ohm detection
[0] = Enable 14 ohm detection
Note that the impedance values quoted assume that a microphone
(475ohm-30kohm) is also present on the MICDET pin.
Mic Detect Level
(indicates the measured impedance)
[8] = Not used
[7] = >475 ohm, <30k ohm
[6] = 326 ohm
[5] = 152 ohm
[4] = 77 ohm
[3] = 47.6 ohm
[2] = 29.4 ohm
[1] = 14 ohm
[0] = <3 ohm
Note that the impedance values quoted assume that a microphone
(475ohm-30kohm) is also present on the MICDET pin.
Mic Detect Data Valid
0 = Not Valid
1 = Valid
Mic Detect Status
0 = No Mic Accessory present
(impedance is >30k ohm)
1 = Mic Accessory is present
(impedance is <30k ohm)
Table 85 Microphone Detect Control
The external connections for the Microphone Detect circuit are illustrated in Figure 42. In typical applications, it can be used to detect a microphone or button press.
The microphone detection function uses MICBIAS2 as a reference. The microphone detection function will automatically enable MICBIAS2 when required for MICDET impedance measurement.
If MICBIAS2 is not already enabled (ie. if MICB2_ENA = 0), then MICBIAS2 will be enabled for short periods of time only, every time the impedance measurement is scheduled. To allow time for the
MICBIAS2 source to start-up, a time delay is applied before the measurement is performed; this is configured using the MICD_BIAS_STARTTIME register, as described in Table 85.
The MICD_BIAS_STARTTIME register should be set to 16ms or more if MICB2_RATE = 1 (pop-free start-up / shut-down). The MICD_BIAS_STARTTIME register should be set to 0.25ms or more if
MICB2_RATE = 0 (fast start-up / shut-down).
If the MICBIAS2 reference is not enabled continuously (ie. if MICB2_ENA = 0), then the MICBIAS2 discharge bit (MICB2_DISCH) should be set to 0.
PP, August 2012, Rev 3.4
153
WM8958
Pre-Production
The MICBIAS sources are configured using the registers described in Table 1, in the “Analogue Input
Signal Path” section. w
Figure 42 Microphone Detect Interface
The MICD_LVL_SEL [7:0] register bits allow each of the impedance measurement levels to be enabled or disabled independently. This allows the function to be tailored to the particular application requirements.
If one or more bits within the MICD_LVL_SEL register is set to 0, then the corresponding impedance level will be disabled. Any measured impedance which lies in a disabled level will be reported as the next lowest, enabled level.
For example, the MICD_LVL_SEL [3] bit enables the detection of impedances around 77 . If
MICD_LVL_SEL [3] = 0, then an external impedance of 77 will not be indicated as 77 but will be indicated as 47 ; this would be reported in the MICD_LVL register as MICD_LVL [3] = 1.
With all measurement levels enabled, the WM8958 can detect the presence of a typical microphone and up to 7 push-buttons. The microphone detect function is specifically designed to detect a video accessory (typical 75 ) load if required.
See “Applications Information” for typical recommended external components for microphone, video or push-button accessory detection.
The microphone detection circuit assumes that a 2.2k
(2%) resistor is connected to MICBIAS2, as illustrated. Different resistor values will lead to inaccuracy in the impedance measurement.
The measurement accuracy of the microphone detect function is assured whenever the connected load is within the applicable limits specified in the “Electrical Characteristics”. Note that a 2.2k
(2%) resistor must also be connected between MICDET and MICBIAS2.
Note that the connection of a microphone will change the measured impedance on the MICDET pin; see “Applications Information” for recommended components for typical applications.
The measurement time varies between 100 s and 500s according to the impedance of the external load. A high impedance will be measured faster than a low impedance.
PP, August 2012, Rev 3.4
154
Pre-Production
WM8958
The timing of the microphone detect function is illustrated in Figure 43. Two different cases are shown, according to whether MICBIAS2 is enabled periodically by the impedance measurement function (MICB2_ENA=0), or is enabled at all times (MICB2_ENA=1).
Figure 43 Microphone Detect Timing
ACCESSORY DETECTION WITH LOW FREQUENCY SYSCLK
Clocking for the microphone detection function can be derived from AIF1CLK or AIF2CLK, as described earlier.
Under normal circumstances, the AIFn_SR and AIFnCLK_RATE registers must be set to values that are consistent with the available AIFnCLK frequency. The register settings support AIFnCLK frequencies of 1.024MHz or higher.
The microphone detection function can also be supported using a low frequency (eg. 32kHz) clock. In this case, the selected SYSCLK source (AIF1CLK or AIF2CLK) should be configured with the following register settings:
AIFnCLK_RATE = 0001 (AIFnCLK / fs = 128)
AIFn_SR = 0000 (fs = 8kHz).
The register settings above configure the WM8958 for AIFnCLK = 1.024MHz. If the available clock is a different frequency (eg. 32kHz), then the timings set by the MICD_RATE and
MICD_BIAS_STARTUP registers will be scaled accordingly. In the case of a 32kHz clock, these times will be extended by a factor of 32 (calculated as 1024000 / 32000).
For example, under normal circumstances, setting MICD_RATE = 0011 selects a 1ms delay between successive measurements. Using a 32kHz reference clock, and the register settings above, then
MICD_RATE = 0011 will select a 32ms delay. w PP, August 2012, Rev 3.4
155
WM8958
Pre-Production
GENERAL PURPOSE INPUT/OUTPUT
The WM8958 provides a number of GPIO functions to enable interfacing and detection of external hardware and to provide logic outputs to other devices. The input functions can be polled directly or can be used to generate an Interrupt (IRQ) event. The GPIO and Interrupt circuits support the following functions:
Alternate interface functions (AIF2, AIF3)
Button detect (GPIO input)
Logic ‘1’ and logic ‘0’ output (GPIO output)
Interrupt (IRQ) status output
Over-Temperature
Microphone accessory status detection
Frequency Locked Loop (FLL) Lock status output
Sample Rate Conversion (SRC) Lock status output
Dynamic Range Control (DRC) Signal activity detection
Control Write Sequencer status output
Digital Core FIFO error status output
Clock output (SYSCLK divided by OPCLK_DIV)
Frequency Locked Loop (FLL) Clock output
GPIO CONTROL
For each GPIO, the selected function is determined by the GPn_FN field, where n identifies the GPIO pin (1, 6, 8, 9, 10, 11). The pin direction, set by GPn_DIR, must be set according to function selected by GPn_FN.
The alternate audio interfaces AIF2 and AIF3 are both supported using GPIO pins; the applicable pin functions are selected by setting the corresponding GPn_FN register to 00h. See Table 87 for the definition of which AIF function is available on each GPIO pin.
See “Digital Audio Interface Control” for details of AIF2 and AIF3.
When a pin is configured as a GPIO input (GPn_DIR = 1), the logic level at the pin can be read from the respective GPn_LVL bit. Note that GPn_LVL is not affected by the GPn_POL bit.
A de-bounce circuit can be enabled on any GPIO input, to avoid false event triggers. This is enabled on each pin by setting the respective GPn_DB bit.
When a pin is configured as a Logic Level output (GPn_DIR = 0, GPn_FN = 01h), its level can be set to logic 0 or logic 1 using the GPn_LVL field.
When a pin is configured as an output (GPn_DIR = 0), the polarity can be inverted using the
GPn_POL bit. When GPn_POL = 1, then the selected output function is inverted. In the case of Logic
Level output (GPn_FN = 01h), the external output will be the opposite logic level to GPn_LVL when
GPn_POL = 1.
A GPIO output can be either CMOS driven or Open Drain. This is selected on each pin using the respective GPn_OP_CFG bit.
Internal pull-up and pull-down resistors may be enabled using the GPn_PU and GPn_PD fields; this allows greater flexibility to interface with different signals from other devices. (Note that if GPn_PU and GPn_PD are both set for any GPIO pin, then the pull-up and pull-down will be disabled.)
Each of the GPIO pins is an input to the Interrupt control circuit and can be used to trigger an Interrupt event. An interrupt event is triggered on the rising and falling edge of the GPIO input. The associated interrupt bit is latched once set; it can be polled at any time or used to control the IRQ signal. See
“Interrupts” for more details of the Interrupt event handling. w PP, August 2012, Rev 3.4
156
Pre-Production w
WM8958
The register fields that control the GPIO pins are described in Table 86.
REGISTER
ADDRESS
R1792
(0700h)
GPIO 1
R1797
(0705h)
GPIO 6
R1799
(0707h)
GPIO 8 to
R1802
(070Ah)
GPIO 11
15
14
13
10
GPn_DIR
GPn_PU
GPn_PD
GPn_POL
1
0
1
0
DESCRIPTION
GPIOn Pin Direction
0 = Output
1 = Input
GPIOn Pull-Up Enable
0 = Disabled
1 = Enabled
GPIOn Pull-Down Enable
0 = Disabled
1 = Enabled
GPIOn Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9
8
6
GPn_OP_CFG
GPn_DB
GPn_LVL
0
1
0
GPIOn Output Configuration
0 = CMOS
1 = Open Drain
GPIOn Input De-bounce
0 = Disabled
1 = Enabled
GPIOn level. Write to this bit to set a GPIO output. Read from this bit to read GPIO input level.
For output functions only, when
GPn_POL is set, the register contains the opposite logic level to the external pin.
4:0 GPn_FN [4:0] GPIOn Pin Function
(see Table 87 for details)
GP1_FN default = 0000
GP6_FN default = 0001
GP8_FN default = 0001
GP9_FN default = 0001
GP10_FN default = 0001
GP11_FN default = 0001
Note: n is a number (1, 6, 8, 9, 10, 11) that identifies the individual GPIO.
Table 86 GPIO1, GPIO6, GPIO8, GPIO9, GPIO10 to GPIO11 Control
GPIO FUNCTION SELECT
The available GPIO functions are described in Table 87. The function of each GPIO is set using the
GPn_FN register, where n identifies the GPIO pin (1, 6, 8, 9, 10, 11). Note that the respective
GPn_DIR must also be set according to whether the function is an input or output.
GPn_FN DESCRIPTION
00h GPIO1 - ADCLRCLK1
GPIO6 - ADCLRCLK2
GPIO8 - DACDAT3
GPIO9 - ADCDAT3
GPIO10 - LRCLK3
GPIO11 - BCLK3
COMMENTS
Alternate Audio Interface connections.
01h
02h
Button detect input /
Logic level output
GPn_DIR = 0: GPIO pin logic level is set by GPn_LVL.
GPn_DIR = 1: Button detect or logic level input.
Reserved
PP, August 2012, Rev 3.4
157
WM8958 w
Pre-Production
GPn_FN DESCRIPTION
03h IRQ Interrupt (IRQ) output
0 = IRQ not asserted
1 = IRQ asserted
04h Temperature
(Shutdown) status output
COMMENTS
Indicates Temperature Shutdown Sensor status
0 = Temperature is below shutdown level
1 = Temperature is above shutdown level
05h
06h
07h
08h
09h
Microphone Detect Microphone Detect (MICDET accessory) IRQ output
A single 31 s pulse is output whenever an accessory insertion, removal or impedance change is detected.
Reserved
Reserved
0Ah
0Bh
Reserved
FLL1 Lock Indicates FLL1 Lock status
0 = Not locked
1 = Locked
FLL2 Lock
SRC1 Lock
Indicates FLL2 Lock status
0 = Not locked
1 = Locked
Indicates SRC1 Lock status
0 = Not locked
1 = Locked
0Ch
0Dh
0Eh
0Fh
SRC2 Lock
AIF1 DRC1 Signal
Detect
AIF1 DRC2 Signal
Detect
AIF2 DRC Signal
Detect
Indicates SRC2 Lock status
0 = Not locked
1 = Locked
Indicates AIF1 DRC1 Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
Indicates AIF1 DRC2 Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
Indicates AIF2 DRC Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
10h
11h
12h
13h
14h
Write Sequencer
Status
FIFO Error
Clock Output OPCLK
Temperature (Warning) status output
DC Servo Done
Indicates Write Sequencer status
0 = Write Sequencer Idle
1 = Write Sequence Busy
Indicates a Digital Core FIFO Error condition
0 = Normal operation
1 = FIFO Error
GPIO Clock derived from SYSCLK
Indicates Temperature Warning Sensor status
0 = Temperature is below warning level
1 = Temperature is above warning level
Indicates DC Servo status on HPOUT1L and HPOUT1R
0 = DC Servo not complete
1 = DC Servo complete
15h
16h
FLL1 Clock Output
FLL2 Clock Output
Clock output from FLL1
Clock output from FLL2
17h to 1Fh Reserved
Table 87 GPIO Function Select
PP, August 2012, Rev 3.4
158
Pre-Production
WM8958
BUTTON DETECT (GPIO INPUT)
Button detect functionality can be selected on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”. The same functionality can be used to support a Jack Detect input function.
It is recommended to enable the GPIO input de-bounce feature when using GPIOs as button input or
Jack Detect input.
The GPn_LVL fields may be read to determine the logic levels on a GPIO input, after the selectable de-bounce controls. Note that GPn_LVL is not affected by the GPn_POL bit.
The de-bounced GPIO signals are also inputs to the Interrupt control circuit. An interrupt event is triggered on the rising and falling edge of the GPIO input. The associated interrupt bits are latched once set; it can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
LOGIC ‘1’ AND LOGIC ‘0’ OUTPUT (GPIO OUTPUT)
The WM8958 can be programmed to drive a logic high or logic low level on any GPIO pin by selecting the “GPIO Output” function as described in “GPIO Control”. The output logic level is selected using the respective GPn_LVL bit.
Note that the polarity of the GPIO output can be inverted using the GPn_POL registers. If GPn_POL =
1, then the external output will be the opposite logic level to GPn_LVL.
INTERRUPT (IRQ) STATUS OUTPUT
The WM8958 has an Interrupt Controller which can be used to indicate when any selected Interrupt events occur. An interrupt can be generated by any of the events described throughout the GPIO function definition above. Individual interrupts may be masked in order to configure the Interrupt as required. See “Interrupts” for further details.
The Interrupt (IRQ) status may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
OVER-TEMPERATURE DETECTION
The WM8958 incorporates a temperature sensor which detects when the device temperature is within normal limits or if the device is approaching a hazardous temperature condition.
The Temperature status may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”. Any GPIO pin can be used to indicate either a Warning
Temperature event or the Shutdown Temperature event. De-bounce can be applied to the applicable signal using the register bits described in Table 88.
The Warning Temperature and Shutdown Temperature status are inputs to the Interrupt control circuit, after the selectable de-bounce. An interrupt event may be triggered on the rising and falling edges of these signals. The associated interrupt bit is latched once set; it can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
Note that the temperature sensor can be configured to automatically disable the audio outputs of the
WM8958 (see “Thermal Shutdown”). In some applications, it may be preferable to manage the temperature sensor event through GPIO or Interrupt functions, allowing a host processor to implement a controlled system response to an over-temperature condition.
The temperature sensor must be enabled by setting the TSHUT_ENA register bit. When the
TSHUT_OPDIS is also set, then a device over-temperature condition will cause the speaker outputs
(SPKOUTL and SPKOUTR) of the WM8958 to be disabled. w PP, August 2012, Rev 3.4
159
WM8958
Pre-Production
REGISTER
ADDRESS
R2 (0002h)
Power
Management
(2)
R1864
(0748h)
IRQ
Debounce
BIT LABEL DEFAULT
14
13
0
0
TSHUT_EN
A
TSHUT_OP
DIS
TEMP_WAR
N_DB
TEMP_SHU
T_DB
1
1
0
0
DESCRIPTION
Thermal sensor enable
0 = Disabled
1 = Enabled
Thermal shutdown control
(Causes audio outputs to be disabled if an overtemperature occurs. The thermal sensor must also be enabled.)
0 = Disabled
1 = Enabled
Thermal Warning de-bounce
0 = Disabled
1 = Enabled
Thermal shutdown de-bounce
0 = Disabled
1 = Enabled
Table 88 Temperature Sensor Enable and GPIO/Interrupt Control
MICROPHONE ACCESSORY STATUS DETECTION
The WM8958 provides an impedance measurement circuit on the MICDET pin to detect the connection of a microphone or other external accessory. See “External Accessory Detection” for further details.
A logic signal from the microphone detect circuit may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”. This logic signal is set high for a single pulse duration of 31 s whenever an accessory insertion, removal or impedance change is detected.
The microphone detection circuit is also an input to the Interrupt control circuit. An interrupt event is triggered whenever an accessory insertion, removal or impedance change is detected. The associated interrupt bit is latched once set; it can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
FREQUENCY LOCKED LOOP (FLL) LOCK STATUS OUTPUT
The WM8958 maintains a flag indicating the lock status of each of FLLs, which may be used to control other events if required. See “Clocking and Sample Rates” for more details of the FLL.
The FLL Lock signals may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
The FLL Lock signals are inputs to the Interrupt control circuit. An interrupt event is triggered on the rising and falling edges of the FLL Lock signals. The associated interrupt bits are latched once set; they can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
SAMPLE RATE CONVERTER (SRC) LOCK STATUS OUTPUT
The WM8958 maintains a flag indicating the lock status of each of Sample Rate Converters, which may be used to control other events if required. See “Sample Rate Conversion” for more details of the
Sample Rate Converters. w
The SRC Lock signals may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
The SRC Lock signals are inputs to the Interrupt control circuit, after the selectable de-bounce. An interrupt event is triggered on the rising and falling edges of the SRC Lock signals. The associated interrupt bits are latched once set; they can be polled at any time or used to control the IRQ signal.
See “Interrupts” for more details of the Interrupt event handling.
PP, August 2012, Rev 3.4
160
Pre-Production
WM8958
DYNAMIC RANGE CONTROL (DRC) SIGNAL ACTIVITY DETECTION
Signal activity detection is provided on each of the Dynamic Range Controllers (DRCs). These may be configured to indicate when a signal is present on the respective signal path. The signal activity status signals may be used to control other events if required. See “Digital Core Architecture” for more details of the DRCs and the available digital signal paths.
When a DRC is enabled, as described in “Dynamic Range Control (DRC)”, then signal activity detection can be enabled by setting the respective [DRC]_SIG_DET register bit. The applicable threshold can be defined either as a Peak level (Crest Factor) or an RMS level, depending on the
[DRC]_SIG_DET_MODE register bit. When Peak level is selected, the threshold is determined by
[DRC]_SIG_DET_PK, which defines the applicable Crest Factor (Peak to RMS ratio) threshold. If
RMS level is selected, then the threshold is set using [DRC]_SIG_DET_RMS. These register fields are set independently for each of the three Dynamic Range Controllers, as described in Table 89.
When the DRC is enabled in any of the ADC (digital record) paths, the associated High Pass Filter
(HPF) must be enabled also; this ensures that DC offsets are removed prior to the DRC processing.
The output path HPF control registers are described in Table 42 (for AIF1 output paths) and Table 50
(for AIF2 output paths). These are described in the “Digital Volume and Filter Control” section.
The DRC Signal Detect signals may be output directly on any GPIO pin by setting the respective
GPIO registers as described in “GPIO Control”.
The DRC Signal Detect signals are inputs to the Interrupt control circuit. An interrupt event is triggered on the rising edge of the DRC Signal Detect signals. The associated interrupt bits are latched once set; they can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1088
(0440h)
AIF1 DRC1
(1)
15:11
10:9
7
6
AIF1DRC1_SIG_
DET_RMS [4:0]
AIF1DRC1_SIG_
DET_PK [1:0]
AIF1DRC1_SIG_
DET_MODE
AIF1DRC1_SIG_
DET
00000
00
1
0
AIF1 DRC1 Signal Detect RMS
Threshold.
This is the RMS signal level for signal detect to be indicated when
AIF1DRC1_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
11110 = -75dB
11111 = -76.5dB
AIF1 DRC1 Signal Detect Peak
Threshold.
This is the Peak/RMS ratio, or Crest
Factor, level for signal detect to be indicated when
AIF1DRC1_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC1 Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF1 DRC1 Signal Detect Enable
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
161
WM8958 w
Pre-Production
REGISTER
ADDRESS
R1104
(0450h)
AIF1 DRC2
(1)
R1344
(0540h)
AIF2 DRC (1)
BIT LABEL DEFAULT
15:11
10:9
7
6
15:11
10:9
7
6
AIF1DRC2_SIG_
DET_RMS [4:0]
AIF1DRC2_SIG_
DET_PK [1:0]
AIF1DRC2_SIG_
DET_MODE
AIF1DRC2_SIG_
DET
AIF2DRC_SIG_D
ET_RMS [4:0]
AIF2DRC_SIG_D
ET_PK [1:0]
AIF2DRC_SIG_D
ET_MODE
AIF2DRC_SIG_D
ET
00000
00
1
0
00000
00
1
0
Table 89 DRC Signal Activity Detect GPIO/Interrupt Control
DESCRIPTION
AIF1 DRC2 Signal Detect RMS
Threshold.
This is the RMS signal level for signal detect to be indicated when
AIF1DRC2_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
11110 = -75dB
11111 = -76.5dB
AIF1 DRC2 Signal Detect Peak
Threshold.
This is the Peak/RMS ratio, or Crest
Factor, level for signal detect to be indicated when
AIF1DRC2_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC2 Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF1 DRC2 Signal Detect Enable
0 = Disabled
1 = Enabled
AIF2 DRC Signal Detect RMS
Threshold.
This is the RMS signal level for signal detect to be indicated when
AIF2DRC_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
11110 = -75dB
11111 = -76.5dB
AIF2 DRC Signal Detect Peak
Threshold.
This is the Peak/RMS ratio, or Crest
Factor, level for signal detect to be indicated when
AIF2DRC_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF2 DRC Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF2 DRC Signal Detect Enable
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
162
Pre-Production
WM8958
CONTROL WRITE SEQUENCER STATUS DETECTION
The WM8958 Control Write Sequencer (WSEQ) can be used to execute a sequence of register write operations in response to a simple trigger event. When the Control Write Sequencer is executing a sequence, normal access to the register map via the Control Interface is restricted. See “Control Write
Sequencer” for details of the Control Write Sequencer.
The WM8958 generates a signal indicating the status of the Control Write Sequencer, in order to signal to the host processor whether the Control Interface functionality is restricted due to an ongoing
Control Sequence. The WSEQ_DONE flag indicates that the sequencer has completed the commanded sequence.
The Write Sequencer status may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
The Write Sequencer status is an input to the Interrupt control circuit. An interrupt event is triggered on completion of a Control Sequence. The associated interrupt bit is latched once set; it can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
DIGITAL CORE FIFO ERROR STATUS DETECTION
The WM8958 monitors the Digital Core for error conditions which may occur if a clock rate mismatch is detected. Under these conditions, the digital audio may become corrupted.
The most likely cause of a Digital Core FIFO Error condition is an incorrect system clocking configuration. See “Clocking and Sample Rates” for the WM8958 system clocking requirements.
The Digital Core FIFO Error function is provided in order that the system configuration can be verified during product development.
The FIFO Error signal may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
The FIFO Error signal is an input to the Interrupt control circuit. An interrupt event is triggered on the rising edge of the FIFO Error signal. The associated interrupt bit is latched once set; it can be polled at any time or used to control the IRQ signal. See “Interrupts” for more details of the Interrupt event handling.
OPCLK CLOCK OUTPUT
A clock output (OPCLK) derived from SYSCLK may be output on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”. This clock is enabled by register bit
OPCLK_ENA, and its frequency is controlled by OPCLK_DIV. w
See “Clocking and Sample Rates” for more details of the System Clock (SYSCLK).
REGISTER
ADDRESS
R2 (0002h)
Power
Management
(2)
R521 (0209h)
Clocking 1
11 OPCLK_EN
2:0
A
OPCLK_DIV
0
000
GPIO Clock Output (OPCLK) Enable
0 = Disabled
1 = Enabled
DESCRIPTION
GPIO Output Clock (OPCLK) Divider
000 = SYSCLK
001 = SYSCLK / 2
010 = SYSCLK / 3
011 = SYSCLK / 4
100 = SYSCLK / 6
101 = SYSCLK / 8
110 = SYSCLK / 12
111 = SYSCLK / 16
Table 90 OPCLK Control
PP, August 2012, Rev 3.4
163
WM8958
Pre-Production
FLL CLOCK OUTPUT
The FLL Clock outputs may be output directly on any GPIO pin by setting the respective GPIO registers as described in “GPIO Control”.
See “Clocking and Sample Rates” for more details of the WM8958 system clocking and for details of how to enable and configure the Frequency Locked Loops. w PP, August 2012, Rev 3.4
164
Pre-Production
INTERRUPTS
WM8958
The Interrupt Controller has multiple inputs. These include the GPIO input pins, the FLL Lock circuits,
SRC Lock circuit, Microphone activity detection, Over-temperature indication, Digital FIFO error detection and the Write Sequencer status flag. Any combination of these inputs can be used to trigger an Interrupt Request (IRQ) event.
There is an Interrupt register field associated with each of the interrupt inputs. These fields are asserted whenever a logic edge is detected on the respective input. Some inputs are triggered on rising edges only; some are triggered on both edges, as noted in Table 91. The Interrupt register fields are held in Registers R1840 and R1841. The Interrupt flags can be polled at any time from these registers, or else in response to the Interrupt Request (IRQ) output being signalled via a GPIO pin.
All of the Interrupts are edge-triggered, as noted above. Many of these are triggered on both the rising and falling edges and, therefore, the Interrupt registers cannot indicate which edge has been detected. The “Raw Status” fields in Register R1842 provide readback of the current value of selected inputs to the Interrupt Controller. Note that the logic levels of any GPIO inputs can be read using the
GPn_LVL registers, as described in Table 86.
Individual mask bits can select or deselect different functions from the Interrupt controller. These are listed within the Interrupt Status Mask registers, as described in Table 91. Note that the Interrupt register fields remain valid, even when masked, but the masked interrupts will not cause the Interrupt
Request (IRQ) output to be asserted.
The Interrupt Request (IRQ) output represents the logical ‘OR’ of all the unmasked interrupts. The
Interrupt register fields are latching fields and, once they are set, they are not reset until a ‘1’ is written to the respective register bit(s). The Interrupt Request (IRQ) output is not reset until each of the unmasked interrupts has been reset.
De-bouncing of the GPIO inputs can be enabled using the register bits described in Table 86. Debouncing is also available on the Temperature Warning and Temperature Shutdown inputs to the
Interrupt Controller, in order to avoid false detections - see Table 91 for the associated registers.
The Interrupt Request (IRQ) output can be globally masked by setting the IM_IRQ register. Under default conditions, the Interrupt Request (IRQ) is not masked.
The Interrupt Request (IRQ) flag may be output on a GPIO pin - see “General Purpose Input/Output”.
The WM8958 Interrupt Controller circuit is illustrated in Figure 44. (Note that not all interrupt inputs are shown.) The associated control fields are described in Table 91.
Figure 44 Interrupt Controller w PP, August 2012, Rev 3.4
165
WM8958 w
REGISTER
ADDRESS
R1840
(0730h)
Interrupt
Status 1
R1841
(0731h)
Interrupt
Status 2
9
8
7
5
0
9
8
7
6
5
Pre-Production
DESCRIPTION
10 GP11_EINT
15
14
13
12
11
GP10_EINT
GP9_EINT
GP8_EINT
GP6_EINT
GP1_EINT
TEMP_WAR
N_EINT
DCS_DONE
_EINT
WSEQ_DO
NE_EINT
FIFOS_ERR
_EINT
AIF2DRC_SI
G_DET_EIN
T
10 AIF1DRC2_
SIG_DET_EI
NT
AIF1DRC1_
SIG_DET_EI
NT
SRC2_LOC
K_EINT
SRC1_LOC
K_EINT
FLL2_LOCK
_EINT
FLL1_LOCK
_EINT
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
GPIO11 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
GPIO10 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
GPIO9 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
GPIO8 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
GPIO6 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
GPIO1 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
Temperature Warning Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
DC Servo Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
Write Sequencer Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
Digital Core FIFO Error Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF2 DRC Activity Detect Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF1 DRC2 (Timeslot 1) Activity Detect
Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF1 DRC1 (Timeslot 0) Activity Detect
Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
SRC2 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
SRC1 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
FLL2 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
FLL1 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
PP, August 2012, Rev 3.4
166
Pre-Production w
REGISTER
ADDRESS
1 MICD_EINT
0 TEMP_SHU
T_EINT
R1842
(0732h)
Interrupt Raw
Status 2
15 TEMP_WAR
N_STS
14 DCS_DONE
_STS
13 WSEQ_DO
NE_STS
12 FIFOS_ERR
_STS
11 AIF2DRC_SI
G_DET_ST
S
10 AIF1DRC2_
SIG_DET_S
TS
9 AIF1DRC1_
SIG_DET_S
TS
8 SRC2_LOC
K_STS
7 SRC1_LOC
K_STS
6 FLL2_LOCK
_STS
5 FLL1_LOCK
_STS
R1848
(0738h)
Interrupt
Status 1
Mask
0 TEMP_SHU
T_STS
10 IM_GP11_EI
NT
9 IM_GP10_EI
NT
8 IM_GP9_EI
NT
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
0
WM8958
DESCRIPTION
Microphone Detection Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
Temperature Shutdown Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
Temperature Warning status
0 = Temperature is below warning level
1 = Temperature is above warning level
DC Servo status
0 = DC Servo not complete
1 = DC Servo complete
Write Sequencer status
0 = Sequencer Busy (sequence in progress)
1 = Sequencer Idle
Digital Core FIFO Error status
0 = Normal operation
1 = FIFO Error
AIF2 DRC Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
AIF1 DRC2 (Timeslot 1) Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
AIF1 DRC1 (Timeslot 0) Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
SRC2 Lock status
0 = Not locked
1 = Locked
SRC1 Lock status
0 = Not locked
1 = Locked
FLL2 Lock status
0 = Not locked
1 = Locked
FLL1 Lock status
0 = Not locked
1 = Locked
Temperature Shutdown status
0 = Temperature is below shutdown level
1 = Temperature is above shutdown level
GPIO11 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
GPIO10 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
GPIO9 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
PP, August 2012, Rev 3.4
167
WM8958 w
REGISTER
ADDRESS
R1849
(0739h)
Interrupt
Status 2
Mask
R1856
(0740h)
Interrupt
Control
7
5
0
9
8
7
6
5
1
0
0
Pre-Production
DESCRIPTION
15
14
13
12
11
IM_GP8_EI
NT
IM_GP6_EI
NT
IM_GP1_EI
NT
IM_TEMP_
WARN_EIN
T
IM_DCS_D
ONE_EINT
IM_WSEQ_
DONE_EINT
IM_FIFOS_
ERR_EINT
IM_AIF2DR
C_SIG_DET
_EINT
10 IM_AIF1DR
C2_SIG_DE
T_EINT
IM_AIF1DR
C1_SIG_DE
T_EINT
IM_SRC2_L
OCK_EINT
IM_SRC1_L
OCK_EINT
IM_FLL2_L
OCK_EINT
IM_FLL1_L
OCK_EINT
IM_MICD_EI
NT
IM_TEMP_S
HUT_EINT
IM_IRQ
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
GPIO8 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
GPIO6 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
GPIO1 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Temperature Warning Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
DC Servo Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Write Sequencer Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Digital Core FIFO Error Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF2 DRC Activity Detect Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF1 DRC2 (Timeslot 1) Activity Detect
Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF1 DRC1 (Timeslot 0) Activity Detect
Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
SRC2 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
SRC1 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
FLL2 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
FLL1 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Microphone Detection Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Temperature Shutdown Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
IRQ Output Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
PP, August 2012, Rev 3.4
168
Pre-Production
REGISTER
ADDRESS
R1864
(0748h)
IRQ
Debounce
5 TEMP_WAR
N_DB
0 TEMP_SHU
T_DB
Table 91 Interrupt Configuration
1
1
WM8958
DESCRIPTION
Temperature Warning de-bounce
0 = Disabled
1 = Enabled
Temperature Shutdown de-bounce
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
169
WM8958
Pre-Production
DIGITAL AUDIO INTERFACE
The WM8958 provides digital audio interfaces for inputting DAC data and outputting ADC or Digital
Microphone data. Flexible routing options also allow digital audio to be switched or mixed between interfaces without involving any DAC or ADC.
The WM8958 provides two full audio interfaces, AIF1 and AIF2. A third interface, AIF3, supports
Mono PCM digital audio paths to/from the AIF2 DSP functions. AIF3 can also be configured using multiplexers to provide alternate connections to AIF1 or AIF2.
The digital audio interfaces provide flexible connectivity with multiple processors (eg. Applications processor, Baseband processor and Wireless transceiver). A typical configuration is illustrated in
Figure 45.
Applications
Processor
Audio Interface 1
Baseband
Processor
Audio Interface 2
Wireless
Transceiver
Audio Interface 3
WM8958
Figure 45 Typical AIF Connections
In the general case, the digital audio interface uses four pins:
ADCDAT: ADC data output
DACDAT: DAC data input
LRCLK: Left/Right data alignment clock
BCLK: Bit clock, for synchronisation
In master interface mode, the clock signals BCLK and LRCLK are outputs from the WM8958. In slave mode, these signals are inputs, as illustrated below.
As an option, a GPIO pin can be configured as the Left/Right clock for the ADC. In this case, the
LRCLK pin is dedicated to the DAC, allowing the ADC and DAC to be clocked independently.
Four different audio data formats are supported each digital audio interface:
I 2 S w
All four of these modes are MSB first. They are described in the following sections. Refer to the
“Signal Timing Requirements” section for timing information.
PP, August 2012, Rev 3.4
170
Pre-Production
WM8958
Time Division Multiplexing (TDM) is available in all four data format modes. On AIF1, the WM8958 can transmit and receive data on two stereo pairs of timeslots simultaneously. On AIF2, the applicable timeslot pair is selectable using register control bits.
Two variants of DSP mode are supported - ‘Mode A’ and ‘Mode B’. Mono operation can be selected on either audio interface in both DSP modes. PCM operation is supported using the DSP mode.
MASTER AND SLAVE MODE OPERATION
The WM8958 digital audio interfaces can operate as a master or slave as shown in Figure 46 and
Figure 47. The associated control bits are described in “Digital Audio Interface Control”.
Figure 46 Master Mode Figure 47 Slave Mode
OPERATION WITH TDM
Time division multiplexing (TDM) allows multiple devices to transfer data simultaneously on the same bus. The WM8958 ADCs and DACs support TDM in master and slave modes for all data formats and word lengths. TDM is enabled and configured using register bits defined in the “Digital Audio Interface
Control” section.
WM8958
BCLK
LRCLK
ADCDAT
DACDAT
Processor WM8958
BCLK
LRCLK
ADCDAT
DACDAT
Processor
WM8958 or similar
CODEC
BCLK
LRCLK
ADCDAT
DACDAT
Figure 48 TDM with WM8958 as Master
WM8958 or similar
CODEC
BCLK
LRCLK
ADCDAT
DACDAT
Figure 49 TDM with Other CODEC as Master w PP, August 2012, Rev 3.4
171
WM8958
WM8958
BCLK
LRCLK
ADCDAT
DACDAT
Pre-Production
Processor
WM8958 or similar
CODEC
BCLK
LRCLK
ADCDAT
DACDAT
Figure 50 TDM with Processor as Master
Note: The WM8958 is a 24-bit device. If the user operates the WM8958 in 32-bit mode then the 8
LSBs will be ignored on the receiving side and not driven on the transmitting side. It is therefore recommended to add a pull-down resistor if necessary to the DACDAT line and the ADCDAT line in
TDM mode.
AUDIO DATA FORMATS (NORMAL MODE)
The audio data modes supported by the WM8958 are described below. Note that the polarity of the
BCLK and LRCLK signals can be inverted if required; the following descriptions all assume the default, non-inverted polarity of these signals.
In Right Justified mode, the LSB is available on the last rising edge of BCLK before a LRCLK transition. All other bits are transmitted before (MSB first). Depending on word length, BCLK frequency and sample rate, there may be unused BCLK cycles after each LRCLK transition. w
Figure 51 Right Justified Audio Interface (assuming n-bit word length)
In Left Justified mode, the MSB is available on the first rising edge of BCLK following a LRCLK transition. The other bits up to the LSB are then transmitted in order. Depending on word length,
BCLK frequency and sample rate, there may be unused BCLK cycles before each LRCLK transition.
PP, August 2012, Rev 3.4
172
Pre-Production
WM8958
Figure 52 Left Justified Audio Interface (assuming n-bit word length)
In I 2 S mode, the MSB is available on the second rising edge of BCLK following a LRCLK transition.
The other bits up to the LSB are then transmitted in order. Depending on word length, BCLK frequency and sample rate, there may be unused BCLK cycles between the LSB of one sample and the MSB of the next.
1/fs
LEFT CHANNEL RIGHT CHANNEL
LRCLK
BCLK
DACDAT/
ADCDAT
1 BCLK
1
MSB
2 3 n-2 n-1 n
LSB
Input Word Length (WL)
1 BCLK
1 2 3 n-2 n-1 n
Figure 53 I2S Justified Audio Interface (assuming n-bit word length)
In DSP mode, the left channel MSB is available on either the 1 st (mode B) or 2 nd (mode A) rising edge of BCLK following a rising edge of LRCLK. Right channel data immediately follows left channel data.
Depending on word length, BCLK frequency and sample rate, there may be unused BCLK cycles between the LSB of the right channel data and the next sample.
The selected mode (Mode A or Mode B) is determined by the AIFnDAC_LRCLK_INV bits for the AIFn digital input (playback) signal paths, and by the AIFnADC_LRCLK_INV bits for the AIFn digital output
(record) signal paths.
Note that the DSP Mode is selected independently for the input/output paths of each digital audio interface.
In device master mode, the LRCLK output will resemble the frame pulse shown in Figure 54 and
Figure 55. In device slave mode, Figure 56 and Figure 57, it is possible to use any length of frame pulse less than 1/fs, providing the falling edge of the frame pulse occurs greater than one BCLK period before the rising edge of the next frame pulse. w PP, August 2012, Rev 3.4
173
WM8958
Pre-Production
Figure 54 DSP Mode A (AIFnDAC_LRCLK_INV / AIFnADC_LRCLK_INV=0, Master)
1/fs
LRCLK
BCLK
DACDAT/
ADCDAT
LEFT CHANNEL
1
MSB
2 3 n-2 n-1 n
LSB
Input Word Length (WL)
1 2 3
RIGHT CHANNEL n-2 n-1 n
Figure 55 DSP B Mode (AIFnDAC_LRCLK_INV / AIFnADC_LRCLK_INV=1, Master)
Figure 56 DSP Mode A (AIFnDAC_LRCLK_INV / AIFnADC_LRCLK_INV =0, Slave) w PP, August 2012, Rev 3.4
174
Pre-Production
WM8958 w
Figure 57 DSP Mode B (AIFnDAC_LRCLK_INV / AIFnADC_LRCLK_INV =1, Slave)
Mono mode operation is available in DSP interface mode. When Mono mode is enabled, the audio data is transmitted or received starting on either the 1 st (mode B) or 2 nd (mode A) rising edge of BCLK following a rising edge of LRCLK.
PCM operation is supported in DSP interface mode. WM8958 ADC data that is output on the Left
Channel will be read as mono PCM data by the receiving equipment. Mono PCM data received by the
WM8958 will be treated as Left Channel data. This data may be routed to the Left/Right DACs using the control fields described in the “Digital Mixing” and “Digital Audio Interface Control” sections.
AUDIO DATA FORMATS (TDM MODE)
TDM is supported in master and slave modes. All audio interface data formats support time division multiplexing (TDM) for ADC and DAC data.
When more than one pair of ADC or DAC data channels is enabled on AIF1, the WM8958 will transmit and receive data in both Slot 0 and Slot 1.
In the case of AIF2, the ADC or DAC data can be transmitted or received in either timeslot; the required timeslot is selected using register control bits when TDM is enabled.
When TDM is enabled, the ADCDAT pin will be tri-stated immediately before and immediately after data transmission, to allow another ADC device to drive this signal line for the remainder of the sample period. Note that it is important that two ADC devices do not attempt to drive the data pin simultaneously. A short circuit may occur if the transmission time of the two ADC devices overlap with each other. See “Audio Interface Timing” for details of the ADCDAT output relative to BCLK signal.
Note that it is possible to ensure a gap exists between transmissions by setting the transmitted word length to a value higher than the actual length of the data. For example, if 32-bit word length is selected where only 24-bit data is available, then the WM8958 interface will tri-state after transmission of the 24-bit data, ensuring a gap after the WM8958 TDM slot.
On AIF1, TDM can be used to transmit or receive up to four signal paths. Each enabled signal path is transmitted (on ADCDAT) or received (on DACDAT) sequentially. If one or more of the signal paths is disabled, then the position of remaining data blocks within the LRCLK frame may differ from those illustrated in Figure 58 to Figure 62, as the affected channel(s) will revert to the ‘normal’ (non-TDM) format. When the AIF1ADC_TDM register is set, then the ADCDAT1 output is tri-stated when not outputting data.
On AIF2, the TDM format is enabled by register control (AIF2ADC_TDM and AIF2DAC_TDM for the output and input paths respectively). When TDM is enabled on AIF2, the data formats shown in
Figure 58 to Figure 62 are always selected, and the WM8958 transmits or receives data in one of the two available timeslots; the ADCDAT2 output is tri-stated when not outputting data.
In all cases, the BCLK frequency must be high enough to allow data from the relevant time slots to be transferred. The relative timing of Slot 0 and Slot 1 depends upon the selected data format; the TDM timing for four input or output channels is shown in Figure 58 to Figure 62.
PP, August 2012, Rev 3.4
175
WM8958
Pre-Production
Figure 58 TDM in Right-Justified Mode
Figure 59 TDM in Left-Justified Mode
Figure 60 TDM in I 2 S Mode w
Figure 61 TDM in DSP Mode A
PP, August 2012, Rev 3.4
176
Pre-Production
Figure 62 TDM in DSP Mode B
WM8958 w PP, August 2012, Rev 3.4
177
WM8958
Pre-Production
DIGITAL AUDIO INTERFACE CONTROL
This section describes the configuration of the WM8958 digital audio interface paths.
Interfaces AIF1 and AIF2 can be configured as Master or Slave, or can be tri-stated. Each input and output signal path can be independently enabled or disabled. AIF output (digital record) and AIF input
(digital playback) paths can use a common Left/Right clock, or can use separate clocks for mixed sample rates.
Interfaces AIF1 and AIF2 each support flexible formats, word-length, TDM configuration, channel swapping and input path digital boost functions. 8-bit companding modes and digital loopback is also possible.
A third interface, AIF3, supports Mono PCM digital audio paths to/from the AIF2 DSP functions. AIF3 can also be configured using multiplexers to provide alternate connections to AIF1 or AIF2. Note that
AIF3 operates in Master mode only.
AIF1 - MASTER / SLAVE AND TRI-STATE CONTROL
The Digital Audio Interface AIF1 can operate in Master or Slave modes, selected by AIF1_MSTR. In
Master mode, the BCLK1 and LRCLK1 signals are generated by the WM8958 when one or more
AIF1 channels is enabled.
When AIF1_LRCLK_FRC or AIF1_CLK_FRC is set in Master mode, then LRCLK1 and ADCLRCLK1 are output at all times, including when none of the AIF1 audio channels is enabled. Note that LRCLK1 and ADCLRCLK1 are derived from BCLK1, and either an internal or external BCLK1 signal must also be present to generate LRCLK1 or ADCLRCLK1.
When AIF1_CLK_FRC is set in Master mode, then BCLK1 is output at all times, including when none of the AIF1 audio channels is enabled.
The AIF1 interface can be tri-stated by setting the AIF1_TRI register. When this bit is set, then all of the AIF1 outputs are un-driven (high-impedance). Note that the GPIO1/ADCLRCLK1 pin is a configurable pin which may take different functions independent of AIF1. The AIF1_TRI register only controls the GPIO1/ADCLRCLK1 pin when its function is set to ADCLRCLK1. See “General Purpose
Input/Output” to configure the GPIO1 pin.
DESCRIPTION REGISTER
ADDRESS
R770 (0302h)
AIF1
Master/Slave
15 AIF1_TRI 0
14
13
12
AIF1_MSTR
AIF1_CLK_F
RC
AIF1_LRCL
K_FRC
0
0
0
AIF1 Audio Interface tri-state
0 = AIF1 pins operate normally
1 = Tri-state all AIF1 interface pins
Note that the GPIO1 pin is controlled by this register only when configured as
ADCLRCLK1.
AIF1 Audio Interface Master Mode Select
0 = Slave mode
1 = Master mode
Forces BCLK1, LRCLK1 and ADCLRCLK1 to be enabled when all AIF1 audio channels are disabled.
0 = Normal
1 = BCLK1, LRCLK1 and ADCLRCLK1 always enabled in Master mode
Forces LRCLK1 and ADCLRCLK1 to be enabled when all AIF1 audio channels are disabled.
0 = Normal
1 = LRCLK1 and ADCLRCLK1 always enabled in Master mode
Table 92 AIF1 Master / Slave and Tri-state Control w PP, August 2012, Rev 3.4
178
Pre-Production w
WM8958
AIF1 - SIGNAL PATH ENABLE
The AIF1 interface supports up to four input channels and up to four output channels. All enabled channels are transmitted (on ADCDAT) or received (on DACDAT) sequentially, using time division multiplexing (TDM).
Each of the available channels can be enabled or disabled using the register bits defined in Table 93.
These register controls are illustrated in Figure 67.
DESCRIPTION REGISTER
ADDRESS
R4 (0004h)
Power
Management
(4)
R5 (0005h)
Power
Management
(5)
11
10
9
8
11
10
9
8
AIF1ADC2L
_ENA
AIF1ADC2R
_ENA
AIF1ADC1L
_ENA
AIF1ADC1R
_ENA
AIF1DAC2L
_ENA
AIF1DAC2R
_ENA
AIF1DAC1L
_ENA
AIF1DAC1R
_ENA
0
0
0
0
0
0
0
0
Enable AIF1ADC2 (Left) output path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1ADC2 (Right) output path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1ADC1 (Left) output path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1ADC1 (Right) output path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Left) input path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Right) input path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1DAC1 (Left) input path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1DAC1 (Right) input path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Table 93 AIF1 Signal Path Enable
AIF1 - BCLK AND LRCLK CONTROL
The BCLK1 frequency is controlled relative to AIF1CLK by the AIF1_BCLK_DIV divider. See
“Clocking and Sample Rates” for details of the AIF1 clock, AIF1CLK.
The LRCLK1 frequency is controlled relative to BCLK1 by the AIF1DAC_RATE divider.
In Master mode, the LRCLK1 output is generated by the WM8958 when any of the AIF1 channels is enabled. (Note that, when GPIO1 is configured as ADCLRCLK1, then only the AIF1 DAC channels will cause LRCLK1 to be output.)
In Slave mode, the LRCLK1 output is disabled by default to allow another digital audio interface to drive this pin. It is also possible to force the LRCLK1 signal to be output, using the
AIF1DAC_LRCLK_DIR or AIF1ADC_LRCLK_DIR register bits, allowing mixed master and slave modes. (Note that, when GPIO1 is configured as ADCLRCLK1, then only the AIF1DAC_LRCLK_DIR bit will force the LRCLK1 signal.)
PP, August 2012, Rev 3.4
179
WM8958
Pre-Production
When the GPIO1 pin is configured as ADCLRCLK1, then the ADCLRCLK1 frequency is controlled relative to BCLK1 by the AIF1ADC_RATE divider. In this case, the ADCLRCLK1 is dedicated to AIF1 output, and the LRCLK1 pin is dedicated to AIF1 input, allowing different sample rates to be supported in the two paths.
In Master mode, with GPIO1 pin configured as ADCLRCLK1, this output is enabled when any of the
AIF1 ADC channels is enabled. The ADCLRCLK1 signal can also be enabled in Slave mode, using the AIF1ADC_LRCLK_DIR bit, allowing mixed master and slave modes.
When the GPIO1 pin is not configured as ADCLRCLK1, then the LRCLK1 signal applies to the ADC and DAC channels, at a rate set by AIF1DAC_RATE.
See “General Purpose Input/Output” for the configuration of GPIO1. Note that, in Ultrasonic (4FS) mode, the GPIO1 pin must be configured as ADCLRCLK1.
The BCLK1 output can be inverted using the AIF1_BCLK_INV register bit. The LRCLK1 and
ADCLRCLK1 output (when selected) can be inverted using the AIF1DAC_LRCLK_INV and
AIF1ADC_LRCLK_INV register controls respectively.
Note that in Slave mode, when BCLK1 is an input, the AIF1_BCLK_INV register selects the polarity of the received BCLK1 signal. Under default conditions, DACDAT1 input is captured on the rising edge of BCLK1, as illustrated in Figure 5. When AIF1_BCLK_INV = 1, DACDAT1 input is captured on the falling edge of BCLK1.
The AIF1 clock generators are controlled as illustrated in Figure 63. w
Figure 63 Audio Interface 1 - BCLK and LRCLK Control
PP, August 2012, Rev 3.4
180
Pre-Production w
REGISTER
ADDRESS
R768 (0300h)
AIF1 Control
(1)
8 AIF1_BCLK
_INV
R771 (0303h)
AIF1 BCLK
8:4 AIF1_BCLK
_DIV [4:0]
R772 (0304h)
AIF1ADC
LRCLK
12 AIF1ADC_L
RCLK_INV
11 AIF1ADC_L
RCLK_DIR
10:0 AIF1ADC_R
ATE [10:0]
0
00100
0
0
040h
WM8958
DESCRIPTION
BCLK1 Invert
0 = BCLK1 not inverted
1 = BCLK1 inverted
Note that AIF1_BCLK_INV selects the BCLK1 polarity in Master mode and in Slave mode.
BCLK1 Rate
00000 = AIF1CLK
00001 = AIF1CLK / 1.5
00010 = AIF1CLK / 2
00011 = AIF1CLK / 3
00100 = AIF1CLK / 4
00101 = AIF1CLK / 5
00110 = AIF1CLK / 6
00111 = AIF1CLK / 8
01000 = AIF1CLK / 11
01001 = AIF1CLK / 12
01010 = AIF1CLK / 16
01011 = AIF1CLK / 22
01100 = AIF1CLK / 24
01101 = AIF1CLK / 32
01110 = AIF1CLK / 44
01111 = AIF1CLK / 48
10000 = AIF1CLK / 64
10001 = AIF1CLK / 88
10010 = AIF1CLK / 96
10011 = AIF1CLK / 128
10100 = AIF1CLK / 176
10101 = AIF1CLK / 192
10110 - 11111 = Reserved
Right, left and I 2 S modes – ADCLRCLK1 polarity
0 = normal ADCLRCLK1 polarity
1 = invert ADCLRCLK1 polarity
Note that AIF1ADC_LRCLK_INV selects the
ADCLRCLK1 polarity in Master mode and in
Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK1 rising edge after ADCLRCLK1 rising edge (mode A)
1 = MSB is available on 1st BCLK1 rising edge after ADCLRCLK1 rising edge (mode B)
Allows ADCLRCLK1 to be enabled in Slave mode
0 = Normal
1 = ADCLRCLK1 enabled in Slave mode
ADCLRCLK1 Rate
ADCLRCLK1 clock output =
BCLK1 / AIF1ADC_RATE
Integer (LSB = 1)
Valid from 8..2047
PP, August 2012, Rev 3.4
181
WM8958 w
REGISTER
ADDRESS
R773 (0305h)
AIF1DAC
LRCLK
Pre-Production
DESCRIPTION
12 AIF1DAC_L
RCLK_INV
11 AIF1DAC_L
RCLK_DIR
10:0 AIF1DAC_R
ATE [10:0]
0
0
040h
Right, left and I 2 S modes – LRCLK1 polarity
0 = normal LRCLK1 polarity
1 = invert LRCLK1 polarity
Note that AIF1DAC_LRCLK_INV selects the
LRCLK1 polarity in Master mode and in Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK1 rising edge after LRCLK1 rising edge (mode A)
1 = MSB is available on 1st BCLK1 rising edge after LRCLK1 rising edge (mode B)
Allows LRCLK1 to be enabled in Slave mode
0 = Normal
1 = LRCLK1 enabled in Slave mode
LRCLK1 Rate
LRCLK1 clock output =
BCLK1 / AIF1DAC_RATE
Integer (LSB = 1)
Valid from 8..2047
Table 94 AIF1 BCLK and LRCLK Control
AIF1 - DIGITAL AUDIO DATA CONTROL
The register bits controlling the audio data format, word length, left/right channel selection and TDM control for AIF1 are described in Table 95.
In DSP mode, the left channel MSB is available on either the 1 st (mode B) or 2 nd (mode A) rising edge of BCLK following a rising edge of LRCLK (assuming default BCLK polarity).
When the AIF1DAC_LRCLK_INV bit is set in DSP mode, then DSP Mode B is selected for the AIF1 digital input (playback) signal path. When the AIF1DAC_LRCLK_INV bit is not set, then DSP Mode A is selected.
When the AIF1ADC_LRCLK_INV bit is set in DSP mode, then DSP Mode B is selected for the AIF1 digital output (record) signal path. When the AIF1ADC_LRCLK_INV bit is not set, then DSP Mode A is selected.
Note that the DSP Mode is selected independently for the input/output paths of each digital audio interface. Also note that the AIF1ADCLRCLK_INV bits remain valid even when the LRCLK signal is common for both paths. See Table 94 for details of the AIF1DAC_LRCLK_INV and
AIF1ADC_LRCLK_INV register fields.
A digital gain function is available at the audio interface input path to boost the DAC volume when a small signal is received on DACDAT1. This is controlled using the AIF1DAC_BOOST register. To prevent clipping, this function should not be used when the boosted data is expected to be greater than 0dBFS.
DESCRIPTION REGISTER
ADDRESS
R768 (0300h)
AIF1 Control
(1)
15
14
AIF1ADCL_
SRC
AIF1ADCR_
SRC
0
1
AIF1 Left Digital Audio interface source
0 = Left ADC data is output on left channel
1 = Right ADC data is output on left channel
AIF1 Right Digital Audio interface source
0 = Left ADC data is output on right channel
1 = Right ADC data is output on right channel
PP, August 2012, Rev 3.4
182
Pre-Production w
WM8958
REGISTER
ADDRESS
DESCRIPTION
R769 (0301h)
AIF1 Control
(2)
R774 (0306h)
AIF1 DAC
Data
R775 (0307h)
AIF1 ADC
Data
13 AIF1ADC_T
DM
6:5 AIF1_WL
[1:0]
4:3 AIF1_FMT
[1:0]
15 AIF1DACL_
SRC
14 AIF1DACR_
SRC
11:10 AIF1DAC_B
OOST [1:0]
1
0
1
0
AIF1DACL_
DAT_INV
AIF1DACR_
DAT_INV
AIF1ADCL_
DAT_INV
AIF1ADCR_
DAT_INV
0
10
10
0
1
00
0
0
0
0
AIF1 transmit (ADC) TDM Control
0 = ADCDAT1 drives logic ‘0’ when not transmitting data
1 = ADCDAT1 is tri-stated when not transmitting data
AIF1 Digital Audio Interface Word Length
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
AIF1 Digital Audio Interface Format
00 = Right justified
01 = Left justified
10 = I 2 S Format
11 = DSP Mode
AIF1 Left Receive Data Source Select
0 = Left DAC receives left interface data
1 = Left DAC receives right interface data
AIF1 Right Receive Data Source Select
0 = Right DAC receives left interface data
1 = Right DAC receives right interface data
AIF1 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
AIF1 Left Receive Data Invert
0 = Not inverted
1 = Inverted
AIF1 Right Receive Data Invert
0 = Not inverted
1 = Inverted
AIF1 Left Transmit Data Invert
0 = Not inverted
1 = Inverted
AIF1 Right Transmit Data Invert
0 = Not inverted
1 = Inverted
Table 95 AIF1 Digital Audio Data Control
AIF1 - MONO MODE
AIF1 can be configured to operate in mono DSP mode by setting AIF1_MONO = 1 as described in
Table 96. Note that mono mode is only supported in DSP mode, ie when AIF1_FMT = 11.
In mono mode, the Left channel data or the Right channel data may be selected for output on
ADCDAT1. The selected channel is determined by the AIF1ADC1L_ENA and AIF1ADC1R_ENA bits.
(If both bits are set, then the Right channel data is selected.)
In mono mode, the DACDAT1 input can be enabled on the Left and/or Right signal paths using the
AIF1DAC1L_ENA and AIF1DAC1R_ENA bits. The mono input can be enabled on both paths at the same time if required.
Note that AIF1 TDM mode and AIF1 Mono mode cannot be supported simultaneously.
PP, August 2012, Rev 3.4
183
WM8958 w
REGISTER
ADDRESS
R769 (0301h)
AIF1 Control
(2)
8
Pre-Production
DESCRIPTION
AIF1_MONO 0 AIF1 DSP Mono Mode
0 = Disabled
1 = Enabled
Note that Mono Mode is only supported when
AIF1_FMT = 11.
Table 96 AIF1 Mono Mode Control
AIF1 - COMPANDING
The WM8958 supports A-law and -law companding on both transmit (ADC) and receive (DAC) sides of AIF1. This is configured using the register bits described in Table 97.
REGISTER
ADDRESS
R769 (0301h)
AIF1 Control
(2)
4 AIF1DAC_C
OMP
0
DESCRIPTION
3
2
1
AIF1DAC_C
OMPMODE
AIF1ADC_C
OMP
AIF1ADC_C
OMPMODE
0
0
0
AIF1 Receive Companding Enable
0 = Disabled
1 = Enabled
AIF1 Receive Companding Type
0 = µ-law
1 = A-law
AIF1 Transmit Companding Enable
0 = Disabled
1 = Enabled
AIF1 Transmit Companding Type
0 = µ-law
1 = A-law
Table 97 AIF1 Companding
Companding involves using a piecewise linear approximation of the following equations (as set out by
ITU-T G.711 standard) for data compression:
-law (where =255 for the U.S. and Japan):
F(x) = ln( 1 + |x|) / ln( 1 + )
A-law (where A=87.6 for Europe):
} for -1 ≤ x ≤ 1
F(x) = A|x| / ( 1 + lnA)
F(x) = ( 1 + lnA|x|) / (1 + lnA)
for x ≤ 1/A
for 1/A ≤ x ≤ 1
The companded data is also inverted as recommended by the G.711 standard (all 8 bits are inverted for -law, all even data bits are inverted for A-law). The data will be transmitted as the first 8 MSBs of data.
Companding converts 13 bits ( -law) or 12 bits (A-law) to 8 bits using non-linear quantization. This provides greater precision for low amplitude signals than for high amplitude signals, resulting in a greater usable dynamic range than 8 bit linear quantization. The companded signal is an 8-bit word comprising sign (1 bit), exponent (3 bits) and mantissa (4 bits).
AIF1 8-bit mode is selected whenever AIF1DAC_COMP=1 or AIF1ADC_COMP=1. The use of 8-bit data allows samples to be passed using as few as 8 BCLK1 cycles per LRCLK1 frame. When using
DSP mode B, 8-bit data words may be transferred consecutively every 8 BCLK1 cycles.
AIF1 8-bit mode (without Companding) may be enabled by setting AIF1DAC_COMPMODE=1 or
AIF1ADC_COMPMODE=1, when AIF1DAC_COMP=0 and AIF1ADC_COMP=0.
BIT7 BIT[6:4]
SIGN EXPONENT
Table 98 8-bit Companded Word Composition
BIT[3:0]
MANTISSA
PP, August 2012, Rev 3.4
184
Pre-Production
120
100
80
60
40
20
0
0 u-law Companding
120
100
80
60
40
20
0
0 0.1
0.2
0.3
0.4
0.5
0.6
Normalised Input
0.7
0.8
0.9
0.7
0.6
0.5
0.4
0.3
0.2
1
0.9
0.8
1
0.1
0
Figure 64 µ-Law Companding
WM8958
A-law Companding
0.2
0.4
0.6
Normalised Input
0.8
1
0.9
0.8
0.7
0.6
1
0
0.2
0.1
0.5
0.4
0.3
Figure 65 A-Law Companding w PP, August 2012, Rev 3.4
185
WM8958
REGISTER
ADDRESS
R769 (0301h)
AIF1 Control
(2)
0
Pre-Production
AIF1 - LOOPBACK
The AIF1 interface can provide a Loopback option. When the AIF1_LOOPBACK bit is set, then AIF1 digital audio output is routed to the AIF1 digital audio input. The normal input (DACDAT1) is not used when AIF1 Loopback is enabled.
AIF1_LOOP
BACK
0
DESCRIPTION
AIF1 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (ADCDAT1 data output is directly input to DACDAT1 data input).
Table 99 AIF1 Loopback
AIF1 - DIGITAL PULL-UP AND PULL-DOWN
The WM8958 provides integrated pull-up and pull-down resistors on each of the DACDAT1, LRCLK1 and BCLK1 pins. This provides a flexible capability for interfacing with other devices.
Each of the pull-up and pull-down resistors can be configured independently using the register bits described in Table 100. Note that if the Pull-up and Pull-down are both enabled for any pin, then the pull-up and pull-down will be disabled.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1824
(0720h)
Pull Control
(1)
5
4
3
2
1
0
DACDAT1_PU
DACDAT1_PD
DACLRCLK1_
PU
DACLRCLK1_
PD
BCLK1_PU
BCLK1_PD
0
0
0
0
0
0
DACDAT1 Pull-up enable
0 = Disabled
1 = Enabled
DACDAT1 Pull-down enable
0 = Disabled
1 = Enabled
LRCLK1 Pull-up enable
0 = Disabled
1 = Enabled
LRCLK1 Pull-down enable
0 = Disabled
1 = Enabled
BCLK1 Pull-up enable
0 = Disabled
1 = Enabled
BCLK1 Pull-down enable
0 = Disabled
1 = Enabled
Table 100 AIF1 Digital Pull-Up and Pull-Down Control w PP, August 2012, Rev 3.4
186
Pre-Production
WM8958
AIF2 - MASTER / SLAVE AND TRI-STATE CONTROL
The Digital Audio Interface AIF2 can operate in Master or Slave modes, selected by AIF2_MSTR. In
Master mode, the BCLK2 and LRCLK2 signals are generated by the WM8958 when one or more
AIF2 channels is enabled.
When AIF2_LRCLK_FRC or AIF2_CLK_FRC is set in Master mode, then LRCLK2 and ADCLRCLK2 are output at all times, including when none of the AIF2 audio channels is enabled. Note that LRCLK2 and ADCLRCLK2 are derived from BCLK2, and either an internal or external BCLK2 signal must also be present to generate LRCLK2 or ADCLRCLK2.
When AIF2_CLK_FRC is set in Master mode, then BCLK2 is output at all times, including when none of the AIF2 audio channels is enabled.
Note that the ADCLRCLK2 pin is also a GPIO pin, whose function is configurable. This pin must be configured for AIF functionality when used as audio interface pin. See “General Purpose
Input/Output”.
The AIF2 interface can be tri-stated by setting the AIF2_TRI register. When this bit is set, then all of the AIF2 outputs are un-driven (high-impedance). The AIF2_TRI register only affects those pins which are configured for AIF2 functions; it does not affect pins which are configured for other functions.
REGISTER
ADDRESS
R786 (0312h)
AIF2
Master/Slave
15
14
13
12
AIF2_TRI
AIF2_MSTR
AIF2_CLK_F
RC
AIF2_LRCL
K_FRC
0
0
0
0
Table 101 AIF2 Master / Slave and Tri-state Control
DESCRIPTION
AIF2 Audio Interface tri-state
0 = AIF2 pins operate normally
1 = Tri-state all AIF2 interface pins
Note that pins not configured as AIF2 functions are not affected by this register.
AIF2 Audio Interface Master Mode Select
0 = Slave mode
1 = Master mode
Forces BCLK2, LRCLK2 and ADCLRCLK2 to be enabled when all AIF2 audio channels are disabled.
0 = Normal
1 = BCLK2, LRCLK2 and ADCLRCLK2 always enabled in Master mode
Forces LRCLK2 and ADCLRCLK2 to be enabled when all AIF2 audio channels are disabled.
0 = Normal
1 = LRCLK2 and ADCLRCLK2 always enabled in Master mode w PP, August 2012, Rev 3.4
187
WM8958
AIF2 - SIGNAL PATH ENABLE
The AIF2 interface supports two input channels and two output channels. Each of the available channels can be enabled or disabled using the register bits defined in Table 102. These register controls are illustrated in Figure 67.
REGISTER
ADDRESS
R4 (0004h)
Power
Management
(4)
R5 (0005h)
Power
Management
(5)
R784 (0310h)
AIF2 Control
(1)
13 AIF2ADCL_
ENA
12 AIF2ADCR_
ENA
13 AIF2DACL_
ENA
12 AIF2DACR_
ENA
1
0
Pre-Production
AIF2TXL_E
NA
AIF2TXR_E
NA
0
0
0
0
1
1
DESCRIPTION
Enable AIF2ADC (Left) output path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 or AIF3 output of the AIF2ADC (Left) signal.
Enable AIF2ADC (Right) output path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 or AIF3 output of the AIF2ADC (Left) signal.
Enable AIF2DAC (Left) input path
0 = Disabled
1 = Enabled
Enable AIF2DAC (Right) input path
0 = Disabled
1 = Enabled
Enable AIF2DAC (Left) input path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 output of the
AIF2ADC (Left) signal. For AIF3 output only, this bit can be set to 0.
Enable AIF2DAC (Right) input path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 output of the
AIF2ADC (Left) signal. For AIF3 output only, this bit can be set to 0.
Table 102 AIF2 Signal Path Enable
AIF2 - BCLK AND LRCLK CONTROL
The BCLK2 frequency is controlled relative to AIF2CLK by the AIF2_BCLK_DIV divider. See
“Clocking and Sample Rates” for details of the AIF2 clock, AIF2CLK.
The LRCLK2 frequency is controlled relative to BCLK2 by the AIF2DAC_RATE divider.
In Master mode, the LRCLK2 output is generated by the WM8958 when any of the AIF2 channels is enabled. (Note that, when GPIO6 is configured as ADCLRCLK2, then only the AIF2 DAC channels will cause LRCLK2 to be output.)
In Slave mode, the LRCLK2 output is disabled by default to allow another digital audio interface to drive this pin. It is also possible to force the LRCLK2 signal to be output, using the
AIF2DAC_LRCLK_DIR or AIF2ADC_LRCLK_DIR register bits, allowing mixed master and slave modes. (Note that, when GPIO6 is configured as ADCLRCLK2, then only the AIF2DAC_LRCLK_DIR bit will force the LRCLK2 signal.)
When the GPIO6 pin is configured as ADCLRCLK2, then the ADCLRCLK2 frequency is controlled relative to BCLK2 by the AIF2ADC_RATE divider. In this case, the ADCLRCLK2 is dedicated to AIF2 output, and the LRCLK2 pin is dedicated to AIF2 input, allowing different sample rates to be supported in the two paths. w
In Master mode, with GPIO6 pin configured as ADCLRCLK2, this output is enabled when any of the
AIF2 ADC channels is enabled. The ADCLRCLK2 signal can also be enabled in Slave mode, using the AIF2ADC_LRCLK_DIR bit, allowing mixed master and slave modes.
PP, August 2012, Rev 3.4
188
Pre-Production
WM8958
See “General Purpose Input/Output” for the configuration of GPIO6.
The BCLK2 output can be inverted using the AIF2_BCLK_INV register bit. The LRCLK2 and
ADCLRCLK2 output (when selected) can be inverted using the AIF2DAC_LRCLK_INV and
AIF2ADC_LRCLK_INV register controls respectively.
Note that in Slave mode, when BCLK2 is an input, the AIF2_BCLK_INV register selects the polarity of the received BCLK2 signal. Under default conditions, DACDAT2 input is captured on the rising edge of BCLK2, as illustrated in Figure 5. When AIF2_BCLK_INV = 1, DACDAT2 input is captured on the falling edge of BCLK2.
The AIF2 clock generators are controlled as illustrated in Figure 66. w
Figure 66 Audio Interface 2 - BCLK and LRCLK Control
PP, August 2012, Rev 3.4
189
WM8958 w
REGISTER
ADDRESS
R784 (0310h)
AIF2 Control
(1)
R787 (0313h)
AIF2 BCLK
R788 (0314h)
AIF2ADC
LRCLK
R789 (0315h)
AIF2DAC
LRCLK
8
Pre-Production
DESCRIPTION
8:4
12
11
10:0
12
AIF2_BCLK
_INV
AIF2_BCLK
_DIV [4:0]
AIF2ADC_L
RCLK_INV
AIF2ADC_L
RCLK_DIR
AIF2ADC_R
ATE [10:0]
AIF2DAC_L
RCLK_INV
0
00100
0
0
040h
0
BCLK2 Invert
0 = BCLK2 not inverted
1 = BCLK2 inverted
Note that AIF2_BCLK_INV selects the BCLK2 polarity in Master mode and in Slave mode.
BCLK2 Rate
00000 = AIF2CLK
00001 = AIF2CLK / 1.5
00010 = AIF2CLK / 2
00011 = AIF2CLK / 3
00100 = AIF2CLK / 4
00101 = AIF2CLK / 5
00110 = AIF2CLK / 6
00111 = AIF2CLK / 8
01000 = AIF2CLK / 11
01001 = AIF2CLK / 12
01010 = AIF2CLK / 16
01011 = AIF2CLK / 22
01100 = AIF2CLK / 24
01101 = AIF2CLK / 32
01110 = AIF2CLK / 44
01111 = AIF2CLK / 48
10000 = AIF2CLK / 64
10001 = AIF2CLK / 88
10010 = AIF2CLK / 96
10011 = AIF2CLK / 128
10100 = AIF2CLK / 176
10101 = AIF2CLK / 192
10110 - 11111 = Reserved
Right, left and I 2 S modes – ADCLRCLK2 polarity
0 = normal ADCLRCLK2 polarity
1 = invert ADCLRCLK2 polarity
Note that AIF2ADC_LRCLK_INV selects the
ADCLRCLK2 polarity in Master mode and in
Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK2 rising edge after ADCLRCLK2 rising edge (mode A)
1 = MSB is available on 1st BCLK2 rising edge after ADCLRCLK2 rising edge (mode B)
Allows ADCLRCLK2 to be enabled in Slave mode
0 = Normal
1 = ADCLRCLK2 enabled in Slave mode
ADCLRCLK2 Rate
ADCLRCLK2 clock output =
BCLK2 / AIF2ADC_RATE
Integer (LSB = 1)
Valid from 8..2047
Right, left and I 2 S modes – LRCLK2 polarity
0 = normal LRCLK2 polarity
1 = invert LRCLK2 polarity
Note that AIF2DAC_LRCLK_INV selects the
LRCLK2 polarity in Master mode and in Slave mode.
PP, August 2012, Rev 3.4
190
Pre-Production w
WM8958
REGISTER
ADDRESS
DESCRIPTION
11 AIF2DAC_L
RCLK_DIR
10:0 AIF2DAC_R
ATE [10:0]
0
040h
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK2 rising edge after LRCLK2 rising edge (mode A)
1 = MSB is available on 1st BCLK2 rising edge after LRCLK2 rising edge (mode B)
Allows LRCLK2 to be enabled in Slave mode
0 = Normal
1 = LRCLK2 enabled in Slave mode
LRCLK2 Rate
LRCLK2 clock output =
BCLK2 / AIF2DAC_RATE
Integer (LSB = 1)
Valid from 8..2047
Table 103 AIF2 BCLK and LRCLK Control
AIF2 - DIGITAL AUDIO DATA CONTROL
The register bits controlling the audio data format, word length, left/right channel selection and TDM control for AIF2 are described in Table 104.
When TDM mode is enabled on AIF2, the WM8958 can transmit and receive audio data in Slot 0 or
Slot 1. In this case, the ADCDAT2 output is tri-stated during the unused timeslot, allowing another device to transmit data on the same pin. See “Signal Timing Requirements” for the associated timing details. (Note that, when TDM is not enabled on AIF2, the ADCDAT2 output is driven logic ‘0’ during the unused timeslot.)
In DSP mode, the left channel MSB is available on either the 1 st (mode B) or 2 nd (mode A) rising edge of BCLK following a rising edge of LRCLK (assuming default BCLK polarity).
When the AIF2DAC_LRCLK_INV bit is set in DSP mode, then DSP Mode B is selected for the AIF2 digital input (playback) signal path. When the AIF2DAC_LRCLK_INV bit is not set, then DSP Mode A is selected.
When the AIF2ADC_LRCLK_INV bit is set in DSP mode, then DSP Mode B is selected for the AIF2 digital output (record) signal path. When the AIF2ADC_LRCLK_INV bit is not set, then DSP Mode A is selected.
Note that the DSP Mode is selected independently for the input/output paths of each digital audio interface. Also note that the AIF2ADCLRCLK_INV bits remain valid even when the LRCLK signal is common for both paths. See Table 103 for details of the AIF2DAC_LRCLK_INV and
AIF2ADC_LRCLK_INV register fields.
A digital gain function is available at the audio interface input path to boost the DAC volume when a small signal is received on DACDAT2. This is controlled using the AIF2DAC_BOOST register. To prevent clipping, this function should not be used when the boosted data is expected to be greater than 0dBFS.
DESCRIPTION REGISTER
ADDRESS
R784 (0310h)
AIF2 Control
(1)
15
14
13
AIF2ADCL_
SRC
AIF2ADCR_
SRC
AIF2ADC_T
DM
0
1
0
AIF2 Left Digital Audio interface source
0 = Left ADC data is output on left channel
1 = Right ADC data is output on left channel
AIF2 Right Digital Audio interface source
0 = Left ADC data is output on right channel
1 = Right ADC data is output on right channel
AIF2 transmit (ADC) TDM Enable
0 = Normal ADCDAT2 operation
1 = TDM enabled on ADCDAT2
PP, August 2012, Rev 3.4
191
WM8958
REGISTER
ADDRESS
R785 (0311h)
AIF2 Control
(2)
R790 (0316h)
AIF2 DAC
Data
R791 (0317h)
AIF2 ADC
Data
12 AIF2ADC_T
DM_CHAN
6:5
4:3
15
14
13
12
AIF2DACL_
SRC
AIF2DACR_
SRC
11:10 AIF2DAC_B
1
0
1
0
Pre-Production
DESCRIPTION
AIF2_WL
[1:0]
AIF2_FMT
[1:0]
AIF2DAC_T
DM
AIF2DAC_T
DM_CHAN
OOST [1:0]
AIF2DACL_
DAT_INV
AIF2DACR_
DAT_INV
AIF2ADCL_
DAT_INV
AIF2ADCR_
DAT_INV
0
10
10
0
1
0
0
00
0
0
0
0
AIF2 transmit (ADC) TDM Slot Select
0 = Slot 0
1 = Slot 1
AIF2 Digital Audio Interface Word Length
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
AIF2 Digital Audio Interface Format
00 = Right justified
01 = Left justified
10 = I 2 S Format
11 = DSP Mode
AIF2 Left Receive Data Source Select
0 = Left DAC receives left interface data
1 = Left DAC receives right interface data
AIF2 Right Receive Data Source Select
0 = Right DAC receives left interface data
1 = Right DAC receives right interface data
AIF2 receive (DAC) TDM Enable
0 = Normal DACDAT2 operation
1 = TDM enabled on DACDAT2
AIF2 receive (DAC) TDM Slot Select
0 = Slot 0
1 = Slot 1
AIF2 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
AIF2 Left Receive Data Invert
0 = Not inverted
1 = Inverted
AIF2 Right Receive Data Invert
0 = Not inverted
1 = Inverted
AIF2 Left Transmit Data Invert
0 = Not inverted
1 = Inverted
AIF2 Right Transmit Data Invert
0 = Not inverted
1 = Inverted
Table 104 AIF2 Digital Audio Data Control w PP, August 2012, Rev 3.4
192
Pre-Production w
WM8958
AIF2 - MONO MODE
AIF2 can be configured to operate in mono DSP mode by setting AIF2_MONO = 1 as described in
Table 105. Note that mono mode is only supported in DSP mode, ie when AIF2_FMT = 11.
In mono mode, the Left channel data or the Right channel data may be selected for output on
ADCDAT2. The selected channel is determined by the AIF2ADCL_ENA and AIF2ADCR_ENA bits. (If both bits are set, then the Right channel data is selected.)
In mono mode, the DACDAT2 input can be enabled on the Left and/or Right signal paths using the
AIF2DACL_ENA and AIF2DACR_ENA bits. The mono input can be enabled on both paths at the same time if required.
DESCRIPTION REGISTER
ADDRESS
R785 (0311h)
AIF2 Control
(2)
8 AIF2_MONO 0 AIF2 DSP Mono Mode
0 = Disabled
1 = Enabled
Note that Mono Mode is only supported when
AIF2_FMT = 11.
Table 105 AIF2 Mono Mode Control
AIF2 - COMPANDING
The WM8958 supports A-law and -law companding on both transmit (ADC) and receive (DAC) sides of AIF2. This is configured using the register bits described in Table 106.
For more details on Companding, see the Audio Interface AIF1 description above.
REGISTER
ADDRESS
R785 (0311h)
AIF2 Control
(2)
4
3
2
1
AIF2DAC_C
OMP
AIF2DAC_C
OMPMODE
AIF2ADC_C
OMP
AIF2ADC_C
OMPMODE
0
0
0
0
DESCRIPTION
AIF2 Receive Companding Enable
0 = Disabled
1 = Enabled
AIF2 Receive Companding Type
0 = µ-law
1 = A-law
AIF2 Transmit Companding Enable
0 = Disabled
1 = Enabled
AIF2 Transmit Companding Type
0 = µ-law
1 = A-law
Table 106 AIF2 Companding
AIF2 - LOOPBACK
The AIF2 interface can provide a Loopback option. When the AIF2_LOOPBACK bit is set, then AIF2 digital audio output is routed to the AIF2 digital audio input. The normal input (DACDAT2) is not used when AIF2 Loopback is enabled.
REGISTER
ADDRESS
R785 (0311h)
AIF2 Control
(2)
0 AIF2_LOOP
BACK
0
DESCRIPTION
AIF2 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (ADCDAT2 data output is directly input to DACDAT2 data input).
Table 107 AIF2 Loopback
PP, August 2012, Rev 3.4
193
WM8958
Pre-Production
AIF2 - DIGITAL PULL-UP AND PULL-DOWN
The WM8958 provides integrated pull-up and pull-down resistors on each of the DACDAT2,
DACLRCLK2 and BCLK2 pins. This provides a flexible capability for interfacing with other devices.
Each of the pull-up and pull-down resistors can be configured independently using the register bits described in Table 108. Note that if the Pull-up and Pull-down are both enabled for any pin, then the pull-up and pull-down will be disabled.
Note that pull-up and pull-down resistors are also provided on the GPIO6/ADCLRCLK2 pin; this is described in the “General Purpose Input/Output” section.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R1794
(0702h)
Pull Control
(BCLK2)
R1795
(0703h)
Pull Control
(DACLRCLK2)
R1796
(0704h)
Pull Control
(DACDAT2)
14
13
14
13
14
13
BCLK2_PU
BCLK2_PD
DACLRCLK2_
PU
DACLRCLK2_
PD
DACDAT2_PU
DACDAT2_PD
0
1
0
1
0
1
BCLK2 Pull-up enable
0 = Disabled
1 = Enabled
BCLK2 Pull-down enable
0 = Disabled
1 = Enabled
DACLRCLK2 Pull-up enable
0 = Disabled
1 = Enabled
DACLRCLK2 Pull-down enable
0 = Disabled
1 = Enabled
DACDAT2 Pull-up enable
0 = Disabled
1 = Enabled
DACDAT2 Pull-down enable
0 = Disabled
1 = Enabled
Table 108 AIF2 Digital Pull-Up and Pull-Down Control w PP, August 2012, Rev 3.4
194
Pre-Production
WM8958
AIF3 - SIGNAL PATH CONFIGURATION AND TRI-STATE CONTROL
The AIF3 interface provides Mono PCM digital audio paths to/from the AIF2 DSP functions. The AIF3 interface can also support stereo digital audio paths via multiplexers to provide alternate connections to AIF1 or AIF2. The relevant multiplexers are illustrated in Figure 67.
Left / Right source select / Mono Mix control
AIF2DAC_SRC
Left / Right source select /
Mono Mix control
AIF3ADC_SRC[1:0]
0R 0L 1R 1L 0R 0L 1R 1L
DIGITAL AUDIO
INTERFACE 1 (AIF1)
AIF1_DACDAT_SRC
0R 0L 0R 0L
DIGITAL AUDIO
INTERFACE 2 (AIF2)
MONO PCM
INTERFACE
AIF2_DACDAT_SRC
AIF3_ADCDAT_SRC[1:0]
LRCLK1 LRCLK2
AIF2_ADCDAT_SRC w
Figure 67 Audio Interface AIF3 Configuration
Note that all of the AIF3 connections are supported on pins which also provide GPIO functions. These pins must be configured as AIF functions when used as audio interface pins. See “General Purpose
Input/Output”.
The GPIO8 pin supports the DACDAT3 function, which provides the input to the AIF3 Mono PCM interface.
When AIF3 Mono PCM input is used, this must be configured as an input to the AIF2 input paths using the AIF2DAC_SRC register as described in Table 109. The AIF3 Mono input may be selected on either channel (Left or Right), with AIF2 input enabled on the opposite channel at the same time.
When AIF3 Mono PCM input is used, the AIF2 input paths must be enabled using the
AIF2DACR_ENA and AIF2DACL_ENA register bits defined in Table 102.
The DACDAT3 input pin can also be used as an input (mono or stereo) to AIF1 or AIF2. The data input source for AIF1 is selected using the AIF1_DACDAT_SRC register. The data input source for
AIF2 is selected using the AIF2_DACDAT_SRC register.
The DACDAT3 input pin can also be routed to the ADCDAT2 output. The ADCDAT2 source is selected using the AIF2_ADCDAT_SRC register.
PP, August 2012, Rev 3.4
195
WM8958 w
Pre-Production
The GPIO9 pin supports the ADCDAT3 function, which supports the output from the AIF3 Mono PCM interface. The source for the ADCDAT3 pin is selected using the AIF3_ADCDAT_SRC register.
When AIF3 Mono PCM output is used, the data source must be configured using the AIF3ADC_SRC register; this selects either the Left or Right AIF2 output paths as the data source.
When AIF3 Mono PCM output is used, the AIF2 output paths must be enabled using the
AIF2ADCR_ENA and AIF2ADCL_ENA register bits. Note that, if AIF3 Mono PCM output is required and AIF2 output is not used, then the AIF2 output can be disabled using the AIF2TXL_ENA and
AIF2TXR_ENA registers. See Table 102 for details of these registers.
The ADCDAT3 pin can also be used as an alternate data output (mono or stereo) from AIF1 or AIF2, or can be connected to the DACDAT2 data input.
The AIF3 interface can be tri-stated by setting the AIF3_TRI register. When this bit is set, then all of the AIF3 outputs are un-driven (high-impedance). The AIF3_TRI register only affects those pins which are configured for AIF3 functions; it does not affect pins which are configured for other functions.
The AIF3 control registers are described in Table 109.
REGISTER
ADDRESS
R6 (0006h)
Power
Management
(6)
10:9 AIF3ADC_S
RC [1:0]
00
8:7
5
4:3
2
1
AIF2DAC_S
RC [1:0]
AIF3_TRI
AIF3_ADCD
AT_SRC
[1:0]
AIF2_ADCD
AT_SRC
AIF2_DACD
AT_SRC
00
0
00
0
0
DESCRIPTION
AIF3 Mono PCM output source select
00 = None
01 = AIF2ADC (Left) output path
10 = AIF2ADC (Right) output path
11 = Reserved
AIF2 input path select
00 = Left and Right inputs from AIF2
01 = Left input from AIF2; Right input from
AIF3
10 = Left input from AIF3; Right input from
AIF2
11 = Reserved
AIF3 Audio Interface tri-state
0 = AIF3 pins operate normally
1 = Tri-state all AIF3 interface pins
Note that pins not configured as AIF3 functions are not affected by this register.
GPIO9/ADCDAT3 Source select
00 = AIF1 ADCDAT1
01 = AIF2 ADCDAT2
10 = DACDAT2
11 = AIF3 Mono PCM output
Note that GPIO9 must be configured as
ADCDAT3.
ADCDAT2 Source select
0 = AIF2 ADCDAT2
1 = GPIO8/DACDAT3
For selection 1, the GPIO8 pin must also be configured as DACDAT3.
AIF2 DACDAT Source select
0 = DACDAT2
1 = GPIO8/DACDAT3
For selection 1, the GPIO8 pin must also be configured as DACDAT3.
PP, August 2012, Rev 3.4
196
Pre-Production
WM8958
REGISTER
ADDRESS
DESCRIPTION
0 AIF1_DACD
AT_SRC
0 AIF1 DACDAT Source select
0 = DACDAT1
1 = GPIO8/DACDAT3
Note that, for selection 1, the GPIO8 pin must be configured as DACDAT3.
Table 109 AIF3 Signal Path Configuration
AIF3 - BCLK AND LRCLK CONTROL
The GPIO10 pin supports the LRCLK3 function. When configured as LRCLK3, this pin outputs the
LRCLK signal from AIF1 or AIF2. The applicable AIF source is determined automatically as defined in
Table 110. Note that the LRCLK3 signal is also controlled by the logic illustrated in Figure 63 (AIF1) or Figure 66 (AIF2), depending on the selected AIF source.
The GPIO11 pin supports the BCLK3 function. When configured as BCLK3, this pin outputs the BCLK signal from AIF1 or AIF2. The applicable AIF source is determined automatically as defined in Table
110. Note that the BCLK3 signal is also controlled by the logic illustrated in Figure 63 (AIF1) or Figure
66 (AIF2), depending on the selected AIF source.
AIF1_DACDAT_SRC = 1
(DACDAT3 selected as AIF1 data input) or
CONDITION DESCRIPTION
AIF1 selected as BCLK3 / LRCLK3 source
AIF3_ADCDAT_SRC[1:0] = 00
(AIF1 data output selected on ADCDAT3)
All other conditions AIF2 selected as BCLK3 / LRCLK3 source
Table 110 BCLK3 / LRCLK3 Configuration
The LRCLK3 output can be inverted by setting the AIF3_LRCLK_INV register. Note that AIF3 operates in Master mode only.
DESCRIPTION REGISTER
ADDRESS
R800 (0320h)
AIF3 Control
(1)
7 AIF3_LRCL
K_INV
0 Right, left and I 2 S modes – LRCLK3 polarity
0 = normal LRCLK3 polarity
1 = invert LRCLK3 polarity
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK3 rising edge after LRCLK3 rising edge (mode A)
1 = MSB is available on 1st BCLK3 rising edge after LRCLK3 rising edge (mode B)
Table 111 AIF3 LRCLK Control w PP, August 2012, Rev 3.4
197
WM8958
Pre-Production
AIF3 - DIGITAL AUDIO DATA CONTROL
The register bits controlling the AIF3 Mono PCM interface are described in Table 112.
Note that these registers control the AIF3 Mono PCM interface only; they are not applicable to the
ADCDAT3 and DACDAT3 signal paths when these pins are selected as alternate inputs to the AIF1 or AIF2 interfaces.
The audio data format for AIF3 is set the same as AIF2; this is controlled using the AIF2_FMT register, as described in see Table 104.
In DSP mode, the AIF3 Mono channel MSB is available on either the 1 st (mode B) or 2 nd (mode A) rising edge of BCLK following a rising edge of LRCLK. The applicable DSP mode is selected using the AIF3_LRCLK_INV bit, as described in Table 111.
In Left justified, Right justified and I2S modes, the AIF3 Mono interface data is transmitted and received in the Left channel data bits of the ADCDAT3 and DACDAT3 channels.
A digital gain function is available at the audio interface input path to boost the DAC volume when a small signal is received on DACDAT3. This is controlled using the AIF3DAC_BOOST register. To prevent clipping, this function should not be used when the boosted data is expected to be greater than 0dBFS.
DESCRIPTION REGISTER
ADDRESS
R800 (0320h)
AIF3 Control
(1)
R801 (0321h)
AIF3 Control
(2)
R802 (0322h)
AIF3DAC
Data
R803 (0323h)
AIF3ADC
Data
6:5 AIF3_WL
[1:0]
11:10 AIF3DAC_B
0
0
OOST [1:0]
AIF3DAC_D
AT_INV
AIF3ADC_D
AT_INV
10
00
0
0
AIF3 Digital Audio Interface Word Length
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
Note that this controls the AIF3 Mono PCM interface path only; it does not affect AIF3 inputs/outputs routed to AIF1 or AIF2.
AIF3 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
DACDAT3 input to AIF1 or AIF2.
AIF3 Receive Data Invert
0 = Not inverted
1 = Inverted
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
DACDAT3 input to AIF1 or AIF2.
AIF3 Transmit Data Invert
0 = Not inverted
1 = Inverted
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
ADCDAT3 output from AIF1 or AIF2.
Table 112 AIF3 Digital Audio Data Control w PP, August 2012, Rev 3.4
198
Pre-Production
WM8958
AIF3 - COMPANDING
The WM8958 supports A-law and -law companding on both transmit (ADC) and receive (DAC) sides of AIF3. This is configured using the register bits described in Table 113.
Note that these registers control the AIF3 Mono PCM interface only; they are not applicable to the
ADCDAT3 and DACDAT3 signal paths when these pins are selected as alternate inputs to the AIF1 or AIF2 interfaces.
For more details on Companding, see the Audio Interface AIF1 description above.
REGISTER
ADDRESS
R801 (0321h)
AIF3 Control
(2)
4
3
2
1
AIF3DAC_C
OMP
AIF3DAC_C
OMPMODE
AIF3ADC_C
OMP
AIF3ADC_C
OMPMODE
0
0
0
0
DESCRIPTION
AIF3 Receive Companding Enable
0 = Disabled
1 = Enabled
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
DACDAT3 input to AIF1 or AIF2.
AIF3 Receive Companding Type
0 = µ-law
1 = A-law
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
DACDAT3 input to AIF1 or AIF2.
AIF3 Transmit Companding Enable
0 = Disabled
1 = Enabled
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
ADCDAT3 output from AIF1 or AIF2.
AIF3 Transmit Companding Type
0 = µ-law
1 = A-law
Note that this controls the AIF3 Mono PCM interface path only; it does not affect
ADCDAT3 output from AIF1 or AIF2.
Table 113 AIF3 Companding
AIF3 - LOOPBACK
The AIF3 interface can provide a Loopback option. When the AIF3_LOOPBACK bit is set, then AIF3
Mono PCM output is routed to the AIF3 Mono PCM input. The normal input (DACDAT3) is not used when AIF3 Loopback is enabled.
REGISTER
ADDRESS
R801 (0321h)
AIF3 Control
(2)
0 AIF3_LOOP
BACK
0
DESCRIPTION
AIF3 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (AIF3 Mono PCM data output is directly input to AIF3 Mono PCM data input).
Table 114 AIF3 Loopback w PP, August 2012, Rev 3.4
199
WM8958
Pre-Production
CLOCKING AND SAMPLE RATES
The WM8958 requires a clock for each of the Digital Audio Interfaces (AIF1 and AIF2). These may be derived from a common clock reference, or from independent references. Under typical clocking configurations, many commonly-used audio sample rates can be derived directly from the external reference; for additional flexibility, the WM8958 incorporates two Frequency Locked Loop (FLL) circuits to perform frequency conversion and filtering.
External clock signals may be connected via MCLK1 and MCLK2. In AIF Slave modes, the BCLK or
LRCLK signals may be used as a reference for the AIF clocks.
The WM8958 performs stereo full-duplex sample rate conversion between the audio interfaces AIF1 and AIF2, enabling digital audio to be routed between the interfaces, and asynchronous audio data to be mixed together. See “Sample Rate Conversion” for further details.
In AIF Slave modes, it is important to ensure the applicable AIF clock (AIF1CLK or AIF2CLK) is synchronised with the associated external LRCLK. This can be achieved by selecting an MCLK input that is derived from the same reference as the LRCLK, or can be achieved by selecting the external
BCLK or LRCLK signals as a reference input to one of the FLLs, as a source for the AIF clock.
If the AIF clock is not synchronised with the LRCLK, then clicks arising from dropped or repeated audio samples will occur, due to the inherent tolerances of multiple, asynchronous, system clocks.
See “Applications Information” for further details on valid clocking configurations.
Clocking for the Audio Interfaces is provided by AIF1CLK and AIF2CLK for AIF1 and AIF2 respectively. An additional internal clock, SYSCLK is derived from either AIF1CLK or AIF2CLK in order to support the DSP core functions, Charge Pump, Class D switching amplifier, DC servo control, Control Write Sequencer and other internal functions. A further clock, DSP2CLK, is derived from either AIF1CLK or AIF2CLK in order to support the Multiband Compressor (MBC) function.
The following operating limits must be observed when configuring the WM8958 clocks. Failure to observe these limits will result in degraded performance and/or incorrect system functionality. Latency in the WM8958 signal paths is reduced at high SYSCLK frequencies; power consumption is reduced at low SYSCLK frequencies.
SYSCLK
SYSCLK
SYSCLK
AIF1CLK
AIF1CLK
AIF2CLK
AIF2CLK
DSP2CLK
Note that, if DAC_OSR128 = 0 and ADC_OSR128 = 0, then a slower SYSCLK frequency is possible; in this case, the requirement is SYSCLK 2.048MHz.
Note that, under specific operating conditions, clocking ratios of 128 x fs and 192 x fs are possible; this is described in the “Digital to Analogue Converter (DAC)” section. w
The SYSCLK frequency must be 256 x fs, (where fs is the faster rate of AIF1_SR or AIF2_SR). The
SYSCLK frequency is derived from AIF1CLK or AIF2CLK, as selected by the SYSCLK_SRC register
(see Table 119).
PP, August 2012, Rev 3.4
200
Pre-Production
WM8958
Note that the bandwidth of the digital audio mixing paths will be determined by the sample rate of whichever AIF is selected as the SYSCLK source. When using only one audio interface, the active interface should be selected as the SYSCLK source. For best audio performance when using AIF1 and AIF2 simultaneously, the SYSCLK source must select the AIF with the highest sample rate
(AIFn_SR).
The AIFnCLK / fs ratio is the ratio of AIFnCLK to the AIFn sample rate, where ‘n’ identifies the applicable audio interface AIF1 or AIF2. The AIF clocking ratio and sample rate are set by the
AIFnCLK_RATE and AIFn_SR register fields, defined in Table 116 and Table 118.
Note that, in the case of mixed input/output path sample rates on either interface, then
AIFnCLK_RATE and AIFn_SR are set according to the higher of the two sample rates.
The clocking configuration for AIF1CLK, AIF2CLK, SYSCLK and DSP2CLK is illustrated in Figure 68.
The SYSCLK_SRC register is defined in Table 119.
The WM8958 provides integrated pull-up and pull-down resistors on the MCLK1 and MCLK2 pins.
This provides a flexible capability for interfacing with other devices. This is configured as described in
Table 119. Note that if the Pull-up and Pull-down are both enabled for any pin, then the pull-up and pull-down will be disabled.
Figure 68 Audio Interface Clock Control
AIF1CLK ENABLE
The AIF1CLK_SRC register is used to select the AIF1CLK source. The source may be MCLK1,
MCLK2, FLL1 or FLL2. If either of the Frequency Locked Loops is selected as the source, then the
FLL(s) must be enabled and configured as described later.
The AIF1CLK clock may be adjusted by the AIF1CLK_DIV divider, which provides a divide-by-two option. The selected source may also be inverted by setting the AIF1CLK_INV bit.
The maximum AIF1CLK frequency is specified in the “Electrical Characteristics” section. Note that, when AIF1CLK_DIV = 1, the maximum frequency limit applies to the divided-down AIF1CLK frequency.
The AIF1CLK is enabled by the register bit AIF1CLK_ENA. This bit should be set to 0 when reconfiguring the clock sources. It is not recommended to change AIF1CLK_SRC while the
AIF1CLK_ENA bit is set. w PP, August 2012, Rev 3.4
201
WM8958
REGISTER
ADDRESS
R512 (0200h)
AIF 1
Clocking (1)
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
4:3 AIF1CLK_SR
C
2
1
0
AIF1CLK_INV
AIF1CLK_DIV
AIF1CLK_EN
A
00
0
0
0
AIF1CLK Source Select
00 = MCLK1
01 = MCLK2
10 = FLL1
11 = FLL2
AIF1CLK Invert
0 = AIF1CLK not inverted
1 = AIF1CLK inverted
AIF1CLK Divider
0 = AIF1CLK
1 = AIF1CLK / 2
AIF1CLK Enable
0 = Disabled
1 = Enabled
Table 115 AIF1CLK Enable
AIF1 CLOCKING CONFIGURATION
The WM8958 supports a wide range of standard audio sample rates from 8kHz to 96kHz. The AIF1 clocking configuration is selected using 4 control fields, which are set according to the required AIF digital audio sample rate, and the ADC/DAC clocking rate.
The AIF1_SR register is set according to the AIF1 sample rate. Note that 88.2kHz and 96kHz modes are supported for AIF1 input (DAC playback) only.
The AIF1CLK_RATE register is set according to the ratio of AIF1CLK to the AIF1 sample rate. Note that there a some restrictions on the supported clocking ratios, depending on the selected sample rate and operating conditions. The supported configurations are detailed in the “Digital Microphone
Interface”, “Analogue to Digital Converter (ADC)” and “Digital to Analogue Converter (DAC)” sections, according to each applicable function.
The audio interface can support different sample rates for the input data (DAC path) and output data
(ADC path) simultaneously. In this case, the AIF1_SR and AIF1CLK_RATE fields should be set according to the faster of the two sample rates.
When different sample rates are used for input data (DAC path) and output data (ADC path), the clocking of the slower path is set using AIF1DAC_DIV (if the AIF input path has the slower sample rate) or AIF1ADC_DIV (if the AIF output path has the slower sample rate). The appropriate divider is set according to the ratio of the two sample rates.
For example, if AIF1 input uses 48kHz sample rate, and AIF1 output uses 8kHz, then AIF1ADC_DIV should be set to 110b (divide by 6).
Note that the audio interface cannot support every possible combination of input and output sample rate simultaneously, but only where the ratio of the sample rates matches the available AIF1ADC_DIV or AIF1DAC_DIV divider settings.
Note that the WM8958 performs sample rate conversion, where necessary, to provide digital mixing and interconnectivity between the Audio Interfaces and the DSP Core functions. One stereo Sample
Rate Converter (SRC) is provided for audio input; a second stereo SRC is provided for audio output.
Each SRC is automatically configured on AIF1 or AIF2, depending on the selected Clocking and
Sample Rate settings. The WM8958 cannot support configurations that would require SRC on the input or output paths of both interfaces simultaneously. See “Sample Rate Conversion” for further details. w PP, August 2012, Rev 3.4
202
Pre-Production w
REGISTER
ADDRESS
R513 (0201h)
AIF 1
Clocking (2)
BIT LABEL DEFAULT
5:3 AIF1DAC_DIV 000
R528 (0210h)
AIF1 Rate
2:0
7:4
3:0
AIF1ADC_DIV
AIF1_SR
AIF1CLK_RAT
E
000
1000
0011
WM8958
DESCRIPTION
Selects the AIF1 input path sample rate relative to the AIF1 output path sample rate.
This field should only be changed from default in modes where the AIF1 input path sample rate is slower than the AIF1 output path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF1 output path sample rate relative to the AIF1 input path sample rate.
This field should only be changed from default in modes where the AIF1 output path sample rate is slower than the AIF1 input path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF1 Sample Rate (fs)
0000 = 8kHz
0001 = 11.025kHz
0010 = 12kHz
0011 = 16kHz
0100 = 22.05kHz
0101 = 24kHz
0110 = 32kHz
0111 = 44.1kHz
1000 = 48kHz
1001 = 88.2kHz
1010 = 96kHz
All other codes = Reserved
Note that 88.2kHz and 96kHz modes are supported for AIF1 input (DAC playback) only.
Selects the AIF1CLK / fs ratio
0000 = Reserved
0001 = 128
0010 = 192
0011 = 256
0100 = 384
0101 = 512
0110 = 768
0111 = 1024
1000 = 1408
1001 = 1536
All other codes = Reserved
Table 116 AIF1 Clocking Configuration
PP, August 2012, Rev 3.4
203
WM8958 w
AIF2CLK ENABLE
Pre-Production
The AIF2CLK_SRC register is used to select the AIF2CLK source. The source may be MCLK1,
MCLK2, FLL1 or FLL2. If either of the Frequency Locked Loops is selected as the source, then the
FLL(s) must be enabled and configured as described later.
The AIF2CLK clock may be adjusted by the AIF2CLK_DIV divider, which provides a divide-by-two option. The selected source may also be inverted by setting the AIF2CLK_INV bit.
The maximum AIF2CLK frequency is specified in the “Electrical Characteristics” section. Note that, when AIF2CLK_DIV = 1, the maximum frequency limit applies to the divided-down AIF2CLK frequency.
The AIF2CLK is enabled by the register bit AIF2CLK_ENA. This bit should be set to 0 when reconfiguring the clock sources. It is not recommended to change AIF2CLK_SRC while the
AIF2CLK_ENA bit is set.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R516 (0204h)
AIF 2
Clocking (1)
4:3
2
1
0
AIF2CLK_SR
C
AIF2CLK_INV
AIF2CLK_DIV
AIF2CLK_EN
A
00
0
0
0
AIF2CLK Source Select
00 = MCLK1
01 = MCLK2
10 = FLL1
11 = FLL2
AIF2CLK Invert
0 = AIF2CLK not inverted
1 = AIF2CLK inverted
AIF2CLK Divider
0 = AIF2CLK
1 = AIF2CLK / 2
AIF2CLK Enable
0 = Disabled
1 = Enabled
Table 117 AIF2CLK Enable
AIF2 CLOCKING CONFIGURATION
The WM8958 supports a wide range of standard audio sample rates from 8kHz to 96kHz. The AIF2 clocking configuration is selected using 4 control fields, which are set according to the required AIF digital audio sample rate, and the ADC/DAC clocking rate.
The AIF2_SR register is set according to the AIF2 sample rate. Note that 88.2kHz and 96kHz modes are supported for AIF2 input (DAC playback) only.
The AIF2CLK_RATE register is set according to the ratio of AIF2CLK to the AIF2 sample rate. Note that there a some restrictions on the supported clocking ratios, depending on the selected sample rate and operating conditions. The supported configurations are detailed in the “Digital Microphone
Interface”, “Analogue to Digital Converter (ADC)” and “Digital to Analogue Converter (DAC)” sections, according to each applicable function.
The audio interface can support different sample rates for the input data (DAC path) and output data
(ADC path) simultaneously. In this case, the AIF2_SR and AIF2CLK_RATE fields should be set according to the faster of the two sample rates.
When different sample rates are used for input data (DAC path) and output data (ADC path), the clocking of the slower path is set using AIF2DAC_DIV (if the AIF input path has the slower sample rate) or AIF2ADC_DIV (if the AIF output path has the slower sample rate). The appropriate divider is set according to the ratio of the two sample rates.
For example, if AIF2 input uses 48kHz sample rate, and AIF2 output uses 8kHz, then AIF2ADC_DIV should be set to 110b (divide by 6).
Note that the audio interface cannot support every possible combination of input and output sample rate simultaneously, but only where the ratio of the sample rates matches the available AIF2ADC_DIV or AIF2DAC_DIV divider settings.
PP, August 2012, Rev 3.4
204
Pre-Production
WM8958
Note that the WM8958 performs sample rate conversion, where necessary, to provide digital mixing and interconnectivity between the Audio Interfaces and the DSP Core functions. One stereo Sample
Rate Converter (SRC) is provided for audio input; a second stereo SRC is provided for audio output.
Each SRC is automatically configured on AIF1 or AIF2, depending on the selected Clocking and
Sample Rate settings. The WM8958 cannot support configurations that would require SRC on the input or output paths of both interfaces simultaneously. See “Sample Rate Conversion” for further details.
DESCRIPTION REGISTER
ADDRESS
R517
(0205h)
AIF 2
Clocking
(2)
BIT LABEL DEFAULT
5:3 AIF2DAC_DIV 000
R529
(0211h)
AIF2
Rate
2:0
7:4
AIF2ADC_DIV
AIF2_SR
000
1000
Selects the AIF2 input path sample rate relative to the AIF2 output path sample rate.
This field should only be changed from default in modes where the AIF2 input path sample rate is slower than the AIF2 output path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF2 output path sample rate relative to the AIF2 input path sample rate.
This field should only be changed from default in modes where the AIF2 output path sample rate is slower than the AIF2 input path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF2 Sample Rate (fs)
0000 = 8kHz
0001 = 11.025kHz
0010 = 12kHz
0011 = 16kHz
0100 = 22.05kHz
0101 = 24kHz
0110 = 32kHz
0111 = 44.1kHz
1000 = 48kHz
1001 = 88.2kHz
1010 = 96kHz
All other codes = Reserved
Note that 88.2kHz and 96kHz modes are supported for AIF2 input (DAC playback) only. w PP, August 2012, Rev 3.4
205
WM8958 w
Pre-Production
REGISTER
ADDRESS
BIT LABEL DEFAULT
3:0 AIF2CLK_RAT
E
0011
DESCRIPTION
Selects the AIF2CLK / fs ratio
0000 = Reserved
0001 = 128
0010 = 192
0011 = 256
0100 = 384
0101 = 512
0110 = 768
0111 = 1024
1000 = 1408
1001 = 1536
All other codes = Reserved
Table 118 AIF2 Clocking Configuration
MISCELLANEOUS CLOCK CONTROLS
SYSCLK provides clocking for many of the WM8958 functions. SYSCLK clock is required to support
DSP Core functions and also the Charge Pump, Class D switching amplifier, DC servo control,
Control Write Sequencer and other internal functions.
The SYSCLK_SRC register is used to select the SYSCLK source. The source may be AIF1CLK or
AIF2CLK, as illustrated in Figure 69. Note that the bandwidth of the digital audio mixing paths will be determined by the sample rate of whichever AIF is selected as the SYSCLK source. When using only one audio interface, the active interface should be selected as the SYSCLK source. For best audio performance when using AIF1 and AIF2 simultaneously, the SYSCLK source must select the AIF with the highest sample rate (AIFn_SR).
The SYSCLK_SRC register is also used to select the DSP2CLK source; the DSP2CLK clock is required for the Multiband Compressor (MBC) function. The MBC can enabled on the AIF1 or AIF2 input paths, regardless of the SYSCLK_SRC setting, provided that the minimum clocking requirement for the MBC is satisfied. See “Multiband Compressor” for further details.
The MBC clocking is enabled using the DSP2CLK_ENA bit, as illustrated in Figure 69. See
“Multiband Compressor” for further details of the MBC clocking requirements.
The AIF1 DSP processing clock is derived from SYSCLK, and enabled by AIF1DSPCLK_ENA.
The AIF2 DSP processing clock is derived from SYSCLK, and enabled by AIF2DSPCLK_ENA.
The clocking of the WM8958 ADC, DAC, digital mixer and digital microphone functions is enabled by setting SYSDSPCLK_ENA. See “Digital Microphone Interface” for details of the DMICCLK frequency.
Two modes of ADC / Digital Microphone operation can be selected using the ADC_OSR128 bit. This bit is enabled by default, giving best audio performance. De-selecting this bit provides a low power alternative setting.
A high performance mode of DAC operation can be selected by setting the DAC_OSR128 bit. When the DAC_OSR128 bit is set, the audio performance is improved, but power consumption is also increased.
A clock is required for the Charge Pump circuit when the ground-referenced headphone outputs
(HPOUT1L and HPOUT1R) are enabled. The Charge Pump clock is derived from SYSCLK whenever the Charge Pump is enabled. The Charge Pump clock division is configured automatically.
A clock is required for the Class D speaker driver circuit when the speaker outputs (SPKOUTL and
SPKOUTR) are enabled. The Class D clock is derived from SYSCLK whenever these outputs are enabled in Class D mode. The Class D clock division is configured automatically. See “Analogue
Outputs” for details of the Class D switching frequency.
A clock output (OPCLK) derived from SYSCLK may be output on a GPIO pin. This clock is enabled by register big OPCLK_ENA, and its frequency of this clock is controlled by OPCLK_DIV. See General
Purpose Input/Output” to configure a GPIO pin for this function.
PP, August 2012, Rev 3.4
206
Pre-Production
WM8958
A slow clock (TOCLK) is derived internally in order to control volume update timeouts when the zerocross option is selected. This clock is enabled by register bit TOCLK_ENA, and its frequency is controlled by TOCLK_DIV.
A de-bounce control is provided for GPIO inputs and for other functions that may be selected as
GPIO outputs. The de-bounced clock frequency is controlled by DBCLK_DIV.
The WM8958 generates a 256kHz clock for internal functions; TOCLK and DBCLK are derived from this 256kHz clock. In order to generate this clock correctly when SYSCLK_SRC = 0, valid settings are required for AIF1_SR and AIF1CLK_RATE. To generate this clock correctly when SYSCLK_SRC = 1, valid settings are required for AIF2_SR and AIF2CLK_RATE.
The WM8958 Clocking is illustrated in Figure 69. w
Figure 69 System Clocking
PP, August 2012, Rev 3.4
207
WM8958
REGISTER
ADDRESS
R2 (0002h)
Power
Management
(2)
R520 (0208h)
Clocking (1)
R521 (0209h)
Clocking (2)
Pre-Production
BIT LABEL DEFAULT DESCRIPTION
11
14
4
3
2
1
0
10:8
6:4
OPCLK_ENA
DSP2CLK_EN
A
TOCLK_ENA
AIF1DSPCLK
_ENA
AIF2DSPCLK
_ENA
SYSDSPCLK_
ENA
SYSCLK_SRC
TOCLK_DIV
DBCLK_DIV
0
0
0
0
0
0
0
000
000
GPIO Clock Output (OPCLK) Enable
0 = Disabled
1 = Enabled
MBC Processor Clock Enable
0 = Disabled
1 = Enabled
Slow Clock (TOCLK) Enable
0 = Disabled
1 = Enabled
This clock is required for zero-cross timeout.
AIF1 Processing Clock Enable
0 = Disabled
1 = Enabled
AIF2 Processing Clock Enable
0 = Disabled
1 = Enabled
Digital Mixing Processor Clock Enable
0 = Disabled
1 = Enabled
SYSCLK Source Select
0 = AIF1CLK
1 = AIF2CLK
Slow Clock (TOCLK ) Divider
(Sets TOCLK rate relative to 256kHz.)
000 = Divide by 256 (1kHz)
001 = Divide by 512 (500Hz)
010 = Divide by 1024 (250Hz)
011 = Divide by 2048 (125Hz)
100 = Divide by 4096 (62.5Hz)
101 = Divide by 8192 (31.2Hz)
110 = Divide by 16384 (15.6Hz)
111 = Divide by 32768 (7.8Hz)
De-bounce Clock (DBCLK) Divider
(Sets DBCLK rate relative to 256kHz.)
000 = Divide by 256 (1kHz)
001 = Divide by 2048 (125Hz)
010 = Divide by 4096 (62.5Hz)
011 = Divide by 8192 (31.2Hz)
100 = Divide by 16384 (15.6Hz)
101 = Divide by 32768 (7.8Hz)
110 = Divide by 65536 (3.9Hz)
111 = Divide by 131072 (1.95Hz) w PP, August 2012, Rev 3.4
208
Pre-Production
WM8958
REGISTER
ADDRESS
BIT LABEL DEFAULT DESCRIPTION
R1568
(0620h)
Oversampling
R1793
(0701h)
Pull Control
(MCLK2)
R1824
(0720h)
Pull Control
(1)
2:0 OPCLK_DIV
1
0
14
13
7
6
ADC_OSR128
DAC_OSR128
MCLK2_PU
MCLK2_PD
MCLK1_PU
MCLK1_PD
000
1
0
0
1
0
0
GPIO Output Clock (OPCLK) Divider
0000 = SYSCLK
0001 = SYSCLK / 2
0010 = SYSCLK / 3
0011 = SYSCLK / 4
0100 = SYSCLK / 5.5
0101 = SYSCLK / 6
0110 = SYSCLK / 8
0111 = SYSCLK / 12
1000 = SYSCLK / 16
1001 to 1111 = Reserved
ADC / Digital Microphone Oversample
Rate Select
0 = Low Power
1 = High Performance
DAC Oversample Rate Select
0 = Low Power
1 = High Performance
MCLK2 Pull-up enable
0 = Disabled
1 = Enabled
MCLK2 Pull-down enable
0 = Disabled
1 = Enabled
MCLK1 Pull-up enable
0 = Disabled
1 = Enabled
MCLK1 Pull-down enable
0 = Disabled
1 = Enabled
Table 119 System Clocking
BCLK AND LRCLK CONTROL
The digital audio interfaces (AIF1 and AIF2) use BCLK and LRCLK signals for synchronisation. In master mode, these are output signals, generated by the WM8958. In slave mode, these are input signals to the WM8958. It is also possible to support mixed master/slave operation.
The BCLK and LRCLK signals are controlled as illustrated in Figure 70. See the “Digital Audio
Interface Control” section for further details of the relevant control registers. w PP, August 2012, Rev 3.4
209
WM8958
AIF1CLK
AIF2CLK
Pre-Production
AIF1_BCLK_DIV [4:0]
AIF1DAC_RATE [10:0] f/N f/N
AIF1_MSTR
AIF1ADC_RATE [10:0] MASTER
MODE
CLOCK
OUTPUTS f/N
BCLK1
LRCLK1
ADCLRCLK1/
GPIO1
AIF2_BCLK_DIV [4:0]
AIF2DAC_RATE [10:0] f/N
AIF2_MSTR f/N
AIF2ADC_RATE [10:0]
MASTER
MODE
CLOCK
OUTPUTS f/N
BCLK2
LRCLK2
GPIO6/
ADCLRCLK2
Figure 70 BCLK and LRCLK Control w PP, August 2012, Rev 3.4
210
Pre-Production
WM8958
CONTROL INTERFACE CLOCKING
Register map access is possible with or without a system clock. Clocking is provided from SYSCLK; the SYSCLK_SRC register selects either AIF1CLK or AIF2CLK as the applicable SYSCLK source.
When AIF1CLK is the SYSCLK source (ie. SYSCLK_SRC = 0), and AIF1CLK_ENA = 1, then an active clock source for AIF1CLK must be present for control interface clocking. If the AIF1CLK source is stopped, then AIF1CLK_ENA must be set to 0 for control register access.
When AIF2CLK is the SYSCLK source (ie. SYSCLK_SRC = 1), and AIF2CLK_ENA = 1, then an active clock source for AIF2CLK must be present for control interface clocking. If the AIF2CLK source is stopped, then AIF2CLK_ENA must be set to 0 for control register access.
FREQUENCY LOCKED LOOP (FLL)
Two integrated FLLs are provided to support the clocking requirements of the WM8958. These can be enabled and configured independently according to the available reference clocks and the application requirements. The reference clock may be a high frequency (eg. 12.288MHz) or low frequency (eg.
32.768kHz).
The FLL is tolerant of jitter and may be used to generate a stable AIF clock from a less stable input reference. The FLL characteristics are summarised in “Electrical Characteristics”. Note that the FLL can be used to generate a free-running clock in the absence of an external reference source. This is described in the “Free-Running FLL Clock” section below.
The input reference for FLL1 is selected using FLL1_REFCLK_SRC. The available options are
MCLK1, MCLK2, BCLK1 or LRCLK1. The input reference for FLL2 is selected using
FLL2_REFCLK_SRC. The available options are MCLK1, MCLK2, BCLK2 or LRCLK2.
The FLLs can be bypassed using the FLL1_BYP or FLL2_BYP registers. This allows the BCLKn clock to be used as the AIFnCLK reference, without enabling the respective FLL.
The FLL input reference and bypass configurations are illustrated in Figure 71. w
Figure 71 FLL Input Reference Selection
The following description is applicable to FLL1 and FLL2. The associated register control fields are described in Table 122 for FLL1 and Table 123 for FLL2.
The FLL control registers are illustrated in Figure 72.
PP, August 2012, Rev 3.4
211
WM8958
Pre-Production
Figure 72 FLL Configuration
The FLL is enabled using the FLLn_ENA register bit (where n = 1 for FLL1 and n = 2 for FLL2). Note that the other FLL registers should be configured before enabling the FLL; the FLLn_ENA register bit should be set as the final step of the FLLn enable sequence.
When changing FLL settings, it is recommended that the digital circuit be disabled via FLLn_ENA and then re-enabled after the other register settings have been updated. When changing the input reference frequency F
REF
, it is recommended that the FLL be reset by setting FLLn_ENA to 0.
The field FLLn_REFCLK_DIV provides the option to divide the input reference (MCLK, BCLK or
LRCLK) by 1, 2, 4 or 8. This field should be set to bring the reference down to 13.5MHz or below. For best performance, it is recommended that the highest possible frequency - within the 13.5MHz limit - should be selected.
The FLL output frequency is directly determined from FLLn_FRATIO, FLLn_OUTDIV and the real number represented by N.K.
The integer value, N, is held in the FLLn_N register field. The fractional portion, K, is determined by the ratio FLLn_THETA / FLLn_LAMBDA.
Note that the FLLn_EFS_ENA register bit must be enabled in fractional mode (ie. whenever
FLLn_THETA > 0).
The FLL output frequency is generated according to the following equation:
F
OUT
= (F
VCO
/ FLLn_OUTDIV)
The FLL operating frequency, F
VCO
is set according to the following equation:
F
VCO
= (F
REF
x N.K x FLLn_FRATIO)
F
REF
is the input frequency, as determined by FLLn_REFCLK_DIV.
F
VCO
must be in the range 90-100 MHz. Frequencies outside this range cannot be supported.
Note that the output frequencies that do not lie within the ranges quoted above cannot be guaranteed across the full range of device operating conditions. w PP, August 2012, Rev 3.4
212
Pre-Production
WM8958
In order to follow the above requirements for F
VCO
, the value of FLLn_OUTDIV should be selected according to the desired output F
OUT
. The divider, FLLn_OUTDIV, must be set so that F
VCO
is in the range 90-100MHz. The available divisions are integers from 4 to 64. Some typical settings of
FLLn_OUTDIV are noted in Table 120.
OUTPUT FREQUENCY F
OUT
FLLn_OUTDIV
1.875 MHz - 2.0833 MHz 101111 (divide by 48)
2.8125 MHz - 3.125 MHz
3.75 MHz - 4.1667 MHz
011111 (divide by 32)
010111 (divide by 24)
5.625 MHz - 6.25 MHz
11.25 MHz - 12.5 MHz
18 MHz - 20 MHz
22.5 MHz - 25 MHz
001111 (divide by 16)
000111 (divide by 8)
000100 (divide by 5)
000011 (divide by 4)
Table 120 Selection of FLLn_OUTDIV
The value of FLLn_FRATIO should be selected as described in Table 121.
REFERENCE FREQUENCY F
REF
FLLn_FRATIO
1MHz - 13.5MHz 0h (divide by 1)
256kHz - 1MHz
128kHz - 256kHz
64kHz - 128kHz
Less than 64kHz
1h (divide by 2)
2h (divide by 4)
3h (divide by 8)
4h (divide by 16)
Table 121 Selection of FLLn_FRATIO
In order to determine the remaining FLL parameters, the FLL operating frequency, F
VCO
, must be calculated, as given by the following equation:
F
VCO
= (F
OUT
x FLLn_OUTDIV)
The value of N.K can then be determined as follows:
N.K = F
VCO
/ (FLLn_FRATIO x F
REF
)
Note that, in the above equations:
FLLn_OUTDIV is the F
OUT
clock ratio.
F
REF
is the input frequency, after division by FLL_REFCLK_DIV, where applicable.
FLLn_FRATIO is the F
VCO
clock ratio (1, 2, 4, 8 or 16).
The value of N is held in the FLLn_N register field.
The value of K is determined by the ratio FLLn_THETA / FLLn_LAMBDA.
The FLLn_N, FLLn_THETA and FLLn_LAMBDA fields are all coded as integers (LSB = 1). w PP, August 2012, Rev 3.4
213
WM8958 w
Pre-Production
In Fractional Mode (FLLn_THETA > 0 and FLLn_EFS_ENA = 1), the register fields FLLn_THETA and
FLLn_LAMBDA can be calculated as follows:
Calculate GCD(FLL) using the ‘Greatest Common Denominator’ function:
GCD(FLL) = GCD(FLLn_FRATIO x F
REF
, F
VCO
) where GCD(x, y) is the greatest common denominator of x and y
Next, calculate FLLn_THETA and FLLn_LAMBDA using the following equations:
FLLn_THETA = (F
VCO
- (FLLn_N x FLLn_FRATIO x F
REF
)) / GCD(FLL)
FLLn_LAMBDA = (FLLn_FRATIO x F
REF
) / GCD(FLL)
Note that, in Fractional Mode, the values of FLLn_THETA and FLLn_LAMBDA must be co-prime (ie. not divisible by any common integer). The calculation above ensures that the values will be co-prime.
The value of K must be a fraction less than 1 (ie. FLLn_THETA must be less than FLLn_LAMBDA).
The FLL1 control registers are described in Table 122. The FLL2 control registers are described in
Table 123. Example settings for a variety of reference frequencies and output frequencies are shown in Table 125.
DESCRIPTION REGISTER
ADDRESS
R544 (0220h)
FLL1 Control (1)
0 FLL1_ENA 0
R545 (0221h)
FLL1 Control (2)
R546 (0222h)
FLL1 Control (3)
R547 (0223h)
FLL1 Control (4)
13:8
2:0
15:0
14:5
FLL1_OUTDIV
[5:0]
FLL1_FRATIO
[2:0]
FLL1_THETA
[15:0]
FLL1_N [9:0]
000000
000
0000h
000h
FLL1 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the FLL1 enable sequence, ie. after the other FLL registers have been configured.
FLL1 F
OUT
clock divider
000000 = Reserved
000001 = Reserved
000010 = Reserved
000011 = 4
000100 = 5
000101 = 6
…
111110 = 63
111111 = 64
(F
OUT
= F
VCO
/ FLL1_OUTDIV)
FLL1 F
VCO
clock divider
000 = 1
001 = 2
010 = 4
011 = 8
1XX = 16
FLL Fractional multiply for F
REF
This field sets the numerator
(multiply) part of the FLL1_THETA /
FLL1_LAMBDA ratio.
Coded as LSB = 1.
FLL1 Integer multiply for F
REF
(LSB = 1)
PP, August 2012, Rev 3.4
214
Pre-Production
REGISTER
ADDRESS
R548 (0224h)
FLL1 Control (5)
R550 (0226h)
FLL1 EFS1
R551 (0227h)
FLL1 EFS2
15
4:3
1:0
2:1
0
FLL1_BYP
FLL1_REFCLK_
DIV [1:0]
FLL1_REFCLK_
SRC [1:0]
15:0 FLL1_LAMBDA
[15:0]
FLL1_EFS_ENA
0
00
00
0000h
11
0
WM8958
DESCRIPTION
FLL1 Bypass Select
0 = Disabled
1 = Enabled
When FLL1_BYP is set, the FLL1 output is derived directly from
BCLK1. In this case, FLL1 can be disabled.
FLL1 Clock Reference Divider
00 = MCLK / 1
01 = MCLK / 2
10 = MCLK / 4
11 = MCLK / 8
MCLK (or other input reference) must be divided down to <=13.5MHz.
For lower power operation, the reference clock can be divided down further if desired.
FLL1 Clock source
00 = MCLK1
01 = MCLK2
10 = LRCLK1
11 = BCLK1
FLL Fractional multiply for F
REF
This field sets the denominator
(dividing) part of the FLL1_THETA /
FLL1_LAMBDA ratio.
Coded as LSB = 1.
Reserved - Do not change
FLL Fractional Mode EFS enable
0 = Integer Mode
1 = Fractional Mode
This bit should be set to 1 when
FLL1_THETA > 0.
Table 122 FLL1 Register Map
REGISTER
ADDRESS
R576 (0240h)
FLL2 Control (1)
0 FLL2_ENA 0
DESCRIPTION
FLL2 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the FLL2 enable sequence, ie. after the other FLL registers have been configured. w PP, August 2012, Rev 3.4
215
WM8958 w
REGISTER
ADDRESS
R577 (0241h)
FLL2 Control (2)
R578 (0242h)
FLL2 Control (3)
R579 (0243h)
FLL2 Control (4)
R580 (0244h)
FLL2 Control (5)
R582 (0246h)
FLL2 EFS1
R583 (0247h)
Pre-Production
DESCRIPTION
13:8
2:0
15:0
14:5
15
4:3
1:0
15:0
2:1
FLL2_OUTDIV
[5:0]
FLL2_FRATIO
[2:0]
FLL2_THETA
[15:0]
FLL2_N [9:0]
FLL2_BYP
FLL2_REFCLK_
DIV [1:0]
FLL2_REFCLK_
SRC [1:0]
FLL2_LAMBDA
[15:0]
000000
000
0000h
000h
0
00
00
0000h
11
FLL2 F
OUT
clock divider
000000 = Reserved
000001 = Reserved
000010 = Reserved
000011 = 4
000100 = 5
000101 = 6
…
111110 = 63
111111 = 64
(F
OUT
= F
VCO
/ FLL2_OUTDIV)
FLL2 F
VCO
clock divider
000 = 1
001 = 2
010 = 4
011 = 8
1XX = 16
FLL Fractional multiply for F
REF
This field sets the numerator
(multiply) part of the FLL2_THETA /
FLL2_LAMBDA ratio.
Coded as LSB = 1.
FLL2 Integer multiply for F
REF
(LSB = 1)
FLL2 Bypass Select
0 = Disabled
1 = Enabled
When FLL2_BYP is set, the FLL2 output is derived directly from
BCLK2. In this case, FLL2 can be disabled.
FLL2 Clock Reference Divider
00 = MCLK / 1
01 = MCLK / 2
10 = MCLK / 4
11 = MCLK / 8
MCLK (or other input reference) must be divided down to <=13.5MHz.
For lower power operation, the reference clock can be divided down further if desired.
FLL2 Clock source
00 = MCLK1
01 = MCLK2
10 = LRCLK2
11 = BCLK2
FLL Fractional multiply for F
REF
This field sets the denominator
(dividing) part of the FLL2_THETA /
FLL2_LAMBDA ratio.
Coded as LSB = 1.
Reserved - Do not change
PP, August 2012, Rev 3.4
216
Pre-Production
WM8958
REGISTER
ADDRESS
FLL2 EFS2
DESCRIPTION
0 FLL2_EFS_ENA 0 FLL Fractional Mode EFS enable
0 = Integer Mode
1 = Fractional Mode
This bit should be set to 1 when
FLL2_THETA > 0.
Table 123 FLL2 Register Map
FREE-RUNNING FLL CLOCK
The FLL can generate a clock signal even when no external reference is available. However, it should be noted that the accuracy of this clock is reduced, and a reference source should always be used where possible. The free-running FLL modes are not sufficiently accurate for hi-fi ADC or DAC operations, but are suitable for clocking most other functions, including the Write Sequencer, Charge
Pump, DC Servo and Class D loudspeaker driver. The free-running FLL operation is ideal for clocking the accessory detection function during low-power standby operating conditions (see “External
Accessory Detection”).
If an accurate reference clock is initially available, then the FLL should be configured as described above. The FLL will continue to generate a stable output clock after the reference input is stopped or disconnected.
If no reference clock is available at the time of starting up the FLL, then an internal clock frequency of approximately 12MHz can be generated by implementing the following sequence:
Enable the FLL Analogue Oscillator (FLLn_OSC_ENA = 1)
Set the F
OUT
clock divider to divide by 8 (FLLn_OUTDIV = 000111)
Configure the oscillator frequency by setting FLLn_FRC_NCO = 1 and
FLLn_FRC_NCO_VAL = 19h
Note that the free-running FLL mode is not suitable for hi-fi CODEC applications. In the absence of any reference clock, the FLL output is subject to a very wide tolerance; see “Electrical Characteristics” for details of the FLL accuracy.
Note that the free-running FLL clock is selected as SYSCLK using the registers noted in Figure 68.
The free-running FLL clock may be used to support analogue functions, for which the digital audio interface is not used, and there is no applicable Sample Rate (fs). When SYSCLK is required for circuits such the Class D, DC Servo, Control Write Sequencer or Charge Pump, then valid Sample
Rate register settings are still required, even though the digital audio interface is not active.
For correct functionality when SYSCLK_SRC = 0, valid settings are required for AIF1_SR and
AIF1CLK_RATE. In the case where SYSCLK_SRC = 1, then valid settings are required for AIF2_SR and AIF2CLK_RATE.
The control registers applicable to FLL free-running modes are described in Table 124. w PP, August 2012, Rev 3.4
217
WM8958
REGISTER
ADDRESS
R544 (0220h)
FLL1 Control (1)
R548 (0224h)
FLL1 Control (5)
R576 (0240h)
FLL2 Control (1)
R580 (0244h)
FLL2 Control (5)
Pre-Production
DESCRIPTION
1
12:7
6
1
12:7
6
FLL1_OSC_ENA
FLL1_FRC_NCO
_VAL [5:0]
FLL1_FRC_NCO
FLL2_OSC_ENA
FLL2_FRC_NCO
_VAL [5:0]
FLL2_FRC_NCO
0
19h
0
0
19h
0
FLL1 Oscillator enable
0 = Disabled
1 = Enabled
(Note that this field is required for freerunning FLL1 modes only)
FLL1 Forced oscillator value
Valid range is 000000 to 111111
0x19h (011001) = 12MHz approx
(Note that this field is required for freerunning FLL modes only)
FLL1 Forced control select
0 = Normal
1 = FLL1 oscillator controlled by
FLL1_FRC_NCO_VAL
(Note that this field is required for freerunning FLL modes only)
FLL2 Oscillator enable
0 = Disabled
1 = Enabled
(Note that this field is required for freerunning FLL2 modes only)
FLL2 Forced oscillator value
Valid range is 000000 to 111111
0x19h (011001) = 12MHz approx
(Note that this field is required for freerunning FLL modes only)
FLL2 Forced control select
0 = Normal
1 = FLL2 oscillator controlled by
FLL2_FRC_NCO_VAL
(Note that this field is required for freerunning FLL modes only)
Table 124 FLL Free-Running Mode
GPIO OUTPUTS FROM FLL
For each FLL, the WM8958 has an internal signal which indicates whether the FLL Lock has been achieved. The FLL Lock status is an input to the Interrupt control circuit and can be used to trigger an
Interrupt event - see “Interrupts”.
The FLL Lock signal can be output directly on a GPIO pin as an external indication of FLL Lock. See
“General Purpose Input/Output” for details of how to configure a GPIO pin to output the FLL Lock signal.
The FLL Clock can be output directly on a GPIO pin as a clock signal for other circuits. Note that the
FLL Clock may be output even if the FLL is not selected as the WM8958 SYSCLK source. The FLL clocking configuration is illustrated in Figure 71. See “General Purpose Input/Output” for details of how to configure a GPIO pin to output the FLL Clock. w PP, August 2012, Rev 3.4
218
Pre-Production
WM8958
EXAMPLE FLL CALCULATION
The following example illustrates how to derive the FLL1 registers to generate 12.288 MHz output
(F
OUT
) from a 12.000 MHz reference clock (F
REF
):
Set FLL1_REFCLK_DIV in order to generate F
REF
<=13.5MHz:
FLL1_REFCLK_DIV = 00 (divide by 1)
Set FLL1_OUTDIV for the required output frequency as shown in Table 120:-
F
OUT
= 12.288 MHz, therefore FLL1_OUTDIV = 7h (divide by 8)
Set FLL1_FRATIO for the given reference frequency as shown in Table 121:
F
REF
= 12MHz, therefore FLL1_FRATIO = 0h (divide by 1)
VCO
as given by F
VCO
= F
OUT
x FLL1_OUTDIV:-
F
VCO
= 12.288 x 8 = 98.304MHz
Calculate N.K as given by N.K = F
VCO
/ (FLL1_FRATIO x F
REF
):
N.K = 98.304 / (1 x 12) = 8.192
Set FLL1_EFS_ENA according to whether N.K is an integer.
N.K has a fractional part, therefore FLL1_EFS_ENA = 1
Determine FLL1_N from the integer portion of N.K:-
FLL_N = 8.
Determine GCD(FLL), as given by GCD(FLL) = GCD(FLL1_FRATIO x F
REF
, F
VCO
):
GCD(FLL) = GCD(1 x 12000000, 98304000) = 96000
Determine FLL1_THETA, as given by
FLL1_THETA = (F
VCO
- (FLL1_N x FLL1_FRATIO x F
REF
)) / GCD(FLL):
FLL1_THETA = (98304000 - (8 x 1 x 12000000)) / 96000
FLL1_THETA = 24 (0018h)
Determine FLL_LAMBDA, as given by
FLL1_LAMBDA = (FLL1_FRATIO x F
REF
) / GCD(FLL):
FLL1_LAMBDA = (1 x 12000000) / 96000
FLL1_LAMBDA = 125 (007Dh) w PP, August 2012, Rev 3.4
219
WM8958
Pre-Production
EXAMPLE FLL SETTINGS
Table 125 provides example FLL settings for generating common SYSCLK frequencies from a variety of low and high frequency reference inputs.
F
SOURCE
32.000 kHz
32.000 kHz
32.768 kHz
32.768 kHz
44.1 kHz
48 kHz
128 kHz
128 kHz
512 kHz
F
OUT
(MHz) F
REF
Divider
12.288
N.K FRATIO F
VCO
(MHz) OUTDIV FLLn_N FLLn_
EFS_ENA
1 192 16 98.304 8 0C0h 0
FLLn_
THETA
FLLn_
LAMBDA
11.2896
12.288
11.2896
11.2896
12.288
2.048
12.288
2.048
512 kHz 12.288
1.4112 MHz 11.2896
1 128 16
1 128 16 98.304 8 080h 0
1 96 8 98.304 48 060h 0
1 96 8 98.304 8 060h 0
1 96 2 98.304 48 060h 0
1 96 2 98.304 8 060h 0
1 64 1
2.8224 MHz 11.2896
1.536 MHz
3.072 MHz
12.288
12.288
1 32 1
1 64 1 98.304 8 040h 0
1 32 1 98.304 8 020h 0
11.2896 12.288 1 8.7075 1 98.304 8 008h 1 0068h 0093h
12.000 MHz 12.288
12.000 MHz
12.288 MHz
11.2896
12.288
12.288 MHz 11.2896
13.000 MHz 12.288
13.000 MHz 11.2896
1 7.5264 1 90.3168 8 007h 1 0149h 0271h
1 8 1
1 7.5618 1 98.304 8 007h 1 0391h 0659h
1 6.9474 1 90.3168 8 006h 1 1E12h 1FBDh
19.200 MHz 12.288
19.200 MHz 11.2896
24 MHz
24 MHz
26 MHz
26 MHz
27 MHz
12.288
11.2896
12.288
11.2896
12.288
2 7.5264 1 90.3168 8 007h 1 0149h 0271h
2 7.5618 1 98.304 8 007h 1 0391h 0659h
2 6.9474 1 90.3168 8 006h 1 1E12h 1FBDh
2 7.2818 1 98.304 8 007h 1 013Dh 0465h
27 MHz 11.2896 2 6.6901 1 90.3168 8 006h 1 050Eh 0753h
F
OUT
= (F
SOURCE
/ F
REF
Divider) * N.K * FRATIO / OUTDIV
The values of N and K are contained in the FLLn_N, FLLn_THETA and FLLn_LAMBDA registers as shown above.
See Table 122 and Table 123 for the coding of the FLLn_REFCLK_DIV, FLLn_FRATIO and FLLn_OUTDIV registers.
Table 125 Example FLL Settings w PP, August 2012, Rev 3.4
220
Pre-Production
WM8958
SAMPLE RATE CONVERSION
The WM8958 supports two main digital audio interfaces, AIF1 and AIF2. These interfaces are configured independently and may operate entirely asynchronously to each other. The WM8958 performs stereo full-duplex sample rate conversion between the audio interfaces, allowing digital audio to be routed between the interfaces, and allowing asynchronous audio data to be mixed together.
The Sample Rate Converters (SRCs) are configured automatically within the WM8958, and no user settings are required. The SRCs are enabled automatically when required and are disabled at other times. Synchronisation between the audio interfaces is not instantaneous when the clocking or sample rate configurations are updated; the lock status of the SRCs is signalled via the GPIO or
Interrupt circuits, as described in “General Purpose Input/Output” and “Interrupts”.
Separate clocks can be used for AIF1 and AIF2, allowing asynchronous operation on these interfaces. The digital mixing core is clocked by SYSCLK, which is linked to either AIF1CLK or
AIF2CLK, as described in “Clocking and Sample Rates”. The digital mixing core is, therefore, always synchronised to AIF1, or to AIF2, or to both interfaces at once.
SAMPLE RATE CONVERTER 1 (SRC1)
SRC1 performs sample rate conversion of digital audio data input to the WM8958. Sample Rate
Conversion is required when digital audio data is received on an audio interface that is not synchronised to the digital mixing core.
SRC1 is automatically configured on AIF1 or AIF2, depending on the selected Clocking and Sample
Rate configuration. Note that SRC1 cannot convert input data on AIF1 and AIF2 simultaneously.
Sample Rate conversion on AIF1 is only supported on TDM Timeslot 0.
The SRC1 Lock status indicates when audio data can be received on the interface channel that is not synchronised to the digital mixing core. No audio will be present on this signal path until SRC1 Lock is achieved.
SAMPLE RATE CONVERTER 2 (SRC2)
SRC2 performs sample rate conversion of digital audio data output from the WM8958. Sample Rate
Conversion is required when digital audio data is transmitted on an audio interface that is not synchronised to the digital mixing core.
SRC2 is automatically configured on AIF1 or AIF2, depending on the selected Clocking and Sample
Rate configuration. Note that SRC2 cannot convert output data on AIF1 and AIF2 simultaneously.
Sample Rate conversion on AIF1 is only supported on TDM Timeslot 0.
The SRC2 Lock status indicates when audio data can be transmitted on the interface channel that is not synchronised to the digital mixing core. No audio will be present on this signal path until SRC2
Lock is achieved.
SAMPLE RATE CONVERTER RESTRICTIONS
The following restrictions apply to the configuration of the WM8958 Sample Rate Converters.
No SRC on AIF1 Timeslot 1. Sample Rate Conversion on audio interface AIF1 is not supported on the TDM Timeslot 1. Therefore, it is not possible to route digital audio between AIF1 Timeslot 1 and
AIF2, or to mix together audio from these interface paths. Note that this only applies when the SRC is applied to AIF1.
Maximum of three sample rates in the system. The audio sample rate of AIF1 input and AIF1 output may be different to each other. The audio sample rate of AIF2 input and AIF2 output may be different to each other. However, it is not possible to have four different sample rates operating simultaneously, as this would require sample rate conversion in too many paths. A maximum of three different sample rates can be supported in the system. w PP, August 2012, Rev 3.4
221
WM8958
Pre-Production
No SRC capability when using 88.2kHz or 96kHz AIF input (DAC playback). If either interface is configured for 88.2kHz or 96kHz sample rate, then the digital mixing core must also be configured for this sample rate. Sample Rate Conversion cannot be supported in this mode, therefore AIF output is not supported at any sample rate under these conditions.
Restricted Sample Rate options when AIF1 and AIF2 are synchronised. When the same clock source is used for AIF1CLK and AIF2CLK, and mixed sample rates are selected on both interfaces, then the DAC sample rate of one interface must be the same as the ADC sample rate of the other.
If AIF1CLK_SRC = AIF2CLK_SRC
Then the DAC sample rate of one interface must be the same as the ADC sample rate of the other.
Restricted Sample Rate options when AIF1 and AIF2 are not synchronised. When a different clock source is used for AIF1CLK and AIF2CLK, then the AIF to which the SYSCLK is synchronised cannot be mixed sample rates.
And SYSCLK_SRC =0
Then AIF1DAC_DIV and AIF1ADC_DIV must be set to 000 w
And SYSCLK_SRC =1
Then AIF2DAC_DIV and AIF2ADC_DIV must be set to 000
SAMPLE RATE CONVERTER CONFIGURATION ERROR INDICATION
The WM8958 verifies the register settings relating to Clocking, Sample Rates and Sample Rate
Conversion. If an invalid configuration is attempted, then the SR_ERROR register will indicate the error by showing a non-zero value. This read-only field may be checked to confirm that the WM8958 can support the selected Clocking and Sample Rate settings.
REGISTER
ADDRESS
R530 (0212h)
Rate Status
BIT LABEL DEFAULT
3:0 SR_ERROR
[3:0]
0000
DESCRIPTION
Sample Rate Configuration status
Indicates an error with the register settings related to sample rate configuration
0000 = No errors
0001 = Invalid sample rate
0010 = Invalid AIF divide
0011 = ADC and DAC divides both set in an interface
0100 = Invalid combination of AIF divides and sample-rate
0101 = Invalid set of enables for 96kHz mode
0110 = Invalid SYSCLK rate (derived from
AIF1CLK_RATE or AIF2CLK_RATE)
0111 = Mixed ADC and DAC rates in
SYSCLK AIF when AIFs are asynchronous
1000 = Invalid combination of sample rates when both AIFs are from the same clock source
1001 = Invalid combination of mixed
PP, August 2012, Rev 3.4
222
Pre-Production
WM8958
REGISTER
ADDRESS
BIT LABEL DEFAULT DESCRIPTION
ADC/DAC AIFs when both from the same clock source
1010 = AIF1DAC2 (Timeslot 1) ports enabled when SRCs connected to AIF1
Table 126 Sample Rate Converter Configuration Status w PP, August 2012, Rev 3.4
223
WM8958
Pre-Production
CONTROL INTERFACE
The WM8958 is controlled by writing to registers through a 2-wire serial control interface. Readback is available for all registers, including Chip ID and power management status.
Note that the Control Interface function can be supported with or without system clocking. Where possible, the register map access is synchronised with SYSCLK in order to ensure predictable operation of cross-domain functions. See “Clocking and Sample Rates” for further details of Control
Interface clocking.
The WM8958 is a slave device on the control interface; SCLK is a clock input, while SDAT is a bidirectional data pin. To allow arbitration of multiple slaves (and/or multiple masters) on the same interface, the WM8958 transmits logic 1 by tri-stating the SDAT pin, rather than pulling it high. An external pull-up resistor is required to pull the SDAT line high so that the logic 1 can be recognised by the master.
In order to allow many devices to share a single 2-wire control bus, every device on the bus has a unique 8-bit device ID (this is not the same as the address of each register in the WM8958). The device ID is selectable on the WM8958, using the ADDR pin as shown in Table 127. The LSB of the
Device ID is the Read/Write bit; this bit is set to logic 1 for “Read” and logic 0 for “Write”.
An internal pull-down resistor is enabled by default on the ADDR pin; this can be configured using the
ADDR_PD register bit described in Table 129.
Low
High
0011 0100 (34h)
0011 0110 (36h)
Table 127 Control Interface Device ID Selection
The WM8958 operates as a slave device only. The controller indicates the start of data transfer with a high to low transition on SDAT while SCLK remains high. This indicates that a device ID, register address and data will follow. The WM8958 responds to the start condition and shifts in the next eight bits on SDAT (8-bit device ID, including Read/Write bit, MSB first). If the device ID received matches the device ID of the WM8958, then the WM8958 responds by pulling SDAT low on the next clock pulse (ACK). If the device ID is not recognised or the R/W bit is set incorrectly, the WM8958 returns to the idle condition and waits for a new start condition and valid address.
If the device ID matches the device ID of the WM8958, the data transfer continues as described below. The controller indicates the end of data transfer with a low to high transition on SDAT while
SCLK remains high. After receiving a complete address and data sequence the WM8958 returns to the idle state and waits for another start condition. If a start or stop condition is detected out of sequence at any point during data transfer (i.e. SDAT changes while SCLK is high), the device returns to the idle condition.
The WM8958 supports the following read and write operations:
Single
Single
Multiple write using auto-increment
Multiple read using auto-increment
The sequence of signals associated with a single register write operation is illustrated in Figure 73.
Figure 73 Control Interface 2-wire (I2C) Register Write w PP, August 2012, Rev 3.4
224
Pre-Production
WM8958
The sequence of signals associated with a single register read operation is illustrated in Figure 74.
Figure 74 Control Interface 2-wire (I2C) Register Read
The Control Interface also supports other register operations, as listed above. The interface protocol for these operations is summarised below. The terminology used in the following figures is detailed in
Table 128.
Note that, for multiple write and multiple read operations, the auto-increment option must be enabled.
This feature is enabled by default, as noted in Table 129.
TERMINOLOGY DESCRIPTION
A Acknowledge (SDA Low)
Not Acknowledge (SDA High)
[White field]
[Grey field]
1 = Read
Data flow from bus master to WM8958
Data flow from WM8958 to bus master
Table 128 Control Interface Terminology
Figure 75 Single Register Write to Specified Address
Figure 76 Single Register Read from Specified Address w PP, August 2012, Rev 3.4
225
WM8958
Pre-Production
Figure 77 Multiple Register Write to Specified Address using Auto-increment
Figure 78 Multiple Register Read from Specified Address using Auto-increment
Figure 79 Multiple Register Read from Last Address using Auto-increment
Multiple Write and Multiple Read operations enable the host processor to access sequential blocks of the data in the WM8958 register map faster than is possible with single register operations. The autoincrement option is enabled when the AUTO_INC register bit is set. This bit is defined in Table 129.
Auto-increment is enabled by default.
DESCRIPTION REGISTER
ADDRESS
R257 (0101h)
Control Interface
2 AUTO_INC 1
R1825 (0721h)
Pull Control (2)
8 ADDR_PD 1
Enables address auto-increment
0 = Disabled
1 = Enabled
ADDR Pull-down enable
0 = Disabled
1 = Enabled
Table 129 Control Interface Configuration w PP, August 2012, Rev 3.4
226
Pre-Production
WM8958
CONTROL WRITE SEQUENCER
The Control Write Sequencer is a programmable unit that forms part of the WM8958 control interface logic. It provides the ability to perform a sequence of register write operations with the minimum of demands on the host processor - the sequence may be initiated by a single operation from the host processor and then left to execute independently.
Default sequences for Start-Up of each output driver and Shut-Down are provided (see “Default
Sequences” section). It is recommended that these default sequences are used unless changes become necessary.
When a sequence is initiated, the sequencer performs a series of pre-defined register writes. The host processor informs the sequencer of the start index of the required sequence within the sequencer’s memory. At each step of the sequence, the contents of the selected register fields are read from the sequencer’s memory and copied into the WM8958 control registers. This continues sequentially through the sequencer’s memory until an “End of Sequence” bit is encountered; at this point, the sequencer stops and an Interrupt status flag is asserted. For cases where the timing of the write sequence is important, a programmable delay can be set for specific steps within the sequence.
Note that the Control Write Sequencer’s internal clock is derived from the internal clock SYS_CLK which must be enabled as described in “Clocking and Sample Rates”. The clock division from
SYS_CLK is handled transparently by the WM8958 without user intervention, provided that SYS_CLK is configured as specified in “Clocking and Sample Rates”.
INITIATING A SEQUENCE
The Register fields associated with running the Control Write Sequencer are described in Table 130.
Note that the operation of the Control Write Sequencer also requires the internal clock SYS_CLK to be configured as described in “Clocking and Sample Rates”.
The Write Sequencer is enabled by setting the WSEQ_ENA bit. The start index of the required sequence must be written to the WSEQ_START_INDEX field.
The Write Sequencer stores up to 128 register write commands. These are defined in Registers
R12288 to R12799. There are 4 registers used to define each of the 128 possible commands. The value of WSEQ_START_INDEX selects the registers applicable to the first write command in the selected sequence.
Setting the WSEQ_START bit initiates the sequencer at the given start index. The Write Sequencer can be interrupted by writing a logic 1 to the WSEQ_ABORT bit.
The current status of the Write Sequencer can be read using two further register fields - when the
WSEQ_BUSY bit is asserted, this indicates that the Write Sequencer is busy. Note that, whilst the
Control Write Sequencer is running a sequence (indicated by the WSEQ_BUSY bit), normal read/write operations to the Control Registers cannot be supported. The index of the current step in the Write Sequencer can be read from the WSEQ_CURRENT_INDEX field; this is an indicator of the sequencer’s progress. On completion of a sequence, this field holds the index of the last step within the last commanded sequence.
When the Write Sequencer reaches the end of a sequence, it asserts the WSEQ_DONE_EINT flag in
Register R1841 (see Table 91). This flag can be used to generate an Interrupt Event on completion of the sequence. Note that the WSEQ_DONE_EINT flag is asserted to indicate that the WSEQ is NOT
Busy. w PP, August 2012, Rev 3.4
227
WM8958 w
Pre-Production
REGISTER
ADDRESS
R272 (0110h)
Write
Sequencer
Ctrl (1)
R273 (0111h)
Write
Sequencer
Ctrl (2)
BIT LABEL DEFAULT
15
9
8
6:0
8
6:0
WSEQ_ENA
WSEQ_ABORT
WSEQ_START
WSEQ_START_
INDEX [6:0]
WSEQ_BUSY
(read only)
WSEQ_CURRE
NT_INDEX [6:0]
(read only)
DESCRIPTION
0
0
0
Write Sequencer Enable.
0 = Disabled
1 = Enabled
Writing a 1 to this bit aborts the current sequence and returns control of the device back to the serial control interface.
Writing a 1 to this bit starts the write sequencer at the index location selected by WSEQ_START_INDEX.
The sequence continues until it reaches an “End of sequence” flag.
At the end of the sequence, this bit will be reset by the Write Sequencer.
000_0000 Sequence Start Index. This field determines the memory location of the first command in the selected sequence. There are 127 Write
Sequencer RAM addresses:
00h = WSEQ_ADDR0 (R12288)
01h = WSEQ_ADDR1 (R12292)
02h = WSEQ_ADDR2 (R12296)
….
7Fh = WSEQ_ADDR127 (R12796)
0 Sequencer Busy flag (Read Only).
0 = Sequencer idle
1 = Sequencer busy
Note: it is not possible to write to control registers via the control interface while the Sequencer is
Busy.
000_0000 Sequence Current Index. This indicates the memory location of the most recently accessed command in the write sequencer memory.
Coding is the same as
WSEQ_START_INDEX.
Table 130 Write Sequencer Control - Initiating a Sequence
PROGRAMMING A SEQUENCE
A sequence consists of write operations to data bits (or groups of bits) within the control registers.
Each write operation is defined by a block of 4 registers, which contain 6 fields as described in this section.
The block of 4 registers is the same for up to 128 steps held in the sequencer memory. Multiple sequences can be held in the memory at the same time; each sequence occupies its own range within the 128 available register blocks.
The following 6 fields are replicated 128 times - one for each of the sequencer’s 128 steps. In the following descriptions, the term ‘n’ is used to denote the step number, from 0 to 127.
WSEQ_ADDRn is a 14-bit field containing the Control Register Address in which the data should be written.
WSEQ_DATAn is an 8-bit field which contains the data to be written to the selected Control Register.
The WSEQ_DATA_WIDTHn field determines how many of these bits are written to the selected register; the most significant bits (above the number indicated by WSEQ_DATA_WIDTHn) are ignored.
PP, August 2012, Rev 3.4
228
Pre-Production
WM8958
WSEQ_DATA_STARTn is a 4-bit field which identifies the LSB position within the selected Control
Register to which the data should be written. For example, setting WSEQ_DATA_STARTn = 0100 will select bit 4 as the LSB position; in this case, 4-bit data would be written to bits 7:4.
WSEQ_DATA_WIDTHn is a 3-bit field which identifies the width of the data block to be written. This enables selected portions of a Control Register to be updated without any concern for other bits within the same register, eliminating the need for read-modify-write procedures. Values of 0 to 7 correspond to data widths of 1 to 8 respectively. For example, setting WSEQ_DATA_WIDTHn = 010 will cause a
3-bit data block to be written. Note that the maximum value of this field corresponds to an 8-bit data block; writing to register fields greater than 8 bits wide must be performed using two separate operations of the Control Write Sequencer.
WSEQ_DELAYn is a 4-bit field which controls the waiting time between the current step and the next step in the sequence i.e. the delay occurs after the write in which it was called. The total delay time per step (including execution) is defined below, giving a useful range of execution/delay times from
562 s up to 2.048s per step:
T = k × (2 WSEQ_DELAY + 8) where k = 62.5
s (under recommended operating conditions)
WSEQ_EOSn is a 1-bit field which indicates the End of Sequence. If this bit is set, then the Control
Write Sequencer will automatically stop after this step has been executed.
The register definitions for Step 0 are described in Table 131. The equivalent definitions also apply to
Step 1 through to Step 127, in the subsequent register address locations. w
REGISTER
ADDRESS
R12288
(3000h)
Write
Sequencer 0
R12289
(3001h)
Write
Sequencer 1
R12290
(3002h)
Write
Sequencer 2
R12291
(3003h)
Write
Sequencer 3
BIT LABEL DEFAULT
13:0
7:0
10:8
3:0
8
3:0
WSEQ_ADDR
0 [13:0]
WSEQ_DATA
0 [7:0]
WSEQ_DATA
_WIDTH0 [2:0]
WSEQ_DATA
_START0 [3:0]
WSEQ_EOS0
WSEQ_DELA
Y0 [3:0]
0000h
00h
000
0000
0
0000
DESCRIPTION
Control Register Address to be written to in this sequence step.
Data to be written in this sequence step.
When the data width is less than 8 bits, then one or more of the MSBs of
WSEQ_DATAn are ignored. It is recommended that unused bits be set to
0.
Width of the data block written in this sequence step.
000 = 1 bit
001 = 2 bits
010 = 3 bits
011 = 4 bits
100 = 5 bits
101 = 6 bits
110 = 7 bits
111 = 8 bits
Bit position of the LSB of the data block written in this sequence step.
0000 = Bit 0
…
1111 = Bit 15
End of Sequence flag. This bit indicates whether the Control Write Sequencer should stop after executing this step.
0 = Not end of sequence
1 = End of sequence (Stop the sequencer after this step).
Time delay after executing this step.
Total time per step (including execution)
= 62.5µs × (2 WSEQ_DELAY + 8)
Table 131 Write Sequencer Control - Programming a Sequence
PP, August 2012, Rev 3.4
229
WM8958
Pre-Production
Note that a ‘Dummy’ write can be inserted into a control sequence by commanding the sequencer to write a value of 0 to bit 0 of Register R255 (00FFh). This is effectively a write to a non-existent register location. This can be used in order to create placeholders ready for easy adaptation of a control sequence. For example, a sequence could be defined to power-up a mono signal path from
DACL to headphone, with a ‘dummy’ write included to leave space for easy modification to a stereo signal path configuration. Dummy writes can also be used in order to implement additional time delays between register writes. Dummy writes are included in both of the Headphone start-up sequences - see Table 132 and Table 133.
In summary, the Control Register to be written is set by the WSEQ_ADDRn field. The data bits that are written are determined by a combination of WSEQ_DATA_STARTn, WSEQ_DATA_WIDTHn and
WSEQ_DATAn. This is illustrated below for an example case of writing to the VMID_SEL field within
Register R1 (0001h).
In this example, the Start Position is bit 01 (WSEQ_DATA_STARTn = 0001b) and the Data width is 2 bits (WSEQ_DATA_WIDTHn = 0001b). With these settings, the Control Write Sequencer would update the Control Register R1 [2:1] with the contents of WSEQ_DATAn [1:0]. w
Figure 80 Control Write Sequencer Example
DEFAULT SEQUENCES
When the WM8958 is powered up, a number of Control Write Sequences are available through default settings in the sequencer memory locations. The pre-programmed default settings include
Start-Up and Shut-Down sequences for each of the output drivers. Note that the default sequences do not include audio signal path or gain setting configuration; this must be implemented prior to scheduling any of the default Start-Up sequences.
The entire sequencer memory may be programmed to users’ own settings at any time, as described in “Programming a Sequence”. Users’ own settings remain in memory regardless of WSEQ_ENA, and are not affected by software resets (i.e. writing to Register R0). However, any non-default sequences are lost when the device is powered down.
The following default control sequences are provided:
1. Headphone Cold Start-Up - This sequence powers up the headphone driver and charge pump. It commands the DC Servo to perform offset correction. It enables the master bias required for analogue functions. This sequence is intended for enabling the headphone output after initial power-on, when DC offset correction has not previously been run.
PP, August 2012, Rev 3.4
230
Pre-Production
WSEQ
INDEX
0 (00h)
REGISTER
ADDRESS
R57 (0039h)
WM8958
2. Headphone Warm Start-Up - This sequence is similar to the Headphone Cold Start-Up, but does not include the DC Servo operation. This sequence is intended for fast enabling of the headphone output when DC offset correction has previously been scheduled and provided the analogue gain settings have not been updated since scheduling the DC offset correction.
3. Speaker Start-Up - This sequence powers up the stereo speaker driver. It also enables the master bias required for analogue functions.
4. Earpiece Start-Up - This sequence powers up the earpiece driver. It also enables the master bias required for analogue functions. The soft-start VMID option is used in order to suppress pops when the driver is enabled. This sequence is intended for enabling the earpiece driver when the master bias has not previously been enabled.
5. Line Output Start-Up - This sequence powers up the line outputs. Active discharge of the line outputs is selected, followed by the soft-start VMID enable, followed by selection of the master bias and un-muting of the line outputs. This sequence is intended for enabling the line drivers when the master bias has not previously been enabled.
6. Speaker and Headphone Fast Shut-Down - This sequence implements a fast shutdown of the speaker and headphone drivers. It also disables the DC Servo and charge pump circuits, and disables the analogue bias circuits using the soft-start (ramp) feature. This sequence is intended as a shut-down sequence when only the speaker or headphone drivers are enabled.
7. Generic Shut-Down - This sequence shuts down all of the WM8958 output drivers, DC Servo, charge pump and analogue bias circuits. It is similar to the Fast Shut-Down sequence, with the additional control of the earpiece and line output drivers. Active discharge of the line outputs is included and all drivers are disabled as part of this sequence.
Specific details of each of these sequences is provided below.
Headphone Cold Start-Up
The Headphone Cold Start-Up sequence is initiated by writing 8100h to Register 272 (0110h). This single operation starts the Control Write Sequencer at Index Address 0 (00h) and executes the sequence defined in Table 132.
This sequence takes approximately 296ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
5 bits Bit 2 1Bh 0h 0b
1 (01h)
2 (02h)
3 (03h)
4 (04h)
5 (05h)
R1 (0001h)
R76 (004Ch)
R1 (0001h)
R96 (0060h)
R84 (0054h)
3 bits
1 bit
2 bits
5 bits
6 bits
Bit 0
Bit 15
Bit 8
Bit 1
Bit 0
03h
01h
03h
11h
33h
9h
6h
0h
0h
Ch
0b
0b
0b
0b
0b
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 11b
(delay = 0.5625ms)
BIAS_ENA = 1
VMID_SEL[1:0] = 01b
(delay = 32.5ms)
CP_ENA = 1
(delay = 4.5ms)
HPOUT1R_ENA = 1
HPOUT1L_ENA = 1
(delay = 0.5625ms)
HPOUT1R_DLY = 1
HPOUT1L_DLY = 1
(delay = 0.5625ms)
DCS_ENA_CHAN_0 = 1
DCS_ENA_CHAN_1 = 1
DCS_TRIG_STARTUP_0 = 1
DCS_TRIG_STARTUP_1 = 1
(delay = 256.5ms) w PP, August 2012, Rev 3.4
231
WM8958
WSEQ
INDEX
6 (06h)
7 (07h)
REGISTER
ADDRESS
R255
(00FFh)
R96 (0060h)
Pre-Production
WIDTH START DATA DELAY EOS DESCRIPTION
1 bit
6 bits
Bit 0
Bit 2
00h
3Bh
0h
0h
0b
1b
Dummy Write for expansion
(delay = 0.5625ms)
HPOUT1R_OUTP = 1
HPOUT1R_RMV_SHORT =1
HPOUT1_DLY = 1
HPOUT1L_OUTP = 1
HPOUT1L_RMV_SHORT = 1
(delay = 0.5625ms)
Table 132 Headphone Cold Start-Up Default Sequence
WSEQ
INDEX
8 (08h)
9 (09h)
10 (0Ah)
11 (0Bh)
12 (0Ch)
13 (0Dh)
14 (0Eh)
REGISTER
ADDRESS
R57 (0039h)
Headphone Warm Start-Up
The Headphone Warm Start-Up sequence can be initiated by writing 8108h to Register 272 (0110h).
This single operation starts the Control Write Sequencer at Index Address 8 (08h) and executes the sequence defined in Table 133.
This sequence takes approximately 40ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
5 bits Bit 2 1Bh 0h 0b
R1 (0001h)
R76 (004Ch)
R1 (0001h)
R96 (0060h)
R84 (0054h)
R255
(00FFh)
15 (0Fh) R96 (0060h)
3 bits
1 bits
2 bits
5 bits
2 bits
1 bits
6 bits
Bit 0
Bit 15
Bit 8
Bit 1
Bit 0
Bit 0
Bit 2
03h
01h
03h
11h
03h
00h
3Bh
9h
6h
0h
0h
0h
0h
0h
0b
0b
0b
0b
0b
0b
1b
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 11b
(delay = 0.5625ms)
BIAS_ENA = 1
VMID_SEL[1:0] = 01b
(delay = 32.5ms)
CP_ENA = 1
(delay = 4.5ms)
HPOUT1R_ENA = 1
HPOUT1L_ENA = 1
(delay = 0.5625ms)
HPOUT1R_DLY = 1
HPOUT1L_DLY = 1
(delay = 0.5625ms)
DCS_ENA_CHAN_0 = 1
DCS_ENA_CHAN_1 = 1
(delay = 0.5625ms)
Dummy Write for expansion
(delay = 0.5625ms)
HPOUT1R_OUTP = 1
HPOUT1R_RMV_SHORT =1
HPOUT1_DLY = 1
HPOUT1L_OUTP = 1
HPOUT1L_RMV_SHORT = 1
(delay = 0.5625ms)
Table 133 Headphone Warm Start-Up Default Sequence w PP, August 2012, Rev 3.4
232
Pre-Production
WSEQ
INDEX
16 (10h)
REGISTER
ADDRESS
R57 (39h)
WM8958
Speaker Start-Up
The Speaker Start-Up sequence can be initiated by writing 8110h to Register 272 (0110h). This single operation starts the Control Write Sequencer at Index Address 16 (10h) and executes the sequence defined in Table 134.
This sequence takes approximately 34ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
5 bits Bit 2 1Bh 0h 0b
17 (11h)
18 (12h)
R1 (01h)
R1 (01h)
3 bits
2 bits
Bit 0
Bit 12
03h
03h
9h
0h
0b
1b
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 11b
(delay = 0.5625ms)
BIAS_ENA = 1
VMID_SEL[1:0] = 01b
(delay = 32.5ms)
SPKOUTL_ENA = 1
SPKOUTR_ENA = 1
(delay = 0.5625ms)
Table 134 Speaker Start-Up Default Sequence
WSEQ
INDEX
19 (13h)
20 (14h)
21 (15h)
22 (16h)
23 (17h)
24 (18h)
REGISTER
ADDRESS
R57 (39h)
R56 (38h)
R31 (1Fh)
R1 (01h)
R1 (01h)
R57 (39h)
Earpiece Start-Up
The Earpiece Start-Up sequence can be initiated by writing 8113h to Register 272 (0110h). This single operation starts the Control Write Sequencer at Index Address 19 (13h) and executes the sequence defined in Table 135.
This sequence takes approximately 259ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
6 bits
1 bit
1 bit
1 bit
3 bits
1 bit
Bit 1
Bit 6
Bit 5
Bit 11
Bit 0
Bit 1
27h
01h
00h
01h
03h
00h
0h
0h
0h
0h
Ch
0h
0b
0b
1b
0b
0b
0b
BIAS_SRC = 1
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 10b
(delay = 0.5625ms)
HPOUT2_IN_ENA = 1
(delay = 0.5625ms)
HPOUT2_MUTE = 0
(delay = 0.5625ms)
HPOUT2_ENA = 1
(delay = 0.5625ms)
BIAS_ENA = 1
VMID_SEL[1:0] = 01b
(delay = 256.5ms)
BIAS_SRC = 0
(delay = 0.5625ms)
Table 135 Earpiece Start-Up Default Sequence w PP, August 2012, Rev 3.4
233
WM8958
WSEQ
INDEX
25 (19h)
26 (1Ah)
27 (1Bh)
28 (1Ch)
29 (1Dh)
30 (1Eh)
31 (1Fh)
32 (20h)
33 (21h)
REGISTER
ADDRESS
R56 (38h)
Pre-Production
Line Output Start-Up
The Line Output Start-Up sequence can be initiated by writing 8119h to Register 272 (0110h). This single operation starts the Control Write Sequencer at Index Address 25 (19h) and executes the sequence defined in Table 136.
This sequence takes approximately 517ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
2 bits Bit 4 03h 0h 0b
R57 (39h)
R56 (38h)
R3 (03h)
R56 (38h)
R1 (01h)
R57 (39h)
R30 (1Eh)
R30 (1Eh)
6 bits
1 bit
4 bits
2 bits
3 bits
1 bit
2 bits
2 bits
Bit 1
Bit 7
Bit 10
Bit 4
Bit 0
Bit 1
Bit 5
Bit 1
27h
01h
0Fh
00h
03h
00h
00h
00h
0h
0h
0h
0h
Dh
0h
0h
0h
0b
0b
0b
0b
0b
0b
0b
1b
LINEOUT2_DISCH = 1
LINEOUT1_DISCH = 1
(delay = 0.5625ms)
BIAS_SRC = 1
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 10b
(delay = 0.5625ms)
LINEOUT_VMID_BUF_ENA = 1
(delay = 0.5625ms)
LINEOUT2P_ENA = 1
LINEOUT2N_ENA = 1
LINEOUT1P_ENA = 1
LINEOUT1N_ENA = 1
(delay = 0.5625ms)
LINEOUT2_DISCH = 0
LINEOUT1_DISCH = 0
(delay = 0.5625ms)
BIAS_ENA = 1
VMID_SEL = 01b
(delay = 512.5ms)
BIAS_SRC = 0
(delay = 0.5625ms)
LINEOUT1P_MUTE = 0
LINEOUT1N_MUTE = 0
(delay = 0.5625ms)
LINEOUT2P_MUTE = 0
LINEOUT2N_MUTE = 0
(delay = 0.5625ms)
Table 136 Line Output Start-Up Default Sequence w PP, August 2012, Rev 3.4
234
Pre-Production
WSEQ
INDEX
34 (22h)
REGISTER
ADDRESS
R96 (60h)
WM8958
Speaker and Headphone Fast Shut-Down
The Speaker and Headphone Fast Shut-Down sequence can be initiated by writing 8122h to Register
272 (0110h). This single operation starts the Control Write Sequencer at Index Address 34 (22h) and executes the sequence defined in Table 137.
This sequence takes approximately 37ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
7 bits Bit 1 00h 0h 0b
35 (23h)
36 (24h)
37 (25h)
38 (26h)
39 (27h)
40 (28h)
41 (29h)
R84 (54h)
R1 (01h)
R76 (4Ch)
R1 (01h)
R57 (39h)
R1 (01h)
R57 (39h)
2 bits
2 bits
1 bit
2 bits
6 bits
3 bits
6 bits
Bit 0
Bit 8
Bit 15
Bit 12
Bit 1
Bit 0
Bit 1
00h
00h
00h
00h
37h
00h
00h
0h
0h
0h
0h
0h
9h
0h
0b
0b
0b
0b
0b
0b
1b
HPOUT1R_DLY = 0
HPOUT1R_OUTP = 0
HPOUT1R_RMV_SHORT = 0
HPOUT1L_DLY = 0
HPOUT1L_OUTP = 0
HPOUT1L_RMV_SHORT = 0
(delay = 0.5625ms)
DCS_ENA_CHAN_0 = 0
DCS_ENA_CHAN_1 = 0
(delay = 0.5625ms)
HPOUT1R_ENA = 0
HPOUT1L_ENA = 0
(delay = 0.5625ms)
CP_ENA = 0
(delay = 0.5625ms)
SPKOUTL_ENA = 0
SPKOUTR_ENA = 0
(delay = 0.5625ms)
BIAS_SRC = 1
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 11b
(delay = 0.5625ms)
BIAS_ENA = 0
VMID_SEL = 00b
(delay = 32.5ms)
BIAS_SRC = 0
STARTUP_BIAS_ENA = 0
VMID_BUF_ENA = 0
VMID_RAMP[1:0] = 00b
(delay = 0.5625ms)
Table 137 Speaker and Headphone Fast Shut-Down Default Sequence w PP, August 2012, Rev 3.4
235
WM8958
WSEQ
INDEX
42 (2Ah)
REGISTER
ADDRESS
R31 (1Fh)
43 (2Bh)
44 (2Ch)
45 (2Dh)
46 (2Eh)
47 (2Fh)
48 (30h)
49 (31h)
50 (32h)
51 (33h)
52 (34h)
53 (35h)
54 (36h)
55 (37h)
R30 (1Eh)
R96 (60h)
R84 (54h)
R1 (01h)
R76 (4Ch)
R1 (01h)
R57 (39h)
R1 (01h)
R1 (01h)
R56 (38h)
R55 (37h)
R56 (38h)
R3 (03h)
Pre-Production
Generic Shut-Down
The Generic Shut-Down sequence can be initiated by writing 812Ah to Register 272 (0110h). This single operation starts the Control Write Sequencer at Index Address 42 (2Ah) and executes the sequence defined in Table 138.
This sequence takes approximately 522ms to run.
WIDTH START DATA DELAY EOS DESCRIPTION
1 bit
6 bits
7 bits
2 bits
2 bits
1 bit
2 bits
6 bits
3 bits
1 bit
2 bits
1 bit
1 bit
4 bits
Bit 5
Bit 1
Bit 1
Bit 0
Bit 8
Bit 15
Bit 12
Bit 1
Bit 0
Bit 11
Bit 4
Bit 0
Bit 6
Bit 10
01h
33h
00h
00h
00h
00h
00h
17h
00h
00h
03h
01h
00h
00h
0h
0h
0h
0h
0h
0h
0h
0h
Dh
0h
0h
0h
0h
0h
0b
0b
0b
0b
0b
0b
0b
0b
0b
0b
0b
0b
0b
0b
HPOUT2_MUTE = 1
(delay = 0.5625ms)
LINEOUT2P_MUTE = 1
LINEOUT2N_MUTE = 1
LINEOUT1P_MUTE = 1
LINEOUT1N_MUTE = 1
(delay = 0.5625ms)
HPOUT1R_DLY = 0
HPOUT1R_OUTP = 0
HPOUT1R_RMV_SHORT = 0
HPOUT1L_DLY = 0
HPOUT1L_OUTP = 0
HPOUT1L_RMV_SHORT = 0
(delay = 0.5625ms)
DCS_ENA_CHAN_0 = 0
DCS_ENA_CHAN_1 = 0
(delay = 0.5625ms)
HPOUT1R_ENA = 0
HPOUT1L_ENA = 0
(delay = 0.5625ms)
CP_ENA = 0
(delay = 0.5625ms)
SPKOUTL_ENA = 0
SPKOUTR_ENA = 0
(delay = 0.5625ms)
BIAS_SRC = 1
STARTUP_BIAS_ENA = 1
VMID_BUF_ENA = 1
VMID_RAMP[1:0] = 01b
(delay = 0.5625ms)
BIAS_ENA = 0
VMID_SEL = 00b
(delay = 512.5ms)
HPOUT2_ENA = 0
(delay = 0.5625ms)
LINEOUT2_DISCH = 1
LINEOUT1_DISCH = 1
(delay = 0.5625ms)
VROI = 1
(delay = 0.5625ms)
HPOUT2_IN_ENA =0
(delay = 0.5625ms)
LINEOUT2P_ENA = 0
LINEOUT2N_ENA = 0
LINEOUT1P_ENA = 0
LINEOUT1N_ENA = 0
(delay = 0.5625ms) w PP, August 2012, Rev 3.4
236
Pre-Production
WSEQ
INDEX
56 (38h)
REGISTER
ADDRESS
R56 (38h)
WIDTH START DATA DELAY
1 bit Bit 7 00h 0h
57 (39h)
58 (3Ah)
R55 (37h)
R57 (39h)
1 bit
6 bits
Bit 0
Bit 1
00h
00h
0h
0h
WM8958
DESCRIPTION EOS
0b
0b
1b
LINEOUT_VMID_BUF_ENA = 0
(delay = 0.5625ms)
VROI = 0
(delay = 0.5625ms)
BIAS_SRC = 0
STARTUP_BIAS_ENA = 0
VMID_BUF_ENA = 0
VMID_RAMP[1:0] = 00b
(delay = 0.5625ms)
Table 138 Generic Shut-Down Default Sequence
POP SUPPRESSION CONTROL
The WM8958 incorporates a number of features, including Wolfson’s SilentSwitch™ technology, designed to suppress pops normally associated with Start-Up, Shut-Down or signal path control. To achieve maximum benefit from these features, careful attention is required to the sequence and timing of these controls. Note that, under the recommended usage conditions of the WM8958, these features will be configured by running the default Start-Up and Shut-Down sequences as described in the “Control Write Sequencer” section. In these cases, the user does not need to set these register fields directly.
The Pop Suppression controls relating to the Headphone / Line Output drivers are described in the
“Analogue Output Signal Path” section.
Additional bias controls, also pre-programmed into Control Write Sequencer, are described in the
“Reference Voltages and Master Bias” section.
DISABLED LINE OUTPUT CONTROL
The line outputs are biased to VMID in normal operation. To avoid audible pops caused by a disabled signal path dropping to AGND, the WM8958 can maintain these connections at VMID when the relevant output stage is disabled. This is achieved by connecting a buffered VMID reference to the output.
The buffered VMID reference is enabled by setting VMID_BUF_ENA. The output resistance is selectable, using the VROI register bit.
Note that, if LINEOUTn_DISCH=1 (see Table 140), then the respective output will be discharged to
AGND, and will not be connected to VMID.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R55 (0037h)
Additional
Control
R57 (0039h)
AntiPOP (2)
0
3
VROI
VMID_BUF
_ENA
0
0
Buffered VMID to Analogue Line Output
Resistance (Disabled Outputs)
0 = 20k from buffered VMID to output
1 = 500 from buffered VMID to output
VMID Buffer Enable
0 = Disabled
1 = Enabled (provided VMID_SEL > 00)
Table 139 Disabled Line Output Control w PP, August 2012, Rev 3.4
237
WM8958 w
Pre-Production
LINE OUTPUT DISCHARGE CONTROL
The line output paths can be actively discharged to AGND through internal resistors if desired. This is desirable at start-up in order to achieve a known output stage condition prior to enabling the soft-start
VMID reference voltage. This is also desirable in shut-down to prevent the external connections from being affected by the internal circuits.
The line outputs LINEOUT1P and LINEOUT1N are discharged to AGND by setting
LINEOUT1_DISCH. The line outputs LINEOUT2P and LINEOUT2N are discharged to AGND by setting LINEOUT2_DISCH.
The discharge resistance is dependent upon the respective LINEOUTn_ENA bit, and also according to the VROI bit (see Table 139). The discharge resistance is noted in the “Electrical Characteristics” section.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R56 (0038h)
AntiPOP (1)
5 LINEOUT1_DISC
H
0
4 LINEOUT2_DISC
H
0
Discharges LINEOUT1P and
LINEOUT1N outputs
0 = Not active
1 = Actively discharging LINEOUT1P and LINEOUT1N
Discharges LINEOUT2P and
LINEOUT2N outputs
0 = Not active
1 = Actively discharging LINEOUT2P and LINEOUT2N
Table 140 Line Output Discharge Control
VMID REFERENCE DISCHARGE CONTROL
The VMID reference can be actively discharged to AGND through internal resistors. This is desirable at start-up in order to achieve a known initial condition prior to enabling the soft-start VMID reference; this ensures maximum suppression of audible pops associated with start-up. VMID is discharged by setting VMID_DISCH.
REGISTER
ADDRESS
R57 (0039h)
AntiPOP (2)
BIT LABEL DEFAULT
0 VMID_DISCH 0
DESCRIPTION
Connects VMID to ground
0 = Disabled
1 = Enabled
Table 141 VMID Reference Discharge Control
INPUT VMID CLAMPS
The analogue inputs can be clamped to Vmid using the INPUTS_CLAMP bit described below. This allows pre-charging of the input AC coupling capacitors during power-up. Note that all eight inputs are clamped using the same control bit.
Note that INPUTS_CLAMP must be set to 0 when the analogue input signal paths are in use.
REGISTER
ADDRESS
R21 (15h)
Input Mixer (1)
6 INPUTS_CLAMP 0
DESCRIPTION
Input pad VMID clamp
0 = Clamp de-activated
1 = Clamp activated
Table 142 Input VMID Clamps
PP, August 2012, Rev 3.4
238
Pre-Production
WM8958
LDO REGULATORS
The WM8958 provides two integrated Low Drop-Out Regulators (LDOs). These are provided to generate the appropriate power supplies for internal circuits, simplifying and reducing the requirements for external supplies and associated components. A reference circuit powered by
AVDD2 ensures the accuracy of the LDO regulator voltage settings.
Note that the integrated LDOs are only intended for generating the AVDD1 and DCVDD supply rails for the WM8958; they are not suitable for powering any additional or external loads.
LDO1 is intended for generating AVDD1 - the primary analogue power domain of the WM8958. LDO1 is powered by LDO1VDD and is enabled when a logic ‘1’ is applied to the LDO1ENA pin. The logic level is determined with respect to the DBVDD1 voltage domain. The LDO1 start-up time is dependent on the external AVDD1 and VREFC capacitors; the start-up time is noted in the “Electrical
Characteristics” section for the recommended external component conditions.
When LDO1 is enabled, the output voltage is controlled by the LDO1_VSEL register field. Note that the LDO1 voltage difference LDO1VDD - AVDD1 must be higher than the LDO1 Drop-Out voltage
(see “Electrical Characteristics”).
When LDO1 is disabled (by applying a logic ‘0’ to the LDO1ENA pin), the output can be left floating or can be actively discharged, depending on the LDO1_DISCH control bit.
It is possible to supply AVDD1 from an external supply. If AVDD1 is supplied externally, then LDO1 should be disabled, and the LDO1 output left floating (LDO1DISCH = 0). Note that the LDO1VDD voltage must be greater than or equal to AVDD1; this ensures that there is no leakage path through the LDO for the external supply.
Note that the WM8958 can operate with AVDD1 tied to 0V; power consumption may be reduced, but the analogue audio functions will not be supported.
LDO2 is intended for generating the DCVDD power domain which supplies the digital core functions on the WM8958. LDO2 is powered by DBVDD1 and is enabled when a logic ‘1’ is applied to the
LDO2ENA pin. The logic level is determined with respect to the DBVDD1 voltage domain. The LDO2 start-up time is dependent on the external DCVDD and VREFC capacitors; the start-up time is noted in the “Electrical Characteristics” section for the recommended external component conditions.
When LDO2 is enabled, the output voltage is controlled by the LDO2_VSEL register field.
When LDO2 is disabled (by applying a logic ‘0’ to the LDO2ENA pin), the output can be left floating or can be actively discharged, depending on the LDO2_DISCH control bit.
It is possible to supply DCVDD from an external supply. If DCVDD is supplied externally, the
LDO2ENA and LDO2DISCH bits should be set to 0. Note that the DBVDD1 voltage must be greater than or equal to DCVDD; this ensures that there is no leakage path through the LDO for the external supply.
An internal pull-down resistor is enabled by default on the LDO1ENA and LDO2ENA pins. These pulldown resistors can be configured using the register bits described in Table 143.
Decoupling capacitors should be connected to the voltage reference pin, VREFC, and also to the
LDO outputs, AVDD1 and DCVDD. See “Applications Information” for further details.
The LDO Regulator connections and controls are illustrated in Figure 81. The register controls are defined in Table 143. w PP, August 2012, Rev 3.4
239
WM8958
VREFC
Pre-Production
AVDD2
Voltage
Reference
LDO1_VSEL[2:0]
LDO1_DISCH
LDO1
Analogue Supply
LDO2_VSEL[1:0]
LDO2_DISCH
Digital Core Supply
LDO2 w
LDO1VDD LDO1ENA AVDD1
Figure 81 LDO Regulators
REGISTER
ADDRESS
R59 (003Bh)
LDO 1
3:1 LDO1_VSEL [2:0]
R60
(003Ch)
LDO 2
R1825
(0721h)
Pull Control
(2)
0
2:1
0
6
4
LDO1_DISCH
LDO2_VSEL [1:0]
LDO2_DISCH
LDO2ENA_PD
LDO1ENA_PD
DBVDD1 LDO2ENA DCVDD
110
1
10
1
1
1
DESCRIPTION
LDO1 Output Voltage Select
2.4V to 3.1V in 100mV steps
000 = 2.4V
001 = 2.5V
010 = 2.6V
011 = 2.7V
100 = 2.8V
101 = 2.9V
110 = 3.0V
111 = 3.1V
LDO1 Discharge Select
0 = LDO1 floating when disabled
1 = LDO1 discharged when disabled
LDO2 Output Voltage Select
1.1V to 1.3V in 100mV steps
00 = Reserved
01 = 1.1V
10 = 1.2V
11 = 1.3V
LDO2 Discharge Select
0 = LDO2 floating when disabled
1 = LDO2 discharged when disabled
LDO2ENA Pull-down enable
0 = Disabled
1 = Enabled
LDO1ENA Pull-down enable
0 = Disabled
1 = Enabled
Table 143 LDO Regulator Control
PP, August 2012, Rev 3.4
240
Pre-Production
WM8958
REFERENCE VOLTAGES AND MASTER BIAS
This section describes the analogue reference voltage and bias current controls. It also describes the
VMID soft-start circuit for pop suppressed start-up and shut-down.
The analogue circuits in the WM8958 require a mid-rail analogue reference voltage, VMID. This reference is generated from AVDD1 via a programmable resistor chain. Together with the external
VMID decoupling capacitor, the programmable resistor chain determines the charging characteristic on VMID. This is controlled by VMID_SEL[1:0], and can be used to optimise the reference for normal operation or low power standby as described in Table 144.
A buffered mid-rail reference voltage is provided. This is required for the single-ended configuration of the Input PGAs, and also for direct signal paths from the input pins to the Input Mixers, Output Mixers or Speaker Mixers. These requirements are noted in the relevant “Analogue Input Signal Path” and
“Analogue Output Signal Path” sections. The buffered mid-rail reference is enabled by setting the
VMID_BUF_ENA register bit.
The analogue circuits in the WM8958 require a bias current. The normal bias current is enabled by setting BIAS_ENA. Note that the normal bias current source requires VMID to be enabled also.
REGISTER
ADDRESS
R1 (0001h)
Power
Management
(1)
R57 (0039h)
AntiPOP (2)
2:1
0
3
VMID_SEL
[1:0]
BIAS_ENA
VMID_BUF_
ENA
00
0
0
DESCRIPTION
VMID Divider Enable and Select
00 = VMID disabled (for OFF mode)
01 = 2 x 40k divider (for normal operation)
10 = 2 x 240k divider (for low power standby)
11 = Reserved
Enables the Normal bias current generator (for all analogue functions)
0 = Disabled
1 = Enabled
VMID Buffer Enable
0 = Disabled
1 = Enabled (provided VMID_SEL > 00)
Table 144 Reference Voltages and Master Bias Enable
A pop-suppressed start-up requires VMID to be enabled smoothly, without the step change normally associated with the initial stage of the VMID capacitor charging. A pop-suppressed start-up also requires the analogue bias current to be enabled throughout the signal path prior to the VMID reference voltage being applied. The WM8958 incorporates pop-suppression circuits which address these requirements.
An alternate bias current source (Start-Up Bias) is provided for pop-free start-up; this is enabled by the STARTUP_BIAS_ENA register bit. The start-up bias is selected (in place of the normal bias) using the BIAS_SRC bit. It is recommended that the start-up bias is used during start-up, before switching back to the higher quality, normal bias.
A soft-start circuit is provided in order to control the switch-on of the VMID reference. The soft-start control circuit offers two slew rates for enabling the VMID reference; these are selected and enabled by VMID_RAMP. When the soft-start circuit is enabled prior to enabling VMID_SEL, the reference voltage rises smoothly, without the step change that would otherwise occur. It is recommended that the soft-start circuit and the output signal path be enabled before VMID is enabled by VMID_SEL.
A soft shut-down is provided, using the soft-start control circuit and the start-up bias current generator. The soft shut-down of VMID is achieved by setting VMID_RAMP, STARTUP_BIAS_ENA and BIAS_SRC to select the start-up bias current and soft-start circuit prior to setting VMID_SEL=00.
Note that, if the VMID_RAMP function is enabled for soft start-up or soft shut-down then, after setting
VMID_SEL = 00 to disable VMID, the soft-start circuit must be reset before re-enabling VMID. The soft-start circuit is reset by setting VMID_RAMP = 00. After resetting the soft-start circuit, the
VMID_RAMP register may be updated to the required setting for the next VMID transition.
The VMID soft-start register controls are defined in Table 145. w PP, August 2012, Rev 3.4
241
WM8958
REGISTER
ADDRESS
R57 (0039h)
AntiPOP (2)
BIT LABEL DEFAULT
6:5
2
1
Pre-Production
VMID_RAMP [1:0]
STARTUP_BIAS_
ENA
BIAS_SRC
10
0
1
DESCRIPTION
VMID soft start enable / slew rate control
00 = Normal slow start
01 = Normal fast start
10 = Soft slow start
11 = Soft fast start
If VMID_RAMP = 1X is selected for
VMID start-up or shut-down, then the soft-start circuit must be reset by setting VMID_RAMP=00 after VMID is disabled, before VMID is re-enabled.
VMID is disabled / enabled using the
VMID_SEL register.
Enables the Start-Up bias current generator
0 = Disabled
1 = Enabled
Selects the bias current source
0 = Normal bias
1 = Start-Up bias
Table 145 Soft Start Control w PP, August 2012, Rev 3.4
242
Pre-Production
WM8958
POWER MANAGEMENT
The WM8958 has control registers that allow users to select which functions are active. For minimum power consumption, unused functions should be disabled. To minimise pop or click noise, it is important to enable or disable functions in the correct order. See “Control Write Sequencer” for details of recommended control sequences.
REGISTER
ADDRESS
R1 (0001h)
Power
Management
(1)
R2 (0002h)
Power
Management
(2)
13 SPKOUTR_ENA
12 SPKOUTL_ENA
11 HPOUT2_ENA
9
8
5
4
2:1 VMID_SEL
[1:0]
0
HPOUT1L_ENA
HPOUT1R_ENA
MICB2_ENA
MICB1_ENA
BIAS_ENA
14 TSHUT_ENA
13 TSHUT_OPDIS
11 OPCLK_ENA
9 MIXINL_ENA
0
0
0
0
0
0
0
00
0
1
1
0
0
DESCRIPTION
SPKMIXR Mixer, SPKRVOL PGA and
SPKOUTR Output Enable
0 = Disabled
1 = Enabled
SPKMIXL Mixer, SPKLVOL PGA and
SPKOUTL Output Enable
0 = Disabled
1 = Enabled
HPOUT2 and HPOUT2MIX Enable
0 = Disabled
1 = Enabled
Enables HPOUT1L input stage
0 = Disabled
1 = Enabled
Enables HPOUT1R input stage
0 = Disabled
1 = Enabled
Microphone Bias 2 Enable
0 = Disabled
1 = Enabled
Microphone Bias 1 Enable
0 = Disabled
1 = Enabled
VMID Divider Enable and Select
00 = VMID disabled (for OFF mode)
01 = 2 x 40k divider (Normal mode)
10 = 2 x 240k divider (Standby mode)
11 = Reserved
Enables the Normal bias current generator (for all analogue functions)
0 = Disabled
1 = Enabled
Thermal Sensor Enable
0 = Disabled
1 = Enabled
Thermal Shutdown Control
(Causes audio outputs to be disabled if an over-temperature occurs. The thermal sensor must also be enabled.)
0 = Disabled
1 = Enabled
GPIO Clock Output (OPCLK) Enable
0 = Disabled
1 = Enabled
Left Input Mixer Enable
(Enables MIXINL and RXVOICE input to
MIXINL)
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
w
243
WM8958 w
REGISTER
ADDRESS
R3 (0003h)
Power
Management
(3)
Pre-Production
DESCRIPTION
8
7
6
5
4
13
12
11
10
9
8
7
6
5
MIXINR_ENA
IN2L_ENA
IN1L_ENA
IN2R_ENA
IN1R_ENA
LINEOUT1N_ENA
LINEOUT1P_ENA
LINEOUT2N_ENA
LINEOUT2P_ENA
SPKRVOL_ENA
SPKLVOL_ENA
MIXOUTLVOL_E
NA
MIXOUTRVOL_E
NA
MIXOUTL_ENA
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Right Input Mixer Enable
(Enables MIXINR and RXVOICE input to
MIXINR)
0 = Disabled
1 = Enabled
IN2L Input PGA Enable
0 = Disabled
1 = Enabled
IN1L Input PGA Enable
0 = Disabled
1 = Enabled
IN2R Input PGA Enable
0 = Disabled
1 = Enabled
IN1R Input PGA Enable
0 = Disabled
1 = Enabled
LINEOUT1N Line Out and
LINEOUT1NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT1P Line Out and
LINEOUT1PMIX Enable
0 = Disabled
1 = Enabled
LINEOUT2N Line Out and
LINEOUT2NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT2P Line Out and
LINEOUT2PMIX Enable
0 = Disabled
1 = Enabled
SPKMIXR Mixer and SPKRVOL PGA
Enable
0 = Disabled
1 = Enabled
Note that SPKMIXR and SPKRVOL are also enabled when SPKOUTR_ENA is set.
SPKMIXL Mixer and SPKLVOL PGA
Enable
0 = Disabled
1 = Enabled
Note that SPKMIXL and SPKLVOL are also enabled when SPKOUTL_ENA is set.
MIXOUTL Left Volume Control Enable
0 = Disabled
1 = Enabled
MIXOUTR Right Volume Control Enable
0 = Disabled
1 = Enabled
MIXOUTL Left Output Mixer Enable
0 = Disabled
PP, August 2012, Rev 3.4
244
Pre-Production w
REGISTER
ADDRESS
4 MIXOUTR_ENA
R4 (0004h)
Power
Management
(4)
13 AIF2ADCL_ENA
12 AIF2ADCR_ENA
11 AIF1ADC2L_ENA
9 AIF1ADC1L_ENA
8 AIF1ADC1L_ENA
5 DMIC2L_ENA
4 DMIC2R_ENA
3 DMIC1L_ENA
2 DMIC1R_ENA
1 ADCL_ENA
0 ADCR_ENA
R5 (0005h)
Power
Management
13 AIF2DACL_ENA
0
0
0
0
10 AIF1ADC2R_ENA 0
0
0
0
0
0
0
0
0
0
WM8958
DESCRIPTION
1 = Enabled
MIXOUTR Right Output Mixer Enable
0 = Disabled
1 = Enabled
Enable AIF2ADC (Left) output path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 or AIF3 output of the AIF2ADC (Left) signal.
Enable AIF2ADC (Right) output path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 or AIF3 output of the AIF2ADC (Left) signal.
Enable AIF1ADC2 (Left) output path
(AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1ADC2 (Right) output path
(AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1ADC1 (Left) output path
(AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1ADC1 (Right) output path
(AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Digital microphone DMICDAT2 Left channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT2 Right channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT1 Left channel enable
0 = Disabled
1 = Enabled
Digital microphone DMICDAT1 Right channel enable
0 = Disabled
1 = Enabled
Left ADC Enable
0 = ADC disabled
1 = ADC enabled
Right ADC Enable
0 = ADC disabled
1 = ADC enabled
Enable AIF2DAC (Left) input path
0 = Disabled
1 = Enabled
PP, August 2012, Rev 3.4
245
WM8958 w
REGISTER
ADDRESS
(5)
R76 (004Ch)
Charge Pump
(1)
R84 (0054h)
DC Servo (1)
R272 (0110h)
Write
Sequencer
Ctrl (1)
R512 (0200h)
AIF 1 Clocking
(1)
R516 (0204h)
AIF 2 Clocking
(1)
R520 (0208h)
Clocking (1)
Pre-Production
DESCRIPTION
12 AIF2DACR_ENA 0
11
10
9
8
3
2
1
0
15
1
0
8
0
0
4
AIF1DAC2L_ENA
AIF1DAC2R_ENA
AIF1DAC1L_ENA
AIF1DAC1R_ENA
DAC2L_ENA
DAC2R_ENA
DAC1L_ENA
DAC1R_ENA
CP_ENA
DCS_ENA_CHAN
_1
DCS_ENA_CHAN
_0
WSEQ_ENA
AIF1CLK_ENA
AIF2CLK_ENA
TOCLK_ENA
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Enable AIF2DAC (Right) input path
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Left) input path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Right) input path
(AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1DAC1 (Left) input path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1DAC1 (Right) input path
(AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Left DAC2 Enable
0 = DAC disabled
1 = DAC enabled
Right DAC2 Enable
0 = DAC disabled
1 = DAC enabled
Left DAC1 Enable
0 = DAC disabled
1 = DAC enabled
Right DAC1 Enable
0 = DAC disabled
1 = DAC enabled
Enable charge-pump digits
0 = Disable
1 = Enable
DC Servo enable for HPOUT1R
0 = Disabled
1 = Enabled
DC Servo enable for HPOUT1L
0 = Disabled
1 = Enabled
Write Sequencer Enable.
0 = Disabled
1 = Enabled
AIF1CLK Enable
0 = Disabled
1 = Enabled
AIF2CLK Enable
0 = Disabled
1 = Enabled
Slow Clock (TOCLK) Enable
0 = Disabled
1 = Enabled
This clock is required for zero-cross timeout.
PP, August 2012, Rev 3.4
246
Pre-Production
REGISTER
ADDRESS
3 AIF1DSPCLK_EN
A
2 AIF2DSPCLK_EN
A
1 SYSDSPCLK_EN
A
R544 (0220h)
FLL1 Control
(1)
0 FLL1_ENA
R576 (0240h)
FLL2 Control
(1)
0 FLL2_ENA
0
0
0
0
0
WM8958
DESCRIPTION
AIF1 Processing Clock Enable
0 = Disabled
1 = Enabled
AIF2 Processing Clock Enable
0 = Disabled
1 = Enabled
Digital Mixing Processor Clock Enable
0 = Disabled
1 = Enabled
FLL1 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the
FLL1 enable sequence, ie. after the other
FLL registers have been configured.
FLL2 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the
FLL2 enable sequence, ie. after the other
FLL registers have been configured.
Table 146 Power Management w PP, August 2012, Rev 3.4
247
WM8958
Pre-Production
THERMAL SHUTDOWN
The WM8958 incorporates a temperature sensor which detects when the device temperature is within normal limits or if the device is approaching a hazardous temperature condition. The temperature sensor can be configured to automatically disable the audio outputs of the WM8958 in response to an overtemperature condition (approximately 150ºC).
The temperature status can be output directly on a GPIO pin, as described in the “General Purpose
Input/Output” section. The temperature sensor can also be used to generate Interrupt events, as described in the “Interrupts” section. The GPIO and Interrupt functions can be used to indicate either a Warning Temperature event or the Shutdown Temperature event.
The temperature sensor is enabled by setting the TSHUT_ENA register bit. When the TSHUT_OPDIS is also set, then a device over-temperature condition will cause the speaker outputs (SPKOUTL and
SPKOUTR) of the WM8958 to be disabled; this response is likely to prevent any damage to the device attributable to the large currents of the output drivers.
Note that, to prevent pops and clicks, TSHUT_ENA and TSHUT_OPDIS should only be updated whilst the speaker and headphone outputs are disabled.
BIT LABEL DEFAULT DESCRIPTION REGISTER
ADDRESS
R2 (0002h)
Power
Management
(2)
14
13
TSHUT_ENA
TSHUT_OPDIS
1
1
Thermal sensor enable
0 = Disabled
1 = Enabled
Thermal shutdown control
(Causes audio outputs to be disabled if an overtemperature occurs. The thermal sensor must also be enabled.)
0 = Disabled
1 = Enabled
Table 147 Thermal Shutdown w PP, August 2012, Rev 3.4
248
Pre-Production
WM8958
POWER ON RESET
The WM8958 includes a Power-On Reset (POR) circuit, which is used to reset the digital logic into a default state after power up. The POR circuit derives its output from AVDD2 and DCVDD. The internal
The specific behaviour of the circuit will vary, depending on relative timing of the supply voltages.
Typical scenarios are illustrated in Figure 82 and Figure 83.
Figure 82 Power On Reset Timing – AVDD2 enabled/disabled first
Figure 83 Power On Reset Timing - DCVDD enabled/disabled first
¯¯¯ signal is undefined until AVDD2 has exceeded the minimum threshold, V pora
. Once this writes to the control interface are ignored. Once AVDD2 and DCVDD have reached their respective control interface may take place.
Note that a power-on reset period, T
POR
, applies after AVDD2 and DCVDD have reached their respective power on thresholds. This specification is guaranteed by design rather than test. w power-down thresholds.
PP, August 2012, Rev 3.4
249
WM8958 w
Pre-Production
Typical Power-On Reset parameters for the WM8958 are defined in Table 148.
V pora_on
V pora_off
V pord_on
V pord_off
T
POR
Power-On threshold (AVDD2)
Power-Off threshold (AVDD2)
Power-On threshold (DCVDD)
Power-Off threshold (DCVDD)
Minimum Power-On Reset period
Table 148 Typical Power-On Reset Parameters
1.15 V
1.14 V
0.56 V
0.55 V
100 ns
Table 149 describes the status of the WM8958 digital I/O pins when the Power On Reset has completed, prior to any register writes. The same conditions apply on completion of a Software Reset
(described in the “Software Reset and Device ID” section).
TYPE RESET STATUS PIN NO NAME
DBVDD1 power domain
A4
C3
SPKMODE
LDO1ENA
D3
D5
E3
G2
ADDR
LDO2ENA
SCLK
SDA
E1
D2
F2
D4
H1
G1
MCLK1
MCLK2
BCLK1
LRCLK1
GPIO1/ADCLRCLK1
DACDAT1
F1 ADCDAT1
DBVDD2 power domain
G3 BCLK2
H2
H3
E4
F4
LRCLK2
GPIO6/ADCLRCLK2
DACDAT2
ADCDAT2
DBVDD3 power domain
E5 GPIO11/BCLK3
F5
G4
H4
GPIO10/LRCLK3
GPIO8/DACDAT3
GPIO9/ADCDAT3
Digital Input
Digital Input
Digital Input
Digital Input
Digital Input
Digital Input/Output
Digital Input
Digital Input
Digital Input/Output
Digital Input/Output
Digital Input/Output
Digital Input
Digital Output
Digital Input/Output
Digital Input/Output
Digital Input/Output
Digital Input
Digital Output
Digital Input/Output
Digital Input/Output
Digital Input/Output
Digital Input/Output
Pull-up to DBVDD1
Pull-down to DGND
Pull-down to DGND
Pull-down to DGND
Digital input
Digital input
Digital input
Pull-down to DGND
Digital input
Digital input
Digital input
Digital input
Digital output
Digital input,
Pull-down to DGND
Digital input,
Pull-down to DGND
Digital input,
Pull-down to DGND
Pull-down to DGND
Digital output
Digital input,
Pull-down to DGND
Digital input,
Pull-down to DGND
Digital input,
Pull-down to DGND
Digital input,
Pull-down to DGND
MICBIAS1 power domain
D6
A8
B9
DMICCLK
IN2RN/DMICDAT2
IN2LN/DMICDAT1
Digital Output
Analogue Input/Digital Input
Analogue Input/Digital Input
Table 149 WM8958 Digital I/O Status in Reset
Digital output
Analogue input
Analogue input
Note that the dual function IN2LN/DMICDAT1 and IN2RN/DMICDAT2 pins default to IN2LN or IN2RN
(analogue input) after Power On Reset is completed. The IN2LN and IN2RN functions are referenced to the AVDD1 power domain.
PP, August 2012, Rev 3.4
250
Pre-Production
WM8958
QUICK START-UP AND SHUTDOWN
The default control sequences (see “Control Write Sequencer”) contain only the register writes necessary to enable or disable specific output drivers. It is therefore necessary to configure the signal path and gain settings before commanding any of the default start-up sequences.
This section describes minimum control sequences to configure the WM8958 for DAC to Headphone playback. Note that these sequences are provided for guidance only; application software should be verified and tailored to ensure optimum performance.
Table 150 describes an example control sequence to enable DAC playback to HPOUT1L and
HPOUT1R path. This involves DAC enable, signal path configuration and mute control, together with the default “Headphone Cold Start-Up” sequence. Table 151 describes an example control sequence to disable the direct DAC to Headphone path.
REGISTER VALUE
R5 (0005h) 0003h
DESCRIPTION
Enable DAC1L and DAC1R
R45 (002Dh)
R46 (002Eh)
R272 (0110h)
0100h
0100h
8100h
Enable path from DAC1L to HPOUT1L
Enable path from DAC1R to HPOUT1R
Initiate Control Write Sequencer at Index Address 0 (00h)
(Headphone Cold Start-Up sequence)
R1056 (0420h) 0000h
Note: Delay must be inserted in the sequence to allow the
Control Write Sequencer to finish. Any control interface writes to the CODEC will be ignored while the Control Write
Sequencer is running.
Soft un-mute DAC1L and DAC1R
Table 150 DAC to Headphone Direct Start-Up Sequence
REGISTER VALUE
R1056 (0420h) 0200h
DESCRIPTION
Soft mute DAC1L and DAC1R
R272 (0110h) 812Ah Initiate Control Write Sequencer at Index Address 42 (2Ah)
(Generic Shut-Down)
R45 (002Dh)
R46 (002Eh)
R5 (0005h)
0000h
0000h
0000h
Note: Delay must be inserted in the sequence to allow the
Control Write Sequencer to finish. Any control interface writes to the CODEC will be ignored while the Control Write
Sequencer is running.
Disable path from DAC1L to HPOUT1L
Disable path from DAC1R to HPOUT1R
Disable DAC1L and DAC1R
Table 151 DAC to Headphone Direct Shut-Down Sequence
In both cases, the WSEQ_BUSY bit (in Register R272, see Table 130) will be set to 1 while the
Control Write Sequence runs. When this bit returns to 0, the remaining steps of the sequence may be executed. w PP, August 2012, Rev 3.4
251
WM8958
Pre-Production
SOFTWARE RESET AND DEVICE ID
The device ID can be read back from register R0. Writing to this register will reset the device.
The software reset causes most control registers to be reset to their default state. Note that the
Control Write Sequencer registers R12288 (3000h) through to R12799 (31FFh) are not affected by a software reset; the Control Sequences defined in these registers are retained unchanged.
The status of the WM8958 digital I/O pins following a software reset is described in Table 149.
The device revision can be read back from register R256.
REGISTER
ADDRESS
R0 (0000h)
Software
Reset
BIT LABEL DEFAULT
15:0 SW_RESET
[15:0]
8958h
DESCRIPTION
Writing to this register resets all registers to their default state. (Note - Control
Write Sequencer registers are not affected by Software Reset.)
Reading from this register will indicate device ID 8958h.
Chip revision R256
(0100h)
Chip
Revision
3:0 CHIP_REV [3:0]
Table 152 Chip Reset and ID w PP, August 2012, Rev 3.4
252
Pre-Production
WM8958
REGISTER MAP
The WM8958 control registers are listed below. Note that only the register addresses described here should be accessed; writing to other addresses may result in undefined behaviour. Register bits that are not documented should not be changed from the default values.
REG NAME
R0 (0h) Software Reset
R1 (1h)
R2 (2h)
R3 (3h)
R4 (4h)
R5 (5h)
Power
Management (1)
Power
Management (2)
Power
Management (3)
Power
Management (4)
Power
Management (5)
15 14 13 12 11 10 9 8 7
SW_RESET [15:0]
6 5 4 3 2 1 0 DEFAULT
0000h
0 TSHU
T_ENA
UTR_E
NA
SPKO
UTL_E
NA
HPOU
T2_EN
A
TSHU
T_OP
DIS
UT1N_
ENA
LINEO
UT1P_
ENA
LINEO
UT2N_
ENA
LINEO
UT2P_
ENA
SPKR
VOL_E
NA
SPKLV
OL_EN
A
MIXOU
TLVOL
_ENA
MIXOU
TRVO
L_ENA
MIXOU
TL_EN
A
MIXOU
TR_EN
A
DCL_E
NA
ACL_E
NA
0 OPCL
AIF2A
DCR_
ENA
AIF2D
ACR_
ENA
K_ENA
AIF1A
DC2L_
ENA
0 HPOU
0 MIXIN
L_ENA
MIXIN
R_EN
A
AIF1A
DC2R_
ENA
T1L_E
NA
AIF1A
DC1L_
ENA
HPOU
T1R_E
NA
AIF1A
DC1R_
ENA
AIF1D
AC2L_
ENA
AIF1D
AC2R_
ENA
AIF1D
AC1L_
ENA
AIF1D
AC1R_
ENA
IN2L_
ENA
IN1L_
ENA
_ENA
IN2R_
ENA
MICB1
_ENA
IN1R_
ENA
L_ENA
DMIC2
R_EN
A
DMIC1
L_ENA
DMIC1
R_EN
A
ADCL_
ENA
ADCR
_ENA
_ENA
DAC2
R_EN
A
[1:0]
DAC1L
_ENA
ENA
DAC1
R_EN
A
0000h
0 0 0 0 6000h
0 0 0 0 0000h
0000h
0000h
R6 (6h) Power
Management (6)
R21 (15h) Input Mixer (1)
[1:0] [1:0]
_MIXI
NR_B
OOST
IN1LP
_MIXI
NL_BO
OST
INPUT
S_CLA
MP
0 AIF3_T
RI
AIF3_ADCDAT
_SRC [1:0]
AIF2_
ADCD
AT_SR
C
AIF2_
DACD
AT_SR
C
AIF1_
DACD
AT_SR
C
0000h
0 0 0 0 0 0 0000h
U
IN1L_
MUTE
IN1L_Z
C
R24 (18h) Left Line Input 1&2
Volume
R25 (19h) Left Line Input 3&4
Volume
R26 (1Ah) Right Line Input
1&2 Volume
U
IN2L_
MUTE
IN2L_Z
C
IN1R_
MUTE
IN1R_
ZC U
R27 (1Bh) Right Line Input
3&4 Volume
R29 (1Dh) Right Output
Volume
U
IN2R_
MUTE
R28 (1Ch) Left Output Volume 0 0 0 0 0 0 0
T1_VU
HPOU
T1L_Z
C
IN2R_
ZC
HPOU
T1L_M
UTE_N
R30 (1Eh) Line Outputs
Volume
R31 (1Fh) HPOUT2 Volume
HPOUT1L_VOL [5:0] 006Dh
T1_VU
HPOU
T1R_Z
C
HPOU
T1R_M
UTE_N
HPOUT1R_VOL [5:0]
UT1N_
MUTE
LINEO
UT1P_
MUTE
LINEO
UT1_V
OL
T2_MU
TE
HPOU
T2_VO
L
0 LINEO
UT2N_
MUTE
LINEO
UT2P_
MUTE
LINEO
UT2_V
OL
006Dh
0066h
0 0 0 0 0020h
MIXOUTL_VOL [5:0] 0079h R32 (20h) Left OPGA Volume 0 0 0 0 0 0 0
T_VU
MIXOU
TL_ZC
MIXOU
TL_MU
TE_N
R33 (21h) Right OPGA
Volume T_VU
MIXOU
TR_ZC
MIXOU
TR_M
UTE_N
MIXOUTR_VOL [5:0] 0079h
R34 (22h) SPKMIXL
Attenuation B_REF
_SEL
0 DAC2L
_SPK
MIXL_
VOL
MIXIN
L_SPK
MIXL_
VOL
IN1LP
_SPK
MIXL_
VOL
MIXOU
TL_SP
KMIXL
_VOL
DAC1L
_SPK
MIXL_
VOL
SPKMIXL_VOL
[1:0]
0003h w PP, August 2012, Rev 3.4
253
WM8958
REG NAME
R35 (23h) SPKMIXR
Attenuation
R36 (24h) SPKOUT Mixers
R37 (25h) ClassD
R38 (26h) Speaker Volume
Left
R39 (27h) Speaker Volume
Right
R40 (28h) Input Mixer (2)
R41 (29h) Input Mixer (3)
R42 (2Ah) Input Mixer (4)
R43 (2Bh) Input Mixer (5)
R44 (2Ch) Input Mixer (6)
R45 (2Dh) Output Mixer (1)
R46 (2Eh) Output Mixer (2)
R47 (2Fh) Output Mixer (3)
R48 (30h) Output Mixer (4)
R49 (31h) Output Mixer (5)
R50 (32h) Output Mixer (6)
R51 (33h) HPOUT2 Mixer
R52 (34h) Line Mixer (1)
15 14 13 12
Pre-Production
11 10 9 8
UT_CL
ASSA
B
7 6 5 4
0 DAC2
R_SPK
MIXR_
VOL
MIXIN
R_SPK
MIXR_
VOL
IN1RP
_SPK
MIXR_
VOL
3
MIXOU
TR_SP
KMIXR
_VOL
2
DAC1
R_SPK
MIXR_
VOL
1
SPKMIXR_VOL
[1:0]
0 DEFAULT
0003h
P_TO_
SPKO
UTL
SPKMI
XL_TO
_SPK
OUTL
SPKMI
XR_T
O_SP
KOUT
L
IN2LR
P_TO_
SPKO
UTR
SPKMI
XL_TO
_SPK
OUTR
SPKMI
XR_T
O_SP
KOUT
R
0011h
0140h
[2:0]
SPKOUTL_VOL [5:0]
[2:0]
UT_VU
SPKO
UTL_Z
C
SPKO
UTL_M
UTE_N
SPKO
UT_VU UTR_Z
C
SPKO
UTR_
MUTE
_N
SPKOUTR_VOL [5:0]
0079h
0079h
_TO_I
N2L
O_MIX
INL
IN2L_
MIXIN
L_VOL
TO_MI
XINR
IN2R_
MIXIN
R_VOL
IN2LN
_TO_I
N2L
IN1LP
_TO_I
N1L
IN1LN
_TO_I
N1L
0 IN1L_T
O_MIX
INL
IN1L_
MIXIN
L_VOL
0 IN1R_
TO_MI
XINR
IN1R_
MIXIN
R_VOL
IN2RP
_TO_I
N2R
IN2RN
_TO_I
N2R
IN1RP
_TO_I
N1R
IN1RN
_TO_I
N1R
0000h
0 MIXOUTL_MIXINL_VOL 0000h
[2:0]
0 MIXOUTR_MIXINR_VO
L [2:0]
0000h
[2:0] [2:0]
[2:0]
[2:0]
[2:0]
L [2:0]
[2:0] [2:0]
_TO_H
POUT
1L
MIXIN
R_TO_
MIXOU
TL
MIXIN
L_TO_
MIXOU
TL
IN2RN
_TO_
MIXOU
TL
IN2LN
_TO_
MIXOU
TL
IN1R_
TO_MI
XOUT
L
IN1L_T
O_MIX
OUTL
IN2LP
_TO_
MIXOU
TL
DAC1L
_TO_
MIXOU
TL
R_TO_
HPOU
T1R
MIXIN
L_TO_
MIXOU
TR
MIXIN
R_TO_
MIXOU
TR
IN2LN
_TO_
MIXOU
TR
IN2RN
_TO_
MIXOU
TR
IN1L_T
O_MIX
OUTR
IN1R_
TO_MI
XOUT
R
IN2RP
_TO_
MIXOU
TR
DAC1
R_TO_
MIXOU
TR
[2:0]
IN1R_MIXOUTL_VOL
[2:0]
IN1L_MIXOUTL_VOL
[2:0]
0000h
0000h
0000h
[2:0]
IN1L_MIXOUTR_VOL
[2:0]
IN1R_MIXOUTR_VOL
[2:0]
MIXINR_MIXOUTL_VO
L [2:0]
MIXINL_MIXOUTL_VOL
[2:0] [2:0]
IN2LN_MIXOUTR_VOL
[2:0]
MIXINL_MIXOUTR_VO
L [2:0]
MIXINR_MIXOUTR_VO
L [2:0]
0000h
0000h
0000h
P_TO_
HPOU
T2
MIXOU
TLVOL
_TO_H
POUT
2
MIXOU
TRVO
L_TO_
HPOU
T2
TL_TO
_LINE
OUT1
N
MIXOU
TR_TO
_LINE
OUT1
N
LINEO
UT1_M
ODE
0 0 0 0000h
0 IN1R_
TO_LI
NEOU
T1P
IN1L_T
O_LIN
EOUT
1P
MIXOU
TL_TO
_LINE
OUT1
P
0000h w PP, August 2012, Rev 3.4
254
Pre-Production
WM8958
REG NAME
R53 (35h) Line Mixer (2)
R54 (36h) Speaker Mixer
R55 (37h) Additional Control
R56 (38h) AntiPOP (1)
R57 (39h) AntiPOP (2)
15 14 13 12 11 10 9 8 7 6 5
TR_TO
_LINE
OUT2
N
MIXOU
TL_TO
_LINE
OUT2
N
4
LINEO
UT2_M
ODE
3 2 1
0 IN1L_T
O_LIN
EOUT
2P
IN1R_
TO_LI
NEOU
T2P
0 DEFAULT
MIXOU
TR_TO
_LINE
OUT2
P
0000h
_TO_S
PKMIX
L
DAC2
R_TO_
SPKMI
XR
MIXIN
L_TO_
SPKMI
XL
MIXIN
R_TO_
SPKMI
XR
IN1LP
_TO_S
PKMIX
L
IN1RP
_TO_S
PKMIX
R
MIXOU
TL_TO
_SPK
MIXL
MIXOU
TR_TO
_SPK
MIXR
DAC1L
_TO_S
PKMIX
L
DAC1
R_TO_
SPKMI
XR
0000h
UT1_F
B
LINEO
UT2_F
B
UT_V
MID_B
UF_EN
A
HPOU
T2_IN_
ENA
LINEO
UT1_D
ISCH
LINEO
UT2_D
ISCH
0 0 0 0 0000h
[1:0] BUF_E
NA
START
UP_BI
AS_EN
A
BIAS_
SRC
VMID_
DISCH
0180h
R59 (3Bh) LDO 1 000Dh
DISCH
R60 (3Ch) LDO 2
R61 (3Dh) MICBIAS1
R62 (3Eh) MICBIAS2
0005h
_RATE
_RATE
MICB1
_MOD
E
MICB2
_MOD
E
[1:0] DISCH
MICB1_LVL [2:0] MICB1
_DISC
H
MICB2_LVL [2:0] MICB2
_DISC
H
0039h
0039h
0 0 1 1 1 1 1 0 0 1 0 0 1 0 1 1F25h R76 (4Ch) Charge Pump (1) CP_E
NA
R77 (4Dh) Charge Pump (2) CP_DI
SCH
R81 (51h) Class W (1)
R84 (54h) DC Servo (1)
R85 (55h) DC Servo (2)
R87 (57h) DC Servo (4)
R88 (58h) DC Servo
Readback
R96 (60h) Analogue HP (1)
R208 (D0h) Mic Detect 1
R209 (D1h) Mic Detect 2
R210 (D2h) Mic Detect 3
R256 (100h) Chip Revision
R257 (101h) Control Interface
0 1 0 1 0 1 1 0 0 0 1 1 0 0 1 AB19h
RIG_SI
NGLE_
1
DCS_T
RIG_SI
NGLE_
0
0 0 0 0
DCS_DAC_WR_VAL_1 [7:0]
_SEL [1:0]
RIG_S
ERIES
_1
DCS_T
RIG_S
ERIES
_0
RIG_S
TART
UP_1
DCS_T
RIG_S
TART
UP_0
DCS_T
RIG_D
AC_W
R_1
DCS_T
RIG_D
AC_W
R_0
DCS_
ENA_
CHAN
_1
YN_P
WR
DCS_
ENA_
CHAN
_0
0004h
0000h
DCS_TIMER_PERIOD_01 054Ah
DCS_DAC_WR_VAL_0 [7:0] 0000h
MPLETE [1:0]
T1L_R
MV_S
HORT
HPOU
T1L_O
UTP
_COMPLETE
[1:0]
HPOU
T1L_D
LY
P_COMPLETE
[1:0]
0 0000h 0 HPOU
T1R_R
MV_S
HORT
HPOU
T1R_O
UTP
HPOU
T1R_D
LY
[3:0] MICD_RATE MICD_
DBTIM
E
MICD_
ENA
0000h
5600h
0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
MICD_
VALID
MICD_
STS
007Fh
0000h
000Xh
0 0 8004h
_INC w PP, August 2012, Rev 3.4
255
WM8958
Pre-Production
REG NAME
R272 (110h) Write Sequencer
Ctrl (1)
15
WSEQ
_ENA
14 13 12 11 10 9
_ABO
RT
8
WSEQ
_STAR
T
7 6 5 4 3 2 1 0 DEFAULT
R273 (111h) Write Sequencer
Ctrl (2)
R512 (200h) AIF1 Clocking (1)
_BUSY
[1:0] LK_IN
V
AIF1C
LK_DI
V
AIF1C
LK_EN
A
0000h
R513 (201h) AIF1 Clocking (2)
R516 (204h) AIF2 Clocking (1)
[1:0] LK_IN
V
AIF2C
LK_DI
V
AIF2C
LK_EN
A
0000h
R517 (205h) AIF2 Clocking (2)
R520 (208h) Clocking (1) 0 DSP2
CLK_E
NA
AIF1D
K_ENA SPCLK
_ENA
AIF2D
SPCLK
_ENA
SYSD
SPCLK
_ENA
SYSCL
K_SR
C
0000h
R521 (209h) Clocking (2)
R528 (210h) AIF1 Rate
R529 (211h) AIF2 Rate
R530 (212h) Rate Status
R544 (220h) FLL1 Control (1)
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 SR_ERROR 0000h
OSC_
ENA
FLL1_
ENA
0000h
R545 (221h) FLL1 Control (2)
R546 (222h) FLL1 Control (3)
R547 (223h) FLL1 Control (4)
R548 (224h) FLL1 Control (5)
0 0
0
FLL1_
BYP
0 0
FLL1_THETA [15:0]
FLL1_
FRC_
NCO
FLL2_LAMBDA [15:0]
0 FLL1_REFCLK
_DIV [1:0]
0 FLL1_REFCLK
_SRC [1:0]
0000h
0 0 0 0 0 0000h
0C80h
R550 (226h) FLL1 EFS1
R551 (227h) FLL1 EFS2
0000h
0006h
R576 (240h) FLL2 Control (1)
OSC_
ENA
EFS_E
NA
FLL2_
ENA
0000h
R577 (241h) FLL2 Control (2)
R578 (242h) FLL2 Control (3)
R579 (243h) FLL2 Control (4)
R580 (244h) FLL2 Control (5)
R582 (246h) FLL2 EFS1
R583 (247h) FLL2 EFS2
R768 (300h) AIF1 Control (1)
R769 (301h) AIF1 Control (2)
0 0
0
FLL2_
BYP
AIF1A
DCL_S
RC
AIF1D
ACL_S
RC
0 0
AIF1A
DCR_
SRC
AIF1D
ACR_
SRC
AIF1A
DC_T
DM
FLL2_THETA [15:0] 0000h
0 0 0 0 0 0000h
FLL2_
FRC_
NCO
FLL2_LAMBDA [15:0]
0 FLL2_REFCLK
_DIV [1:0]
0 FLL2_REFCLK
_SRC [1:0]
0C80h
0000h
0006h
0 AIF1_WL [1:0] AIF1_FMT [1:0] 0 0
EFS_E
NA
0 4050h
ST [1:0]
BCLK_
INV
0 AIF1_
MONO AC_C
OMP
AIF1D
AC_C
OMPM
ODE
AIF1A
DC_C
OMP
AIF1A
DC_C
OMPM
ODE
AIF1_L
OOPB
ACK
4000h
0 0 0 0 0 0 0 0 0 0 0 0 0000h R770 (302h) AIF1 Master/Slave AIF1_T
RI
R771 (303h) AIF1 BCLK
AIF1_
MSTR
AIF1_
CLK_F
RC
AIF1_L
RCLK_
FRC
0 0 0 0 0 0 0 0 0 0 0 0040h w PP, August 2012, Rev 3.4
256
Pre-Production
WM8958
REG NAME
R772 (304h) AIF1ADC LRCLK
R773 (305h) AIF1DAC LRCLK
15 14 13 10 9 8 7 6 5 4
AIF1ADC_RATE [10:0]
AIF1DAC_RATE [10:0]
3 2 1 0 DEFAULT
0040h
0040h
R774 (306h) AIF1DAC Data
R775 (307h) AIF1ADC Data
ACL_D
AT_IN
V
AIF1D
ACR_
DAT_I
NV
DCL_D
AT_IN
V
AIF1A
DCR_
DAT_I
NV
0 AIF2_WL [1:0] AIF2_FMT [1:0] 0 AIF2T
XL_EN
A
AIF2T
XR_E
NA
R784 (310h) AIF2 Control (1)
R785 (311h) AIF2 Control (2)
R786 (312h) AIF2 Master/Slave AIF2_T
RI
R787 (313h) AIF2 BCLK
R788 (314h) AIF2ADC LRCLK
AIF2A
DCL_S
RC
AIF2D
ACL_S
RC
AIF2A
DCR_
SRC
AIF2A
DC_T
DM
AIF2A
DC_T
DM_C
HAN
BCLK_
INV
AIF2ADC_RATE [10:0]
4053h
AIF2D
ACR_
SRC
AIF2D
AC_TD
M
AIF2D
AC_TD
M_CH
AN
AIF2DAC_BOO
ST [1:0]
0 AIF2_
MONO AC_C
OMP
AIF2D
AC_C
OMPM
ODE
AIF2A
DC_C
OMP
AIF2A
DC_C
OMPM
ODE
AIF2_L
OOPB
ACK
4000h
0 0 0 0 0 0 0 0 0 0 0 0 0000h AIF2_
MSTR
AIF2_
CLK_F
RC
AIF2_L
RCLK_
FRC
0 0 0 0 0040h
0040h
R789 (315h) AIF2DAC LRCLK
0 0 0 0 0 0 0
DC_LR
CLK_I
NV
AIF2A
DC_LR
CLK_D
IR
AC_LR
CLK_I
NV
AIF2D
AC_LR
CLK_D
IR
AIF2DAC_RATE [10:0]
0000h
0000h
0040h
R790 (316h) AIF2DAC Data 0000h
R791 (317h) AIF2ADC Data
ACL_D
AT_IN
V
AIF2D
ACR_
DAT_I
NV
DCL_D
AT_IN
V
AIF2A
DCR_
DAT_I
NV
0000h
R800 (320h) AIF3 Control (1)
12 11
DC_LR
CLK_I
NV
AIF1A
DC_LR
CLK_D
IR
AC_LR
CLK_I
NV
AIF1D
AC_LR
CLK_D
IR
RCLK_
INV
R801 (321h) AIF3 Control (2)
ST [1:0] AC_C
OMP
AIF3D
AC_C
OMPM
ODE
AIF3A
DC_C
OMP
AIF3A
DC_C
OMPM
ODE
AIF3_L
OOPB
ACK
0000h
R802 (322h) AIF3DAC Data 0000h
AC_D
AT_IN
V
R803 (323h) AIF3ADC Data 0000h
DC_D
AT_IN
V
R1024 (400h) AIF1 ADC1 Left
Volume DC1_V
U
AIF1ADC1L_VOL [7:0] 00C0h w PP, August 2012, Rev 3.4
257
WM8958
Pre-Production
REG NAME
R1025 (401h) AIF1 ADC1 Right
Volume
15 14 13 12 11 10 9 8 7 6 5 4 3
AIF1ADC1R_VOL [7:0]
2 1 0 DEFAULT
00C0h
DC1_V
U
R1026 (402h) AIF1 DAC1 Left
Volume AC1_V
U
AIF1DAC1L_VOL [7:0] 00C0h
R1027 (403h) AIF1 DAC1 Right
Volume AC1_V
U
AIF1DAC1R_VOL [7:0] 00C0h
R1028 (404h) AIF1 ADC2 Left
Volume DC2_V
U
AIF1ADC2L_VOL [7:0] 00C0h
R1029 (405h) AIF1 ADC2 Right
Volume
AIF1ADC2R_VOL [7:0] 00C0h
R1030 (406h) AIF1 DAC2 Left
Volume
R1031 (407h) AIF1 DAC2 Right
Volume
AIF1DAC2L_VOL [7:0] 00C0h
AC2_V
U
AIF1DAC2R_VOL [7:0] 00C0h
AC2_V
U
0 0 0 0 0 0 0 0 0 0 0 0000h R1040 (410h) AIF1 ADC1 Filters AIF1A
DC_4F
S
R1041 (411h) AIF1 ADC2 Filters
AIF1ADC1_HP
F_CUT [1:0]
AIF1A
DC1L_
HPF
AIF1A
DC1R_
HPF
0 AIF1ADC2_HP
F_CUT [1:0]
AIF1A
DC2L_
HPF
AIF1A
DC2R_
HPF
R1056 (420h) AIF1 DAC1 Filters
(1)
R1057 (421h) AIF1 DAC1 Filters
(2)
R1058 (422h) AIF1 DAC2 Filters
(1)
R1059 (423h) AIF1 DAC2 Filters
(2)
0 0 0 0 0 0 0 0 0 0 0 0000h
AC1_
MUTE
0 AIF1D
AC1_
MONO
0 AIF1D
AC1_
MUTE
RATE
AIF1D
AC1_U
NMUT
E_RA
MP
0 0 0 0 0200h
0 0 AIF1DAC1_3D_GAIN
AC1_3
D_EN
A
AC2_
MUTE
0 0 0 1 0 0 0 0 0010h
0 AIF1D
AC2_
MONO
0 AIF1D
AC2_
MUTE
RATE
AIF1D
AC2_U
NMUT
E_RA
MP
0 0 0 0 0200h
0 0 AIF1DAC2_3D_GAIN
AC2_3
D_EN
A
0 0 0 1 0 0 0 0 0010h
R1072 (430h) AIF1 DAC1 Noise
Gate
DC2_V
U
_HLD [1:0] [2:0]
0068h
AC1_N
G_EN
A
R1073 (431h) AIF1 DAC2 Noise
Gate
0068h
R1088 (440h) AIF1 DRC1 (1)
R1089 (441h) AIF1 DRC1 (2)
R1090 (442h) AIF1 DRC1 (3)
AIF1DRC1_SIG_DET_RMS [4:0]
_HLD [1:0] [2:0] AC2_N
G_EN
A
AIF1DRC1_SIG
_DET_PK [1:0]
AIF1D
RC1_N
G_EN
A
AIF1D
RC1_S
IG_DE
T_MO
DE
AIF1D
RC1_S
IG_DE
T
AIF1D
RC1_K
NEE2_
OP_E
NA
AIF1D
RC1_
QR
AIF1D
RC1_A
NTICLI
P
AIF1D
AC1_D
RC_E
NA
AIF1A
DC1L_
DRC_
ENA
AIF1A
DC1R_
DRC_
ENA
AIF1DRC1_NG_MINGAIN [3:0] AIF1DRC1_NG
_EXP [1:0]
AIF1DRC1_DCY [3:0]
AIF1DRC1_QR
_THR [1:0]
AIF1DRC1_QR
_DCY [1:0]
AIF1DRC1_MINGAIN
[2:0]
AIF1DRC1_HI_COMP
[2:0]
AIF1DRC1_MA
XGAIN [1:0]
AIF1DRC1_LO_COMP
[2:0]
0098h
0845h
0000h
PP, August 2012, Rev 3.4
w
258
Pre-Production
WM8958
REG NAME
R1091 (443h) AIF1 DRC1 (4)
R1092 (444h) AIF1 DRC1 (5)
R1104 (450h) AIF1 DRC2 (1)
R1105 (451h) AIF1 DRC2 (2)
R1106 (452h) AIF1 DRC2 (3)
R1107 (453h) AIF1 DRC2 (4)
R1108 (454h) AIF1 DRC2 (5)
R1152 (480h) AIF1 DAC1 EQ
Gains (1)
R1153 (481h) AIF1 DAC1 EQ
Gains (2)
R1154 (482h) AIF1 DAC1 EQ
Band 1 A
R1155 (483h) AIF1 DAC1 EQ
Band 1 B
R1156 (484h) AIF1 DAC1 EQ
Band 1 PG
R1157 (485h) AIF1 DAC1 EQ
Band 2 A
R1158 (486h) AIF1 DAC1 EQ
Band 2 B
R1159 (487h) AIF1 DAC1 EQ
Band 2 C
R1160 (488h) AIF1 DAC1 EQ
Band 2 PG
R1161 (489h) AIF1 DAC1 EQ
Band 3 A
R1162 (48Ah) AIF1 DAC1 EQ
Band 3 B
R1163 (48Bh) AIF1 DAC1 EQ
Band 3 C
R1164 (48Ch) AIF1 DAC1 EQ
Band 3 PG
R1165 (48Dh) AIF1 DAC1 EQ
Band 4 A
R1166 (48Eh) AIF1 DAC1 EQ
Band 4 B
R1167 (48Fh) AIF1 DAC1 EQ
Band 4 C
R1168 (490h) AIF1 DAC1 EQ
Band 4 PG
R1169 (491h) AIF1 DAC1 EQ
Band 5 A
R1170 (492h) AIF1 DAC1 EQ
Band 5 B
R1171 (493h) AIF1 DAC1 EQ
Band 5 PG
R1172 (494h) AIF1 DAC1 EQ
Band 1 C w
15 14 13 12 11 10
0 0 0 0 0
9 8 7 6 5 4 3 2 1 0 DEFAULT
AIF1DRC1_KNEE_OP 0000h
0 0 0 0 0 0 AIF1DRC1_KNEE2_IP
AIF1DRC2_SIG_DET_RMS [4:0] AIF1DRC2_SIG
_DET_PK [1:0]
AIF1D
RC2_N
G_EN
A
AIF1D
RC2_S
IG_DE
T_MO
DE
AIF1D
RC2_S
IG_DE
T
AIF1D
RC2_K
NEE2_
OP_E
NA
AIF1D
RC2_
QR
AIF1D
RC2_A
NTICLI
P
AIF1D
AC2_D
RC_E
NA
AIF1A
DC2L_
DRC_
ENA
AIF1A
DC2R_
DRC_
ENA
0098h
AIF1DRC2_NG_MINGAIN [3:0] AIF1DRC2_NG
_EXP [1:0]
0 0 0 0 0
_THR [1:0]
AIF1DRC2_DCY [3:0]
AIF1DRC2_QR AIF1DRC2_QR
_DCY [1:0]
AIF1DRC2_MINGAIN
[2:0]
AIF1DRC2_MA
XGAIN [1:0]
0845h
AIF1DRC2_HI_COMP
[2:0]
AIF1DRC2_LO_COMP
[2:0]
0000h
AIF1DRC2_KNEE_OP 0000h
0 0 0 0 0 0 AIF1DRC2_KNEE2_IP
AIF1DAC1_EQ_B1_GAIN [4:0] AIF1DAC1_EQ_B2_GAIN [4:0] AIF1DAC1_EQ_B3_GAIN [4:0]
AIF1DAC1_EQ_B4_GAIN [4:0] 0 0 0 0 0
AC1_E
Q_MO
DE
AIF1DAC1_EQ_B1_A [15:0]
AIF1D
AC1_E
Q_EN
A
6318h
6300h
0FCAh
AIF1DAC1_EQ_B1_B [15:0]
AIF1DAC1_EQ_B1_PG [15:0]
AIF1DAC1_EQ_B2_A [15:0]
AIF1DAC1_EQ_B2_B [15:0]
AIF1DAC1_EQ_B2_C [15:0]
AIF1DAC1_EQ_B2_PG [15:0]
AIF1DAC1_EQ_B3_A [15:0]
AIF1DAC1_EQ_B3_B [15:0]
AIF1DAC1_EQ_B3_C [15:0]
AIF1DAC1_EQ_B3_PG [15:0]
AIF1DAC1_EQ_B4_A [15:0]
AIF1DAC1_EQ_B4_B [15:0]
AIF1DAC1_EQ_B4_C [15:0]
AIF1DAC1_EQ_B4_PG [15:0]
AIF1DAC1_EQ_B5_A [15:0]
AIF1DAC1_EQ_B5_B [15:0]
AIF1DAC1_EQ_B5_PG [15:0]
AIF1DAC1_EQ_B1_C [15:0]
0400h
00D8h
1EB5h
F145h
0B75h
01C5h
1C58h
F373h
0A54h
0558h
168Eh
F829h
07ADh
1103h
0564h
0559h
4000h
0000h
PP, August 2012, Rev 3.4
259
WM8958
REG NAME
R1184 (4A0h) AIF1 DAC2 EQ
Gains (1)
R1185 (4A1h) AIF1 DAC2 EQ
Gains (2)
R1186 (4A2h) AIF1 DAC2 EQ
Band 1 A
R1187 (4A3h) AIF1 DAC2 EQ
Band 1 B
R1188 (4A4h) AIF1 DAC2 EQ
Band 1 PG
R1189 (4A5h) AIF1 DAC2 EQ
Band 2 A
R1190 (4A6h) AIF1 DAC2 EQ
Band 2 B
R1191 (4A7h) AIF1 DAC2 EQ
Band 2 C
R1192 (4A8h) AIF1 DAC2 EQ
Band 2 PG
R1193 (4A9h) AIF1 DAC2 EQ
Band 3 A
R1194 (4AAh) AIF1 DAC2 EQ
Band 3 B
R1195 (4ABh) AIF1 DAC2 EQ
Band 3 C
R1196 (4ACh) AIF1 DAC2 EQ
Band 3 PG
R1197 (4ADh) AIF1 DAC2 EQ
Band 4 A
R1198 (4AEh) AIF1 DAC2 EQ
Band 4 B
R1199 (4AFh) AIF1 DAC2 EQ
Band 4 C
R1200 (4B0h) AIF1 DAC2 EQ
Band 4 PG
R1201 (4B1h) AIF1 DAC2 EQ
Band 5 A
R1202 (4B2h) AIF1 DAC2 EQ
Band 5 B
R1203 (4B3h) AIF1 DAC2 EQ
Band 5 PG
R1204 (4B4h) AIF1 DAC2 EQ
Band 1 C
R1280 (500h) AIF2 ADC Left
Volume
R1281 (501h) AIF2 ADC Right
Volume
R1282 (502h) AIF2 DAC Left
Volume
R1283 (503h) AIF2 DAC Right
Volume
Pre-Production
15 14 13 12
AIF1DAC2_EQ_B1_GAIN [4:0]
11 10 9 8 7
AIF1DAC2_EQ_B2_GAIN [4:0]
6 5 4 3 2
AIF1DAC2_EQ_B3_GAIN [4:0]
1 0 DEFAULT
AIF1D
AC2_E
Q_EN
A
6318h
6300h AIF1DAC2_EQ_B4_GAIN [4:0] 0 0 0 0 0
AC2_E
Q_MO
DE
AIF1DAC2_EQ_B1_A [15:0] 0FCAh
AIF1DAC2_EQ_B1_B [15:0]
AIF1DAC2_EQ_B1_PG [15:0]
AIF1DAC2_EQ_B2_A [15:0]
AIF1DAC2_EQ_B2_B [15:0]
AIF1DAC2_EQ_B2_C [15:0]
AIF1DAC2_EQ_B2_PG [15:0]
AIF1DAC2_EQ_B3_A [15:0]
AIF1DAC2_EQ_B3_B [15:0]
AIF1DAC2_EQ_B3_C [15:0]
AIF1DAC2_EQ_B3_PG [15:0]
AIF1DAC2_EQ_B4_A [15:0]
AIF1DAC2_EQ_B4_B [15:0]
AIF1DAC2_EQ_B4_C [15:0]
AIF1DAC2_EQ_B4_PG [15:0]
AIF1DAC2_EQ_B5_A [15:0]
AIF1DAC2_EQ_B5_B [15:0]
AIF1DAC2_EQ_B5_PG [15:0]
AIF1DAC2_EQ_B1_C [15:0]
AIF2ADCL_VOL [7:0]
0564h
0559h
4000h
0000h
00C0h
168Eh
F829h
07ADh
1103h
0B75h
01C5h
1C58h
F373h
0A54h
0558h
0400h
00D8h
1EB5h
F145h
DC_V
U
DC_V
U
AC_V
U
AC_V
U
AIF2ADCR_VOL [7:0]
AIF2DACL_VOL [7:0]
AIF2DACR_VOL [7:0]
00C0h
00C0h
00C0h w PP, August 2012, Rev 3.4
260
Pre-Production
WM8958
REG NAME
R1296 (510h) AIF2 ADC Filters
R1312 (520h) AIF2 DAC Filters
(1)
R1313 (521h) AIF2 DAC Filters
(2)
R1328 (531h) AIF2 DAC Noise
Gate
R1344 (540h) AIF2 DRC (1)
R1345 (541h) AIF2 DRC (2)
R1346 (542h) AIF2 DRC (3)
R1347 (543h) AIF2 DRC (4)
R1348 (544h) AIF2 DRC (5)
R1408 (580h) AIF2 EQ Gains (1)
R1409 (581h) AIF2 EQ Gains (2)
R1410 (582h) AIF2 EQ Band 1 A
R1411 (583h) AIF2 EQ Band 1 B
R1412 (584h) AIF2 EQ Band 1
PG
R1413 (585h) AIF2 EQ Band 2 A
R1414 (586h) AIF2 EQ Band 2 B
R1415 (587h) AIF2 EQ Band 2 C
R1416 (588h) AIF2 EQ Band 2
PG
R1417 (589h) AIF2 EQ Band 3 A
R1418 (58Ah) AIF2 EQ Band 3 B
R1419 (58Bh) AIF2 EQ Band 3 C
R1420 (58Ch) AIF2 EQ Band 3
PG
R1421 (58Dh) AIF2 EQ Band 4 A
R1422 (58Eh) AIF2 EQ Band 4 B
R1423 (58Fh) AIF2 EQ Band 4 C
R1424 (590h) AIF2 EQ Band 4
PG
R1425 (591h) AIF2 EQ Band 5 A
R1426 (592h) AIF2 EQ Band 5 B
R1427 (593h) AIF2 EQ Band 5
PG
R1428 (594h) AIF2 EQ Band 1 C
15 14 13 12
0 AIF2ADC_HPF
_CUT [1:0]
AIF2A
DCL_H
PF
11
AIF2A
DCR_
HPF
10 9 8 7 6 5 4 3 2 1 0 DEFAULT
0 0 0 0 0 0 0 0 0 0 0 0000h
0 0
AIF2DRC_SIG_DET_RMS [4:0]
AC_M
UTE
0 AIF2D
AC_M
ONO
0 AIF2D
AC_M
UTER
ATE
AIF2D
AC_U
NMUT
E_RA
MP
0 0 0 0 0200h
0 0 0 1 0 0 0 0 0010h
AC_3D
_ENA
0068h
AIF2DRC_SIG_
DET_PK [1:0]
AIF2D
RC_N
G_EN
A
AIF2D
RC_SI
G_DE
T_MO
DE
HLD [1:0]
AIF2D
RC_SI
G_DE
T
AIF2D
RC_K
NEE2_
OP_E
NA
AIF2D
RC_Q
R
AIF2D
RC_A
NTICLI
P
[2:0]
AIF2D
AC_D
RC_E
NA
AIF2A
DCL_D
RC_E
NA
AC_N
G_EN
A
AIF2A
DCR_
DRC_
ENA
0098h
0 0 0 AIF2DRC_ATK
AIF2DRC_NG_MINGAIN [3:0] AIF2DRC_NG_
EXP [1:0]
0 0 0 0 0
AIF2DRC_QR_
THR [1:0]
AIF2DRC_QR_
DCY [1:0]
0845h
[2:0] GAIN [1:0]
AIF2DRC_HI_COMP
[2:0]
AIF2DRC_LO_COMP
[2:0]
0000h
AIF2DRC_KNEE_OP 0000h
0 0 0 0 0 0 AIF2DRC_KNEE2_IP
AIF2DAC_EQ_B1_GAIN [4:0] AIF2DAC_EQ_B2_GAIN [4:0]
AIF2DRC_KNEE2_OP
AIF2DAC_EQ_B3_GAIN [4:0] AIF2D
AC_E
Q_EN
A
0000h
6318h
6300h
AC_E
Q_MO
DE
AIF2DAC_EQ_B1_A [15:0]
AIF2DAC_EQ_B1_B [15:0]
AIF2DAC_EQ_B1_PG [15:0]
0FCAh
0400h
00D8h
AIF2DAC_EQ_B2_A [15:0]
AIF2DAC_EQ_B2_B [15:0]
AIF2DAC_EQ_B2_C [15:0]
AIF2DAC_EQ_B2_PG [15:0]
AIF2DAC_EQ_B3_A [15:0]
AIF2DAC_EQ_B3_B [15:0]
AIF2DAC_EQ_B3_C [15:0]
AIF2DAC_EQ_B3_PG [15:0]
AIF2DAC_EQ_B4_A [15:0]
AIF2DAC_EQ_B4_B [15:0]
AIF2DAC_EQ_B4_C [15:0]
AIF2DAC_EQ_B4_PG [15:0]
AIF2DAC_EQ_B5_A [15:0]
AIF2DAC_EQ_B5_B [15:0]
AIF2DAC_EQ_B5_PG [15:0]
AIF2DAC_EQ_B1_C [15:0]
1EB5h
F145h
0B75h
01C5h
1C58h
F373h
0A54h
0558h
168Eh
F829h
07ADh
1103h
0564h
0559h
4000h
0000h w PP, August 2012, Rev 3.4
261
WM8958
REG NAME
R1536 (600h) DAC1 Mixer
Volumes
R1537 (601h) DAC1 Left Mixer
Routing
15 14 13 12
Pre-Production
11 10 9 8 7 6 5 4 3 2 1 0 DEFAULT
R1538 (602h) DAC1 Right Mixer
Routing
_TO_D
AC1L
ADCL_
TO_D
AC1L
_TO_D
AC1R
ADCL_
TO_D
AC1R
0 AIF2D
ACL_T
O_DA
C1L
AIF1D
AC2L_
TO_D
AC1L
AIF1D
AC1L_
TO_D
AC1L
0 AIF2D
ACR_T
O_DA
C1R
AIF1D
AC2R_
TO_D
AC1R
AIF1D
AC1R_
TO_D
AC1R
0000h
0000h
R1539 (603h) DAC2 Mixer
Volumes
R1540 (604h) DAC2 Left Mixer
Routing
R1541 (605h) DAC2 Right Mixer
Routing
R1542 (606h) AIF1 ADC1 Left
Mixer Routing
R1543 (607h) AIF1 ADC1 Right
Mixer Routing
R1544 (608h) AIF1 ADC2 Left
Mixer Routing
R1545 (609h) AIF1 ADC2 Right mixer Routing
R1552 (610h) DAC1 Left Volume
_TO_D
AC2L
_TO_D
AC2R
ADCL_
TO_D
AC2L
ADCL_
TO_D
AC2R
0 AIF2D
ACL_T
O_DA
C2L
AIF1D
AC2L_
TO_D
AC2L
AIF1D
AC1L_
TO_D
AC2L
0 AIF2D
ACR_T
O_DA
C2R
AIF1D
AC2R_
TO_D
AC2R
AIF1D
AC1R_
TO_D
AC2R
_TO_A
IF1AD
C1L
AIF2D
ACL_T
O_AIF
1ADC1
L
R_TO_
AIF1A
DC1R
AIF2D
ACR_T
O_AIF
1ADC1
R
_TO_A
IF1AD
C2L
AIF2D
ACL_T
O_AIF
1ADC2
L
R_TO_
AIF1A
DC2R
AIF2D
ACR_T
O_AIF
1ADC2
R
DAC1L_VOL [7:0]
0000h
0000h
0000h
0000h
0000h
0000h
02C0h
R1553 (611h) DAC1 Right
Volume
R1554 (612h) DAC2 Left Volume
R1555 (613h) DAC2 Right
Volume
R1556 (614h) DAC Softmute
_MUT
E
_MUT
E
DAC1_
VU
R_MU
TE
DAC1_
VU
DAC2_
VU
R_MU
TE
DAC2_
VU
DAC1R_VOL [7:0]
DAC2L_VOL [7:0]
DAC2R_VOL [7:0]
02C0h
02C0h
02C0h
R1568 (620h) Oversampling
R1569 (621h) Sidetone
F
SOFT
MUTE
MODE
DAC_
MUTE
RATE
OSR12
8
DAC_
OSR12
8
EL
STL_S
EL
0000h
0002h
0000h w PP, August 2012, Rev 3.4
262
Pre-Production
WM8958
REG NAME
R1792 (700h) GPIO 1
15 14
GP1_D
IR
GP1_P
U
13
GP1_P
D
12 11 10
OL
9
GP1_
OP_C
FG
8
GP1_D
B
7 6
0 GP1_L
VL
5 4 3 2 1 0 DEFAULT
0 0 0 0 1 0 0 0 0 0 0 0 1 A101h R1793 (701h) Pull Control
(MCLK2)
R1794 (702h) Pull Control
(BCLK2)
R1795 (703h) Pull Control
(DACLRCLK2)
R1796 (704h) Pull Control
(DACDAT2)
1 MCLK
2_PU
MCLK
2_PD
1 BCLK2
_PU
BCLK2
_PD
1 DACL
RCLK2
_PU
DACL
RCLK2
_PD
1 DACD
AT2_P
U
DACD
AT2_P
D
GP6_D
IR
GP6_P
U
GP6_P
D
0 0 0 0 1 0 0 0 0 0 0 0 1 A101h
0 0 0 0 1 0 0 0 0 0 0 0 1 A101h
0 0 0 0 1 0 0 0 0 0 0 0 1 A101h
R1797 (705h) GPIO 6
R1799 (707h) GPIO 8
R1800 (708h) GPIO 9
GP8_D
IR
GP9_D
IR
GP8_P
U
GP9_P
U
GP8_P
D
GP9_P
D
OL
OL
OL
GP6_
OP_C
FG
GP8_
OP_C
FG
GP6_D
B
GP8_D
B
GP9_
OP_C
FG
GP9_D
B
0 GP6_L
VL
0 GP8_L
VL
0 GP9_L
VL
R1801 (709h) GPIO 10
R1802 (70Ah) GPIO 11
GP10_
DIR
GP11_
DIR
GP10_
PU
GP11_
PU
GP10_
PD
GP11_
PD
POL
POL
GP10_
OP_C
FG
GP10_
DB
GP11_
OP_C
FG
GP11_
DB
AT2_P
U
DMICD
AT2_P
D
DMICD
AT1_P
U
DMICD
AT1_P
D
0 GP10_
LVL
0 GP11_
LVL
R1824 (720h) Pull Control (1)
R1825 (721h) Pull Control (2)
_PD
MCLK
1_PU
MCLK
1_PD
DACD
AT1_P
U
0 LDO2E
NA_P
D
DACD
AT1_P
D
0 LDO1E
NA_P
D
DACL
RCLK1
_PU
DACL
RCLK1
_PD
BCLK1
_PU
ODE_
PU
BCLK1
_PD
0000h
0 0156h
0000h R1840 (730h) Interrupt Status 1
R1841 (731h) Interrupt Status 2
R1842 (732h) Interrupt Raw
Status 2
TEMP
_WAR
N_EIN
T
DCS_
DONE
_EINT
WSEQ
_DON
E_EIN
T
FIFOS
_ERR_
EINT
AIF2D
RC_SI
G_DE
T_EIN
T
EINT
GP10_
EINT
GP9_E
INT
GP8_E
INT
AIF1D
RC2_S
IG_DE
T_EIN
T
AIF1D
RC1_S
IG_DE
T_EIN
T
SRC2_
LOCK_
EINT
SRC1_
LOCK_
EINT
0 GP6_E
INT
FLL2_
LOCK_
EINT
FLL1_
LOCK_
EINT
TEMP
_WAR
N_STS
DCS_
DONE
_STS
WSEQ
_DON
E_STS
FIFOS
_ERR_
STS
AIF2D
RC_SI
G_DE
T_STS
AIF1D
RC2_S
IG_DE
T_STS
AIF1D
RC1_S
IG_DE
T_STS
SRC2_
LOCK_
STS
SRC1_
LOCK_
STS
FLL2_
LOCK_
STS
FLL1_
LOCK_
STS
R1848 (738h) Interrupt Status 1
Mask
R1849 (739h) Interrupt Status 2
Mask
11_EI
NT
IM_GP
10_EI
NT
IM_GP
9_EIN
T
IM_GP
8_EIN
T
IM_TE
MP_W
ARN_
EINT
IM_DC
S_DO
NE_EI
NT
IM_WS
EQ_D
ONE_
EINT
IM_FIF
OS_E
RR_EI
NT
IM_AIF
2DRC_
SIG_D
ET_EI
NT
IM_AIF
1DRC2
_SIG_
DET_E
INT
IM_AIF
1DRC1
_SIG_
DET_E
INT
IM_SR
C2_LO
CK_EI
NT
IM_SR
C1_LO
CK_EI
NT
1 IM_GP
6_EIN
T
IM_FL
L2_LO
CK_EI
NT
IM_FL
L1_LO
CK_EI
NT
EINT
CD_EI
NT
INT
TEMP
_SHUT
_EINT
_SHUT
_STS
1_EIN
T
IM_TE
MP_S
HUT_E
INT
0000h
0000h
07FFh
FFFFh
R1856 (740h) Interrupt Control 0000h
Q
R1864 (748h) IRQ Debounce 003Fh
_WAR
N_DB
_SHUT
_DB
R2304 (900h) DSP2_Program 1C00h
ENA w PP, August 2012, Rev 3.4
263
WM8958
Pre-Production
REG NAME
R2305 (901h) DSP2_Config
15 14 13
0 0
12 11 10 9 8 7 6 5 4
MBC_SEL
3 2 1 0 DEFAULT
MBC_
ENA
0000h
R2573 (A0Dh) DSP2_ExecControl 0 0 0 0 0 0 0 0 0 0 0 0 0
STOP
DSP2_
RUNR
R12288
(3000h)
Write Sequencer 0 0 0
R12289
(3001h)
Write Sequencer 1 0 0 0 0 0 0 0 0
Write Sequencer 2 R12290
(3002h)
R12291
(3003h)
R12292
(3004h)
Write Sequencer 3
Write Sequencer 4 0 0
[2:0]
_EOS0
Write Sequencer 5 0 0 0 0 0 0 0 0 R12293
(3005h)
R12294
(3006h)
Write Sequencer 6
[2:0]
Write Sequencer 7 R12295
(3007h) _EOS1
0 0000h
0039h
001Bh
0001h
0003h
004Ch
[similar for WSEQ address 2 … 126] 0001h
0006h
0000h R12796
(31FCh)
R12797
(31FDh)
R12798
(31FEh)
R12799
(31FFh)
Write Sequencer
508
Write Sequencer
509
Write Sequencer
510
Write Sequencer
511
0 0 0 0 0 0 0 0
27 [2:0]
_EOS1
27
WSEQ_DATA_START127
0000h
0000h w PP, August 2012, Rev 3.4
264
Pre-Production
REGISTER BITS BY ADDRESS
REGISTER
ADDRESS
R0 (00h)
Software
Reset
[15:0]
DESCRIPTION
_0000_000
0
Writing to this register resets all registers to their default state. (Note - Control Write Sequencer registers are not affected by Software Reset.)
Reading from this register will indicate device ID 8958h.
Register 00h Software Reset
REGISTER
ADDRESS
R1 (01h)
Power
Managemen t (1)
DESCRIPTION
13 SPKOUTR_EN
A
12 SPKOUTL_EN
A
0
0
SPKMIXR Mixer, SPKRVOL PGA and SPKOUTR
Output Enable
0 = Disabled
1 = Enabled
SPKMIXL Mixer, SPKLVOL PGA and SPKOUTL Output
Enable
0 = Disabled
1 = Enabled
11 HPOUT2_ENA 0 HPOUT2 Output Stage Enable
0 = Disabled
1 = Enabled
9 HPOUT1L_EN
A
0 Enables HPOUT1L input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the HPOUT1L Enable sequence.
8 HPOUT1R_EN
A
0 Enables HPOUT1R input stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set as the first step of the HPOUT1R Enable sequence.
5 MICB2_ENA 0 Microphone Bias 2 Enable
0 = Disabled
1 = Enabled
4 MICB1_ENA 0 Microphone Bias 1 Enable
0 = Disabled
1 = Enabled
2:1 VMID_SEL
[1:0]
00 VMID Divider Enable and Select
00 = VMID disabled (for OFF mode)
01 = 2 x 40k divider (for normal operation)
10 = 2 x 240k divider (for low power standby)
11 = Reserved
0 BIAS_ENA 0 Enables the Normal bias current generator (for all analogue functions)
0 = Disabled
1 = Enabled
Register 01h Power Management (1)
WM8958 w PP, August 2012, Rev 3.4
265
WM8958
Pre-Production
REGISTER
ADDRESS
R2 (02h)
Power
Managemen t (2)
DESCRIPTION
14 TSHUT_ENA 1 Thermal sensor enable
0 = Disabled
1 = Enabled
13 TSHUT_OPDIS 1 Thermal shutdown control
(Causes audio outputs to be disabled if an overtemperature occurs. The thermal sensor must also be enabled.)
0 = Disabled
1 = Enabled
11 OPCLK_ENA 0 GPIO Clock Output (OPCLK) Enable
0 = Disabled
1 = Enabled
9 MIXINL_ENA 0 Left Input Mixer Enable
(Enables MIXINL and RXVOICE input to MIXINL)
0 = Disabled
1 = Enabled
8 MIXINR_ENA 0 Right Input Mixer Enable
(Enables MIXINR and RXVOICE input to MIXINR)
0 = Disabled
1 = Enabled
7 IN2L_ENA 0 IN2L Input PGA Enable
0 = Disabled
1 = Enabled
6 IN1L_ENA 0 IN1L Input PGA Enable
0 = Disabled
1 = Enabled
5 IN2R_ENA 0 IN2R Input PGA Enable
0 = Disabled
1 = Enabled
4 IN1R_ENA 0 IN1R Input PGA Enable
0 = Disabled
1 = Enabled
Register 02h Power Management (2)
REGISTER
ADDRESS
R3 (03h)
Power
Managemen t (3)
13 LINEOUT1N_E
NA
12 LINEOUT1P_E
NA
0
0
DESCRIPTION
11 LINEOUT2N_E
NA
10 LINEOUT2P_E
NA
9 SPKRVOL_EN
A w
0
0
0
LINEOUT1N Line Out and LINEOUT1NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT1P Line Out and LINEOUT1PMIX Enable
0 = Disabled
1 = Enabled
LINEOUT2N Line Out and LINEOUT2NMIX Enable
0 = Disabled
1 = Enabled
LINEOUT2P Line Out and LINEOUT2PMIX Enable
0 = Disabled
1 = Enabled
SPKMIXR Mixer and SPKRVOL PGA Enable
0 = Disabled
1 = Enabled
Note that SPKMIXR and SPKRVOL are also enabled when SPKOUTR_ENA is set.
PP, August 2012, Rev 3.4
266
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
8 SPKLVOL_EN
A
7 MIXOUTLVOL_
ENA
6 MIXOUTRVOL
_ENA
5 MIXOUTL_EN
A
4 MIXOUTR_EN
A
0
0
0
0
0
SPKMIXL Mixer and SPKLVOL PGA Enable
0 = Disabled
1 = Enabled
Note that SPKMIXL and SPKLVOL are also enabled when SPKOUTL_ENA is set.
MIXOUTL Left Volume Control Enable
0 = Disabled
1 = Enabled
MIXOUTR Right Volume Control Enable
0 = Disabled
1 = Enabled
MIXOUTL Left Output Mixer Enable
0 = Disabled
1 = Enabled
MIXOUTR Right Output Mixer Enable
0 = Disabled
1 = Enabled
Register 03h Power Management (3)
REGISTER
ADDRESS
R4 (04h)
Power
Managemen t (4)
13 AIF2ADCL_EN
A
12 AIF2ADCR_EN
A
11 AIF1ADC2L_E
NA
10 AIF1ADC2R_E
NA
0
0
0
0
DESCRIPTION
Enable AIF2ADC (Left) output path
0 = Disabled
1 = Enabled
Enable AIF2ADC (Right) output path
0 = Disabled
1 = Enabled
Enable AIF1ADC2 (Left) output path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1ADC2 (Right) output path (AIF1, Timeslot
1)
0 = Disabled
9 AIF1ADC1L_E
NA
8 AIF1ADC1R_E
NA
0
0
1 = Enabled
Enable AIF1ADC1 (Left) output path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1ADC1 (Right) output path (AIF1, Timeslot
0)
0 = Disabled
1 = Enabled
5 DMIC2L_ENA 0 Digital microphone DMICDAT2 Left channel enable
0 = Disabled
1 = Enabled
4 DMIC2R_ENA 0 Digital microphone DMICDAT2 Right channel enable
0 = Disabled
1 = Enabled
3 DMIC1L_ENA 0 Digital microphone DMICDAT1 Left channel enable
0 = Disabled
1 = Enabled
2 DMIC1R_ENA 0 Digital microphone DMICDAT1 Right channel enable
0 = Disabled w
WM8958
PP, August 2012, Rev 3.4
267
WM8958
Pre-Production
REGISTER
ADDRESS
1 = Enabled
1 ADCL_ENA 0 Left ADC Enable
0 = Disabled
1 = Enabled
0 ADCR_ENA 0 Right ADC Enable
0 = Disabled
1 = Enabled
Register 04h Power Management (4)
DESCRIPTION
REGISTER
ADDRESS
R5 (05h)
Power
Managemen t (5)
DESCRIPTION
13 AIF2DACL_EN
A
12 AIF2DACR_EN
A
11 AIF1DAC2L_E
NA
10 AIF1DAC2R_E
NA
0
0
0
0
Enable AIF2DAC (Left) input path
0 = Disabled
1 = Enabled
Enable AIF2DAC (Right) input path
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Left) input path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
Enable AIF1DAC2 (Right) input path (AIF1, Timeslot 1)
0 = Disabled
1 = Enabled
9 AIF1DAC1L_E
NA
8 AIF1DAC1R_E
NA
0
0
Enable AIF1DAC1 (Left) input path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
Enable AIF1DAC1 (Right) input path (AIF1, Timeslot 0)
0 = Disabled
1 = Enabled
3 DAC2L_ENA 0 Left DAC2 Enable
0 = Disabled
1 = Enabled
2 DAC2R_ENA 0 Right DAC2 Enable
0 = Disabled
1 = Enabled
1 DAC1L_ENA 0 Left DAC1 Enable
0 = Disabled
1 = Enabled
0 DAC1R_ENA 0 Right DAC1 Enable
0 = Disabled
1 = Enabled
Register 05h Power Management (5) w PP, August 2012, Rev 3.4
268
Pre-Production
REGISTER
ADDRESS
R6 (06h)
Power
Managemen t (6)
10:9 AIF3ADC_SRC
[1:0]
00
8:7 AIF2DAC_SRC
[1:0]
00
4:3 AIF3_ADCDAT
_SRC [1:0]
2 AIF2_ADCDAT
_SRC
1 AIF2_DACDAT
_SRC
0 AIF1_DACDAT
_SRC
00
0
0
0
DESCRIPTION
AIF3 Mono PCM output source select
00 = None
01 = AIF2ADC (Left) output path
10 = AIF2ADC (Right) output path
11 = Reserved
AIF2 input path select
00 = Left and Right inputs from AIF2
01 = Left input from AIF2; Right input from AIF3
10 = Left input from AIF3; Right input from AIF2
11 = Reserved
0 = AIF3 pins operate normally
1 = Tri-state all AIF3 interface pins
Note that pins not configured as AIF3 functions are not affected by this register.
GPIO9/ADCDAT3 Source select
00 = AIF1 ADCDAT1
01 = AIF2 ADCDAT2
10 = DACDAT2
11 = AIF3 Mono PCM output
Note that GPIO9 must be configured as ADCDAT3.
ADCDAT2 Source select
0 = AIF2 ADCDAT2
1 = GPIO8/DACDAT3
For selection 1, the GPIO8 pin must also be configured as DACDAT3.
AIF2 DACDAT Source select
0 = DACDAT2
1 = GPIO8/DACDAT3
For selection 1, the GPIO8 pin must also be configured as DACDAT3.
AIF1 DACDAT Source select
0 = DACDAT1
1 = GPIO8/DACDAT3
Note that, for selection 1, the GPIO8 pin must be configured as DACDAT3.
Register 06h Power Management (6)
REGISTER
ADDRESS
R21 (15h)
Input Mixer
(1)
8 IN1RP_MIXINR
_BOOST
7 IN1LP_MIXINL
_BOOST
0
0
DESCRIPTION
IN1RP Pin (PGA Bypass) to MIXINR Gain Boost.
This bit selects the maximum gain setting of the
IN1RP_MIXINR_VOL register.
0 = Maximum gain is +6dB
1 = Maximum gain is +15dB
IN1LP Pin (PGA Bypass) to MIXINL Gain Boost.
This bit selects the maximum gain setting of the
IN1LP_MIXINL_VOL register.
0 = Maximum gain is +6dB
1 = Maximum gain is +15dB
WM8958 w PP, August 2012, Rev 3.4
269
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
6 INPUTS_CLAM
P
0 Input pad VMID clamp
0 = Clamp de-activated
1 = Clamp activated
Register 15h Input Mixer (1)
REGISTER
ADDRESS
R24 (18h)
Left Line
Input 1&2
Volume
DESCRIPTION
Writing a 1 to this bit will cause IN1L and IN1R input
PGA volumes to be updated simultaneously
7 IN1L_MUTE 1 IN1L PGA Mute
0 = Disable Mute
1 = Enable Mute
0 = Change gain immediately
1 = Change gain on zero cross only
-16.5dB to +30dB in 1.5dB steps
Register 18h Left Line Input 1&2 Volume
REGISTER
ADDRESS
R25 (19h)
Left Line
Input 3&4
Volume
DESCRIPTION
Writing a 1 to this bit will cause IN2L and IN2R input
PGA volumes to be updated simultaneously
7 IN2L_MUTE 1 IN2L PGA Mute
0 = Disable Mute
1 = Enable Mute
0 = Change gain immediately
1 = Change gain on zero cross only
-16.5dB to +30dB in 1.5dB steps
Register 19h Left Line Input 3&4 Volume
REGISTER
ADDRESS
R26 (1Ah)
Right Line
Input 1&2
Volume
DESCRIPTION
Writing a 1 to this bit will cause IN1L and IN1R input
PGA volumes to be updated simultaneously
7 IN1R_MUTE 1 IN1R PGA Mute
0 = Disable Mute
1 = Enable Mute
0 = Change gain immediately
1 = Change gain on zero cross only
4:0 IN1R_VOL 0_1011 IN1R Volume
[4:0]
-16.5dB to +30dB in 1.5dB steps
Register 1Ah Right Line Input 1&2 Volume w PP, August 2012, Rev 3.4
270
Pre-Production
WM8958
REGISTER
ADDRESS
R27 (1Bh)
Right Line
Input 3&4
Volume
DESCRIPTION
Writing a 1 to this bit will cause IN2L and IN2R input
PGA volumes to be updated simultaneously
7 IN2R_MUTE 1 IN2R PGA Mute
0 = Disable Mute
1 = Enable Mute
0 = Change gain immediately
1 = Change gain on zero cross only
4:0 IN2R_VOL 0_1011 IN2R Volume
[4:0]
-16.5dB to +30dB in 1.5dB steps
Register 1Bh Right Line Input 3&4 Volume
REGISTER
ADDRESS
R28 (1Ch)
Left Output
Volume
DESCRIPTION
8 HPOUT1_VU 0 Headphone Output PGA Volume Update
Writing a 1 to this bit will update HPOUT1LVOL and
HPOUT1RVOL volumes simultaneously.
7 HPOUT1L_ZC 0 HPOUT1LVOL (Left Headphone Output PGA) Zero
Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
6 HPOUT1L_MU
TE_N
1 HPOUT1LVOL (Left Headphone Output PGA) Mute
0 = Mute
1 = Un-mute
5:0 HPOUT1L_VO
L [5:0]
10_1101 HPOUT1LVOL (Left Headphone Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
Register 1Ch Left Output Volume
REGISTER
ADDRESS
R29 (1Dh)
Right Output
Volume
DESCRIPTION
8 HPOUT1_VU 0 Headphone Output PGA Volume Update
Writing a 1 to this bit will update HPOUT1LVOL and
HPOUT1RVOL volumes simultaneously.
7 HPOUT1R_ZC 0 HPOUT1RVOL (Right Headphone Output PGA) Zero
Cross Enable
0 = Zero cross disabled
1 = Zero cross enabled
6 HPOUT1R_MU
TE_N
1 HPOUT1RVOL (Right Headphone Output PGA) Mute
0 = Mute
1 = Un-mute
5:0 HPOUT1R_VO
L [5:0]
10_1101 HPOUT1RVOL (Right Headphone Output PGA)
Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB w PP, August 2012, Rev 3.4
271
WM8958
Register 1Dh Right Output Volume
Pre-Production
REGISTER
ADDRESS
R30 (1Eh)
Line Outputs
Volume
6 LINEOUT1N_M
UTE
5 LINEOUT1P_M
UTE
4 LINEOUT1_VO
L
2 LINEOUT2N_M
UTE
1 LINEOUT2P_M
UTE
0 LINEOUT2_VO
L
1
1
0
1
1
0
DESCRIPTION
LINEOUT1N Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT1P Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT1 Line Output Volume
0 = 0dB
1 = -6dB
Applies to both LINEOUT1N and LINEOUT1P
LINEOUT2N Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT2P Line Output Mute
0 = Un-mute
1 = Mute
LINEOUT2 Line Output Volume
0 = 0dB
1 = -6dB
Applies to both LINEOUT2N and LINEOUT2P
Register 1Eh Line Outputs Volume
REGISTER
ADDRESS
R31 (1Fh)
HPOUT2
Volume
DESCRIPTION
5 HPOUT2_MUT
E
1 HPOUT2 (Earpiece Driver) Mute
0 = Un-mute
1 = Mute
4 HPOUT2_VOL 0 HPOUT2 (Earpiece Driver) Volume
0 = 0dB
1 = -6dB
Register 1Fh HPOUT2 Volume
REGISTER
ADDRESS
R32 (20h)
Left OPGA
Volume
DESCRIPTION
8 MIXOUT_VU 0 Mixer Output PGA Volume Update
Writing a 1 to this bit will update MIXOUTLVOL and
MIXOUTRVOL volumes simultaneously.
7 MIXOUTL_ZC 0 MIXOUTLVOL (Left Mixer Output PGA) Zero Cross
Enable
0 = Zero cross disabled
1 = Zero cross enabled
6 MIXOUTL_MU
TE_N
1 MIXOUTLVOL (Left Mixer Output PGA) Mute
0 = Mute
1 = Un-mute
[5:0]
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps) w PP, August 2012, Rev 3.4
272
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
11_1111 = +6dB
Register 20h Left OPGA Volume
REGISTER
ADDRESS
R33 (21h)
Right OPGA
Volume
DESCRIPTION
8 MIXOUT_VU 0 Mixer Output PGA Volume Update
Writing a 1 to this bit will update MIXOUTLVOL and
MIXOUTRVOL volumes simultaneously.
7 MIXOUTR_ZC 0 MIXOUTRVOL (Right Mixer Output PGA) Zero Cross
Enable
0 = Zero cross disabled
1 = Zero cross enabled
6 MIXOUTR_MU
TE_N
1 MIXOUTLVOL (Right Mixer Output PGA) Mute
0 = Mute
1 = Un-mute
5:0 MIXOUTR_VO
L [5:0]
11_1001 MIXOUTRVOL (Right Mixer Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
Register 21h Right OPGA Volume
REGISTER
ADDRESS
R34 (22h)
SPKMIXL
Attenuation
8 SPKAB_REF_
SEL
0
DESCRIPTION
6 DAC2L_SPKMI
XL_VOL
5 MIXINL_SPKMI
XL_VOL
4 IN1LP_SPKMI
XL_VOL
3 MIXOUTL_SPK
MIXL_VOL
2 DAC1L_SPKMI
XL_VOL
1:0 SPKMIXL_VOL
[1:0]
0
0
0
0
0
11
Selects Reference for Speaker in Class AB mode
0 = SPKVDD/2
1 = VMID
Left DAC2 to SPKMIXL Fine Volume Control
0 = 0dB
1 = -3dB
MIXINL (Left ADC bypass) to SPKMIXL Fine Volume
Control
0 = 0dB
1 = -3dB
IN1LP to SPKMIXL Fine Volume Control
0 = 0dB
1 = -3dB
Left Mixer Output to SPKMIXL Fine Volume Control
0 = 0dB
1 = -3dB
Left DAC1 to SPKMIXL Fine Volume Control
0 = 0dB
1 = -3dB
Left Speaker Mixer Volume Control
00 = 0dB
01 = -6dB
10 = -12dB
11 = Mute
Register 22h SPKMIXL Attenuation
WM8958 w PP, August 2012, Rev 3.4
273
WM8958
REGISTER
ADDRESS
R35 (23h)
SPKMIXR
Attenuation
8 SPKOUT_CLA
SSAB
6 DAC2R_SPKM
IXR_VOL
5 MIXINR_SPKM
IXR_VOL
4 IN1RP_SPKMI
XR_VOL
3 MIXOUTR_SP
KMIXR_VOL
2 DAC1R_SPKM
IXR_VOL
1:0 SPKMIXR_VO
L [1:0]
Pre-Production
0
0
0
0
0
0
11
DESCRIPTION
Speaker Class AB Mode Enable
0 = Class D mode
1 = Class AB mode
Right DAC2 to SPKMIXR Fine Volume Control
0 = 0dB
1 = -3dB
MIXINR (Right ADC bypass) to SPKMIXR Fine Volume
Control
0 = 0dB
1 = -3dB
IN1RP to SPKMIXR Fine Volume Control
0 = 0dB
1 = -3dB
Right Mixer Output to SPKMIXR Fine Volume Control
0 = 0dB
1 = -3dB
Right DAC1 to SPKMIXR Fine Volume Control
0 = 0dB
1 = -3dB
Right Speaker Mixer Volume Control
00 = 0dB
01 = -6dB
10 = -12dB
11 = Mute
Register 23h SPKMIXR Attenuation
REGISTER
ADDRESS
R36 (24h)
SPKOUT
Mixers
5 IN2LRP_TO_S
PKOUTL
4 SPKMIXL_TO_
SPKOUTL
3 SPKMIXR_TO_
SPKOUTL
2 IN2LRP_TO_S
PKOUTR
1 SPKMIXL_TO_
SPKOUTR
0 SPKMIXR_TO_
SPKOUTR
Register 24h SPKOUT Mixers
0
1
0
0
0
1
DESCRIPTION
Direct Voice (VRXN-VRXP) to Left Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXL Left Speaker Mixer to Left Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXR Right Speaker Mixer to Left Speaker Mute
0 = Mute
1 = Un-mute
Direct Voice (VRXN-VRXP) to Right Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXL Left Speaker Mixer to Right Speaker Mute
0 = Mute
1 = Un-mute
SPKMIXR Right Speaker Mixer to Right Speaker Mute
0 = Mute
1 = Un-mute w PP, August 2012, Rev 3.4
274
Pre-Production
REGISTER
ADDRESS
R37 (25h)
ClassD
5:3 SPKOUTL_BO
OST [2:0]
DESCRIPTION
2:0 SPKOUTR_BO
OST [2:0]
000
000
Left Speaker Gain Boost
000 = 1.00x boost (+0dB)
001 = 1.19x boost (+1.5dB)
010 = 1.41x boost (+3.0dB)
011 = 1.68x boost (+4.5dB)
100 = 2.00x boost (+6.0dB)
101 = 2.37x boost (+7.5dB)
110 = 2.81x boost (+9.0dB)
111 = 3.98x boost (+12.0dB)
Right Speaker Gain Boost
000 = 1.00x boost (+0dB)
001 = 1.19x boost (+1.5dB)
010 = 1.41x boost (+3.0dB)
011 = 1.68x boost (+4.5dB)
100 = 2.00x boost (+6.0dB)
101 = 2.37x boost (+7.5dB)
110 = 2.81x boost (+9.0dB)
111 = 3.98x boost (+12.0dB)
Register 25h ClassD
REGISTER
ADDRESS
R38 (26h)
Speaker
Volume Left
DESCRIPTION
8 SPKOUT_VU 0 Speaker Output PGA Volume Update
Writing a 1 to this bit will update SPKLVOL and
SPKRVOL volumes simultaneously.
7 SPKOUTL_ZC 0 SPKLVOL (Left Speaker Output PGA) Zero Cross
Enable
0 = Zero cross disabled
6 SPKOUTL_MU
TE_N
1
1 = Zero cross enabled
SPKLVOL (Left Speaker Output PGA) Mute
0 = Mute
5:0 SPKOUTL_VO
L [5:0]
1 = Un-mute
11_1001 SPKLVOL (Left Speaker Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
Register 26h Speaker Volume Left
WM8958
REGISTER
ADDRESS
R39 (27h)
Speaker
Volume
Right
DESCRIPTION
8 SPKOUT_VU 0 Speaker Output PGA Volume Update
Writing a 1 to this bit will update SPKLVOL and
SPKRVOL volumes simultaneously.
7 SPKOUTR_ZC 0 SPKRVOL (Right Speaker Output PGA) Zero Cross
Enable
0 = Zero cross disabled
1 = Zero cross enabled
6 SPKOUTR_MU
TE_N
1 SPKRVOL (Right Speaker Output PGA) Mute
0 = Mute
1 = Un-mute w PP, August 2012, Rev 3.4
275
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
5:0 SPKOUTR_VO
L [5:0]
11_1001 SPKRVOL (Right Speaker Output PGA) Volume
-57dB to +6dB in 1dB steps
00_0000 = -57dB
00_0001 = -56dB
… (1dB steps)
11_1111 = +6dB
Register 27h Speaker Volume Right
REGISTER
ADDRESS
R40 (28h)
Input Mixer
(2)
7 IN2LP_TO_IN2
L
0
DESCRIPTION
6 IN2LN_TO_IN2
L
5 IN1LP_TO_IN1
L
4 IN1LN_TO_IN1
L
3 IN2RP_TO_IN2
R
2 IN2RN_TO_IN2
R
1 IN1RP_TO_IN1
R
0 IN1RN_TO_IN1
R
0
0
0
0
0
0
0
IN2L PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN2LP
Note that VMID_BUF_ENA must be set when using
IN2L connected to VMID.
IN2L PGA Inverting Input Select
0 = Not connected
1 = Connected to IN2LN
IN1L PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN1LP
Note that VMID_BUF_ENA must be set when using
IN1L connected to VMID.
IN1L PGA Inverting Input Select
0 = Not connected
1 = Connected to IN1LN
IN2R PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN2RP
Note that VMID_BUF_ENA must be set when using
IN2R connected to VMID.
IN2R PGA Inverting Input Select
0 = Not connected
1 = Connected to IN2RN
IN1R PGA Non-Inverting Input Select
0 = Connected to VMID
1 = Connected to IN1RP
Note that VMID_BUF_ENA must be set when using
IN1R connected to VMID.
IN1R PGA Inverting Input Select
0 = Not connected
1 = Connected to IN1RN
Register 28h Input Mixer (2) w PP, August 2012, Rev 3.4
276
Pre-Production
REGISTER
ADDRESS
R41 (29h)
Input Mixer
(3)
8 IN2L_TO_MIXI
NL
7 IN2L_MIXINL_
VOL
5 IN1L_TO_MIXI
NL
4 IN1L_MIXINL_
VOL
2:0 MIXOUTL_MIX
INL_VOL [2:0]
Register 29h Input Mixer (3)
REGISTER
ADDRESS
R42 (2Ah)
Input Mixer
(4)
8 IN2R_TO_MIXI
NR
7 IN2R_MIXINR_
VOL
5 IN1R_TO_MIXI
NR
4 IN1R_MIXINR_
VOL
2:0 MIXOUTR_MIX
INR_VOL [2:0]
0
0
0
0
000
DESCRIPTION
IN2R PGA Output to MIXINR Mute
0 = Mute
1 = Un-Mute
IN2R PGA Output to MIXINR Gain
0 = 0dB
1 = +30dB
IN1R PGA Output to MIXINR Mute
0 = Mute
1 = Un-Mute
IN1R PGA Output to MIXINR Gain
0 = 0dB
1 = +30dB
Record Path MIXOUTR to MIXINR Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB
Register 2Ah Input Mixer (4)
0
0
0
0
000
DESCRIPTION
IN2L PGA Output to MIXINL Mute
0 = Mute
1 = Un-Mute
IN2L PGA Output to MIXINL Gain
0 = 0dB
1 = +30dB
IN1L PGA Output to MIXINL Mute
0 = Mute
1 = Un-Mute
IN1L PGA Output to MIXINL Gain
0 = 0dB
1 = +30dB
Record Path MIXOUTL to MIXINL Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB
WM8958 w PP, August 2012, Rev 3.4
277
WM8958
REGISTER
ADDRESS
R43 (2Bh)
Input Mixer
(5)
8:6 IN1LP_MIXINL
_VOL [2:0]
2:0 IN2LRP_MIXIN
L_VOL [2:0]
Register 2Bh Input Mixer (5)
REGISTER
ADDRESS
R44 (2Ch)
Input Mixer
(6)
8:6 IN1RP_MIXINR
_VOL [2:0]
2:0 IN2LRP_MIXIN
R_VOL [2:0]
000
000
DESCRIPTION
IN1RP Pin (PGA Bypass) to MIXINR Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB (see note below).
When IN1RP_MIXINR_BOOST is set, then the maximum gain setting is increased to +15dB, ie. 111 =
+15dB.
Note that VMID_BUF_ENA must be set when using the
IN1RP (PGA Bypass) input to MIXINR.
RXVOICE Differential Input (VRXP-VRXN) to MIXINR
Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB
Register 2Ch Input Mixer (6)
000
000
Pre-Production
DESCRIPTION
IN1LP Pin (PGA Bypass) to MIXINL Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB (see note below).
When IN1LP_MIXINL_BOOST is set, then the maximum gain setting is increased to +15dB, ie. 111 =
+15dB.
Note that VMID_BUF_ENA must be set when using the
IN1LP (PGA Bypass) input to MIXINL.
RXVOICE Differential Input (VRXP-VRXN) to MIXINL
Gain and Mute
000 = Mute
001 = -12dB
010 = -9dB
011 = -6dB
100 = -3dB
101 = 0dB
110 = +3dB
111 = +6dB w PP, August 2012, Rev 3.4
278
Pre-Production
REGISTER
ADDRESS
R45 (2Dh)
Output Mixer
(1)
8 DAC1L_TO_H
POUT1L
7 MIXINR_TO_M
IXOUTL
6 MIXINL_TO_MI
XOUTL
5 IN2RN_TO_MI
XOUTL
4 IN2LN_TO_MI
XOUTL
3 IN1R_TO_MIX
OUTL
2 IN1L_TO_MIX
OUTL
1 IN2LP_TO_MI
XOUTL
0
0
0
0
0
0
0
0
0
DESCRIPTION
HPOUT1LVOL (Left Headphone Output PGA) Input
Select
0 = MIXOUTL
1 = DAC1L
MIXINR Output (Right ADC bypass) to MIXOUTL Mute
0 = Mute
1 = Un-mute
MIXINL Output (Left ADC bypass) to MIXOUTL Mute
0 = Mute
1 = Un-mute
IN2RN to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2RN input to MIXOUTL.
IN2LN to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2LN input to MIXOUTL.
IN1R PGA Output to MIXOUTL Mute
0 = Mute
1 = Un-mute
IN1L PGA Output to MIXOUTL Mute
0 = Mute
1 = Un-mute
IN2LP to MIXOUTL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2LP input to MIXOUTL.
Left DAC1 to MIXOUTL Mute
0 = Mute
1 = Un-mute
WM8958
0 DAC1L_TO_MI
XOUTL
Register 2Dh Output Mixer (1)
REGISTER
ADDRESS
R46 (2Eh)
Output Mixer
(2)
8 DAC1R_TO_H
POUT1R
7 MIXINL_TO_MI
XOUTR
6 MIXINR_TO_M
IXOUTR
0
0
0
DESCRIPTION
HPOUT1RVOL (Right Headphone Output PGA) Input
Select
0 = MIXOUTR
1 = DAC1R
MIXINL Output (Left ADC bypass) to MIXOUTR Mute
0 = Mute
1 = Un-mute
MIXINR Output (Right ADC bypass) to MIXOUTR Mute
0 = Mute
1 = Un-mute w PP, August 2012, Rev 3.4
279
WM8958
REGISTER
ADDRESS
5 IN2LN_TO_MI
XOUTR
4 IN2RN_TO_MI
XOUTR
3 IN1L_TO_MIX
OUTR
2 IN1R_TO_MIX
OUTR
1 IN2RP_TO_MI
XOUTR
0 DAC1R_TO_MI
XOUTR
0
0
0
0
0
0
Pre-Production
DESCRIPTION
IN2LN to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2LN input to MIXOUTR.
IN2RN to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2RN input to MIXOUTR.
IN1L PGA Output to MIXOUTR Mute
0 = Mute
1 = Un-mute
IN1R PGA Output to MIXOUTR Mute
0 = Mute
1 = Un-mute
IN2RP to MIXOUTR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN2RP input to MIXOUTR.
Right DAC1 to MIXOUTR Mute
0 = Mute
1 = Un-mute
Register 2Eh Output Mixer (2)
REGISTER
ADDRESS
R47 (2Fh)
Output Mixer
(3)
11:9 IN2LP_MIXOU
TL_VOL [2:0]
8:6 IN2LN_MIXOU
TL_VOL [2:0]
5:3 IN1R_MIXOUT
L_VOL [2:0]
2:0 IN1L_MIXOUT
L_VOL [2:0]
000
000
000
000
DESCRIPTION
IN2LP to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2LN to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1R PGA Output to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1L PGA Output to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Register 2Fh Output Mixer (3) w PP, August 2012, Rev 3.4
280
Pre-Production
REGISTER
ADDRESS
R48 (30h)
Output Mixer
(4)
11:9 IN2RP_MIXOU
TR_VOL [2:0]
8:6 IN2RN_MIXOU
TR_VOL [2:0]
5:3 IN1L_MIXOUT
R_VOL [2:0]
2:0 IN1R_MIXOUT
R_VOL [2:0]
000
000
000
000
DESCRIPTION
IN2RP to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2RN to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1L PGA Output to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN1R PGA Output to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Register 30h Output Mixer (4)
REGISTER
ADDRESS
R49 (31h)
Output Mixer
(5)
11:9 DAC1L_MIXO
UTL_VOL [2:0]
8:6 IN2RN_MIXOU
TL_VOL [2:0]
5:3 MIXINR_MIXO
UTL_VOL [2:0]
000
000
000
DESCRIPTION
Left DAC1 to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2RN to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINR Output (Right ADC bypass) to MIXOUTL
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
WM8958 w PP, August 2012, Rev 3.4
281
WM8958
REGISTER
ADDRESS
2:0 MIXINL_MIXO
UTL_VOL [2:0]
000
Pre-Production
DESCRIPTION
MIXINL Output (Left ADC bypass) to MIXOUTL Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Register 31h Output Mixer (5)
REGISTER
ADDRESS
R50 (32h)
Output Mixer
(6)
11:9 DAC1R_MIXO
UTR_VOL [2:0]
8:6 IN2LN_MIXOU
TR_VOL [2:0]
5:3 MIXINL_MIXO
UTR_VOL [2:0]
2:0 MIXINR_MIXO
UTR_VOL [2:0]
000
000
000
000
DESCRIPTION
Right DAC1 to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
IN2LN to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINL Output (Left ADC bypass) to MIXOUTR Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
MIXINR Output (Right ADC bypass) to MIXOUTR
Volume
0dB to -9dB in 3dB steps
X00 = 0dB
X01 = -3dB
X10 = -6dB
X11 = -9dB
Register 32h Output Mixer (6)
REGISTER
ADDRESS
R51 (33h)
HPOUT2
Mixer
5 IN2LRP_TO_H
POUT2
4 MIXOUTLVOL_
TO_HPOUT2
3 MIXOUTRVOL
_TO_HPOUT2
0
0
0
DESCRIPTION
Direct Voice (VRXN-VRXP) to Earpiece Driver
0 = Mute
1 = Un-mute
MIXOUTLVOL (Left Output Mixer PGA) to Earpiece
Driver
0 = Mute
1 = Un-mute
MIXOUTRVOL (Right Output Mixer PGA) to Earpiece
Driver
0 = Mute
1 = Un-mute
Register 33h HPOUT2 Mixer w PP, August 2012, Rev 3.4
282
Pre-Production
REGISTER
ADDRESS
R52 (34h)
Line Mixer
(1)
6 MIXOUTL_TO_
LINEOUT1N
5 MIXOUTR_TO
_LINEOUT1N
4 LINEOUT1_MO
DE
2 IN1R_TO_LINE
OUT1P
1 IN1L_TO_LINE
OUT1P
0 MIXOUTL_TO_
LINEOUT1P
0
0
0
0
0
0
DESCRIPTION
MIXOUTL to Single-Ended Line Output on LINEOUT1N
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 1)
MIXOUTR to Single-Ended Line Output on LINEOUT1N
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 1)
LINEOUT1 Mode Select
0 = Differential
1 = Single-Ended
IN1R Input PGA to Differential Line Output on
LINEOUT1
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 0)
IN1L Input PGA to Differential Line Output on
LINEOUT1
0 = Mute
1 = Un-mute
(LINEOUT1_MODE = 0)
Differential Mode (LINEOUT1_MODE = 0):
MIXOUTL to Differential Output on LINEOUT1
0 = Mute
1 = Un-mute
Single Ended Mode (LINEOUT1_MODE = 1):
MIXOUTL to Single-Ended Line Output on LINEOUT1P
0 = Mute
1 = Un-mute
WM8958
Register 34h Line Mixer (1)
REGISTER
ADDRESS
R53 (35h)
Line Mixer
(2)
6 MIXOUTR_TO
_LINEOUT2N
5 MIXOUTL_TO_
LINEOUT2N
4 LINEOUT2_MO
DE
2 IN1L_TO_LINE
OUT2P
0
0
0
0
DESCRIPTION
MIXOUTR to Single-Ended Line Output on LINEOUT2N
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 1)
MIXOUTL to Single-Ended Line Output on LINEOUT2N
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 1)
LINEOUT2 Mode Select
0 = Differential
1 = Single-Ended
IN1L Input PGA to Differential Line Output on
LINEOUT2
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 0) w PP, August 2012, Rev 3.4
283
WM8958
REGISTER
ADDRESS
1 IN1R_TO_LINE
OUT2P
0 MIXOUTR_TO
_LINEOUT2P
0
0
Pre-Production
DESCRIPTION
IN1R Input PGA to Differential Line Output on
LINEOUT2
0 = Mute
1 = Un-mute
(LINEOUT2_MODE = 0)
Differential Mode (LINEOUT2_MODE = 0):
MIXOUTR to Differential Output on LINEOUT2
0 = Mute
1 = Un-mute
Single-Ended Mode (LINEOUT2_MODE = 0):
MIXOUTR to Single-Ended Line Output on LINEOUT2P
0 = Mute
1 = Un-mute
Register 35h Line Mixer (2)
REGISTER
ADDRESS
R54 (36h)
Speaker
Mixer
9 DAC2L_TO_S
PKMIXL
8 DAC2R_TO_S
PKMIXR
7 MIXINL_TO_S
PKMIXL
6 MIXINR_TO_S
PKMIXR
5 IN1LP_TO_SP
KMIXL
4 IN1RP_TO_SP
KMIXR
3 MIXOUTL_TO_
SPKMIXL
2 MIXOUTR_TO
_SPKMIXR
1 DAC1L_TO_S
PKMIXL
0 DAC1R_TO_S
PKMIXR
Register 36h Speaker Mixer w
0
0
0
0
0
0
0
0
0
0
DESCRIPTION
Left DAC2 to SPKMIXL Mute
0 = Mute
1 = Un-mute
Right DAC2 to SPKMIXR Mute
0 = Mute
1 = Un-mute
MIXINL (Left ADC bypass) to SPKMIXL Mute
0 = Mute
1 = Un-mute
MIXINR (Right ADC bypass) to SPKMIXR Mute
0 = Mute
1 = Un-mute
IN1LP to SPKMIXL Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN1LP input to SPKMIXL.
IN1RP to SPKMIXR Mute
0 = Mute
1 = Un-mute
Note that VMID_BUF_ENA must be set when using the
IN1RP input to SPKMIXR.
Left Mixer Output to SPKMIXL Mute
0 = Mute
1 = Un-mute
Right Mixer Output to SPKMIXR Mute
0 = Mute
1 = Un-mute
Left DAC1 to SPKMIXL Mute
0 = Mute
1 = Un-mute
Right DAC1 to SPKMIXR Mute
0 = Mute
1 = Un-mute
PP, August 2012, Rev 3.4
284
Pre-Production
REGISTER
ADDRESS
R55 (37h)
Additional
Control
DESCRIPTION
7 LINEOUT1_FB 0 Enable ground loop noise feedback on LINEOUT1
0 = Disabled
1 = Enabled
6 LINEOUT2_FB 0 Enable ground loop noise feedback on LINEOUT2
0 = Disabled
1 = Enabled
(Disabled Outputs)
0 = 20k from buffered VMID to output
1 = 500 from buffered VMID to output
Register 37h Additional Control
REGISTER
ADDRESS
R56 (38h)
AntiPOP (1)
7 LINEOUT_VMI
D_BUF_ENA
0
DESCRIPTION
6 HPOUT2_IN_E
NA
5 LINEOUT1_DI
SCH
4 LINEOUT2_DI
SCH
0
0
0
Enables VMID reference for line outputs in singleended mode
0 = Disabled
1 = Enabled
HPOUT2MIX Mixer and Input Stage Enable
0 = Disabled
1 = Enabled
Discharges LINEOUT1P and LINEOUT1N outputs
0 = Not active
1 = Actively discharging LINEOUT1P and LINEOUT1N
Discharges LINEOUT2P and LINEOUT2N outputs
0 = Not active
1 = Actively discharging LINEOUT2P and LINEOUT2N
Register 38h AntiPOP (1)
WM8958
REGISTER
ADDRESS
R57 (39h)
AntiPOP (2)
DESCRIPTION
6:5 VMID_RAMP
[1:0]
3 VMID_BUF_EN
A
2 STARTUP_BIA
S_ENA
00
0
0
VMID soft start enable / slew rate control
00 = Normal slow start
01 = Normal fast start
10 = Soft slow start
11 = Soft fast start
If VMID_RAMP = 1X is selected for VMID start-up or shut-down, then the soft-start circuit must be reset by setting VMID_RAMP=00 after VMID is disabled, before
VMID is re-enabled. VMID is disabled / enabled using the VMID_SEL register.
VMID Buffer Enable
0 = Disabled
1 = Enabled (provided VMID_SEL > 00)
Enables the Start-Up bias current generator
0 = Disabled
1 = Enabled
1 BIAS_SRC 0 Selects the bias current source
0 = Normal bias
1 = Start-Up bias
0 VMID_DISCH 0 Connects VMID to ground w PP, August 2012, Rev 3.4
285
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
0 = Disabled
1 = Enabled
Register 39h AntiPOP (2)
REGISTER
ADDRESS
R59 (3Bh)
LDO 1
DESCRIPTION
3:1 LDO1_VSEL
[2:0]
110 LDO1 Output Voltage Select
2.4V to 3.1V in 100mV steps
000 = 2.4V
001 = 2.5V
010 = 2.6V
011 = 2.7V
100 = 2.8V
101 = 2.9V
110 = 3.0V
111 = 3.1V
0 LDO1_DISCH 1 LDO1 Discharge Select
0 = LDO1 floating when disabled
1 = LDO1 discharged when disabled
Register 3Bh LDO 1
REGISTER
ADDRESS
R60 (3Ch)
LDO 2
DESCRIPTION
2:1 LDO2_VSEL
[1:0]
10 LDO2 Output Voltage Select
1.1V to 1.3V in 100mV steps
00 = Reserved
01 = 1.1V
10 = 1.2V
11 = 1.3V
0 LDO2_DISCH 1 LDO2 Discharge Select
0 = LDO2 floating when disabled
1 = LDO2 discharged when disabled
Register 3Ch LDO 2
REGISTER
ADDRESS
R61 (3Dh)
MICBIAS1
DESCRIPTION
5 MICB1_RATE 1 Microphone Bias 1 Rate
0 = Fast start-up / shut-down
1 = Pop-free start-up / shut-down
4 MICB1_MODE 1 Microphone Bias 1 Mode
0 = Regulator mode
1 = Bypass mode w PP, August 2012, Rev 3.4
286
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
3:1 MICB1_LVL
[2:0]
100 Microphone Bias 1 Voltage Control
(when MICB1_MODE = 0)
000 = 1.5V
001 = 1.8V
010 = 1.9V
011 = 2.0V
100 = 2.2V
101 = 2.4V
110 = 2.5V
111 = 2.6V
0 MICB1_DISCH 1 Microphone Bias 1 Discharge
0 = MICBIAS1 floating when disabled
1 = MICBIAS1 discharged when disabled
Register 3Dh MICBIAS1
REGISTER
ADDRESS
R62 (3Eh)
MICBIAS2
DESCRIPTION
5 MICB2_RATE 1 Microphone Bias 2 Rate
0 = Fast start-up / shut-down
1 = Pop-free start-up / shut-down
4 MICB2_MODE 1 Microphone Bias 2 Mode
0 = Regulator mode
1 = Bypass mode
3:1 MICB2_LVL
[2:0]
100 Microphone Bias 2 Voltage Control
(when MICB2_MODE = 0)
000 = 1.5V
001 = 1.8V
010 = 1.9V
011 = 2.0V
100 = 2.2V
101 = 2.4V
110 = 2.5V
111 = 2.6V
0 MICB2_DISCH 1 Microphone Bias 2 Discharge
0 = MICBIAS2 floating when disabled
1 = MICBIAS2 discharged when disabled
Register 3Eh MICBIAS2
REGISTER
ADDRESS
R76 (4Ch)
Charge
Pump (1)
DESCRIPTION
15 CP_ENA 0 Enable charge-pump digits
0 = Disable
1 = Enable
Register 4Ch Charge Pump (1)
WM8958
REGISTER
ADDRESS
R77 (4Dh)
Charge
Pump (2)
DESCRIPTION
15 CP_DISCH 1 Charge Pump Discharge Select
0 = Charge Pump outputs floating when disabled
1 = Charge Pump outputs discharged when disabled
Register 4Dh Charge Pump (2) w PP, August 2012, Rev 3.4
287
WM8958
Pre-Production
REGISTER
ADDRESS
R81 (51h)
Class W (1)
DESCRIPTION
9:8 CP_DYN_SRC
_SEL [1:0]
00 Selects the digital audio source for envelope tracking
00 = AIF1, DAC Timeslot 0
01 = AIF1, DAC Timeslot 1
10 = AIF2, DAC data
11 = Reserved
0 CP_DYN_PWR 0 Enable dynamic charge pump power control
0 = charge pump controlled by volume register settings
(Class G)
1 = charge pump controlled by real-time audio level
(Class W)
Register 51h Class W (1)
REGISTER
ADDRESS
R84 (54h)
DC Servo
(1)
13 DCS_TRIG_SI
NGLE_1
0
DESCRIPTION
12 DCS_TRIG_SI
NGLE_0
9 DCS_TRIG_SE
RIES_1
8 DCS_TRIG_SE
RIES_0
5 DCS_TRIG_ST
ARTUP_1
4 DCS_TRIG_ST
ARTUP_0
3 DCS_TRIG_DA
C_WR_1
2 DCS_TRIG_DA
C_WR_0
1 DCS_ENA_CH
AN_1
0 DCS_ENA_CH
AN_0
0
0
0
0
0
0
0
0
0
Writing 1 to this bit selects a single DC offset correction for HPOUT1R.
In readback, a value of 1 indicates that the DC Servo single correction is in progress.
Writing 1 to this bit selects a single DC offset correction for HPOUT1L.
In readback, a value of 1 indicates that the DC Servo single correction is in progress.
Writing 1 to this bit selects a series of DC offset corrections for HPOUT1R.
In readback, a value of 1 indicates that the DC Servo
DAC Write correction is in progress.
Writing 1 to this bit selects a series of DC offset corrections for HPOUT1L.
In readback, a value of 1 indicates that the DC Servo
DAC Write correction is in progress.
Writing 1 to this bit selects Start-Up DC Servo mode for
HPOUT1R.
In readback, a value of 1 indicates that the DC Servo
Start-Up correction is in progress.
Writing 1 to this bit selects Start-Up DC Servo mode for
HPOUT1L.
In readback, a value of 1 indicates that the DC Servo
Start-Up correction is in progress.
Writing 1 to this bit selects DAC Write DC Servo mode for HPOUT1R.
In readback, a value of 1 indicates that the DC Servo
DAC Write correction is in progress.
Writing 1 to this bit selects DAC Write DC Servo mode for HPOUT1L.
In readback, a value of 1 indicates that the DC Servo
DAC Write correction is in progress.
DC Servo enable for HPOUT1R
0 = Disabled
1 = Enabled
DC Servo enable for HPOUT1L
0 = Disabled
1 = Enabled
Register 54h DC Servo (1) w PP, August 2012, Rev 3.4
288
Pre-Production
REGISTER
ADDRESS
R85 (55h)
DC Servo
(2)
DESCRIPTION
11:5 DCS_SERIES_
NO_01 [6:0]
3:0 DCS_TIMER_P
ERIOD_01
[3:0]
010_1010 Number of DC Servo updates to perform in a series event.
0 = 1 update
1 = 2 updates
...
127 = 128 updates
1010 Time between periodic updates. Time is calculated as
0.251s x (2^PERIOD), where PERIOD = DCS_TIMER_PERIOD_01.
0000 = Off
0001 = 0.502s
….
1010 = 257s (4min 17s)
1111 = 8225s (2hr 17min)
Register 55h DC Servo (2)
REGISTER
ADDRESS
R87 (57h)
DC Servo
(4)
DESCRIPTION
15:8 DCS_DAC_WR
_VAL_1 [7:0]
7:0 DCS_DAC_WR
_VAL_0 [7:0]
0000_0000 Writing to this field sets the DC Offset value for
HPOUT1R in DAC Write DC Servo mode.
Reading this field gives the current DC Offset value for
HPOUT1R.
Two’s complement format.
LSB is 0.25mV.
Range is -32mV to +31.75mV
0000_0000 Writing to this field sets the DC Offset value for
HPOUT1L in DAC Write DC Servo mode.
Reading this field gives the current DC Offset value for
HPOUT1L.
Two’s complement format.
LSB is 0.25mV.
Range is -32mV to +31.75mV
Register 57h DC Servo (4)
REGISTER
ADDRESS
R88 (58h)
DC Servo
Readback
9:8 DCS_CAL_CO
MPLETE [1:0]
00
DESCRIPTION
5:4 DCS_DAC_WR
_COMPLETE
[1:0]
1:0 DCS_STARTU
P_COMPLETE
[1:0]
00
00
DC Servo Complete status
0 = DAC Write or Start-Up DC Servo mode not completed.
1 = DAC Write or Start-Up DC Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
DC Servo DAC Write status
0 = DAC Write DC Servo mode not completed.
1 = DAC Write DC Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
DC Servo Start-Up status
0 = Start-Up DC Servo mode not completed.
1 = Start-Up DC Servo mode complete.
Bit [1] = HPOUT1R
Bit [0] = HPOUT1L
Register 58h DC Servo Readback w
WM8958
PP, August 2012, Rev 3.4
289
WM8958
REGISTER
ADDRESS
R96 (60h)
Analogue
HP (1)
7 HPOUT1L_RM
V_SHORT
6 HPOUT1L_OU
TP
5 HPOUT1L_DL
Y
3 HPOUT1R_RM
V_SHORT
2 HPOUT1R_OU
TP
1 HPOUT1R_DL
Y
0
0
0
0
0
0
Pre-Production
DESCRIPTION
Removes HPOUT1L short
0 = HPOUT1L short enabled
1 = HPOUT1L short removed
For normal operation, this bit should be set as the final step of the HPOUT1L Enable sequence.
Enables HPOUT1L output stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the DC offset cancellation has been scheduled.
Enables HPOUT1L intermediate stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the output signal path has been configured, and before
DC offset cancellation is scheduled. This bit should be set with at least 20us delay after HPOUT1L_ENA.
Removes HPOUT1R short
0 = HPOUT1R short enabled
1 = HPOUT1R short removed
For normal operation, this bit should be set as the final step of the HPOUT1R Enable sequence.
Enables HPOUT1R output stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the DC offset cancellation has been scheduled.
Enables HPOUT1R intermediate stage
0 = Disabled
1 = Enabled
For normal operation, this bit should be set to 1 after the output signal path has been configured, and before
DC offset cancellation is scheduled. This bit should be set with at least 20us delay after HPOUT1R_ENA.
Register 60h Analogue HP (1)
REGISTER
ADDRESS
R208 (D0h)
Mic Detect 1
15:12 MICD_BIAS_S
TARTTIME
[3:0] w
DESCRIPTION
0101 Mic Detect Bias Startup Delay
(If MICBIAS2 is not enabled already, this field selects the delay time allowed for MICBIAS2 to startup prior to performing the MICDET function.)
0000 = 0ms (continuous)
0001 = 0.25ms
0010 = 0.5ms
0011 = 1ms
0100 = 2ms
0101 = 4ms
0110 = 8ms
0111 = 16ms
1000 = 32ms
1001 = 64ms
1010 = 128ms
PP, August 2012, Rev 3.4
290
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
11:8 MICD_RATE
[3:0]
0110
1011 = 256ms
1100 to 1111 = 512ms
Mic Detect Rate
(Selects the delay between successive Mic Detect measurements.)
0000 = 0ms (continuous)
0001 = 0.25ms
0010 = 0.5ms
0011 = 1ms
0100 = 2ms
0101 = 4ms
0110 = 8ms
0111 = 16ms
1000 = 32ms
1001 = 64ms
1010 = 128ms
1011 = 256ms
1100 to 1111 = 512ms
1 MICD_DBTIME 0 Mic Detect De-bounce
0 = 2 measurements
1 = 4 measurements
0 MICD_ENA 0 Mic Detect Enable
0 = Disabled
1 = Enabled
Register D0h Mic Detect 1
REGISTER
ADDRESS
R209 (D1h)
Mic Detect 2
DESCRIPTION
7:0 MICD_LVL_SE
L [7:0]
0111_1111 Mic Detect Level Select
(enables Mic Detection in specific impedance ranges)
[7] = Not used - must be set to 0
[6] = Enable >475 ohm detection
[5] = Enable 326 ohm detection
[4] = Enable 152 ohm detection
[3] = Enable 77 ohm detection
[2] = Enable 47.6 ohm detection
[1] = Enable 29.4 ohm detection
[0] = Enable 14 ohm detection
Note that the impedance values quoted assume that a microphone (475ohm-30kohm) is also present on the
MICDET pin.
Register D1h Mic Detect 2
WM8958 w PP, August 2012, Rev 3.4
291
WM8958
Pre-Production
REGISTER
ADDRESS
R210 (D2h)
Mic Detect 3
DESCRIPTION
[8:0] 00
Mic Detect Level
(indicates the measured impedance)
[8] = Not used
[7] = >475 ohm, <30k ohm
[6] = 326 ohm
[5] = 152 ohm
[4] = 77 ohm
[3] = 47.6 ohm
[2] = 29.4 ohm
[1] = 14 ohm
[0] = <2 ohm
Note that the impedance values quoted assume that a microphone (475ohm-30kohm) is also present on the
MICDET pin.
1 MICD_VALID 0 Mic Detect Data Valid
0 = Not Valid
1 = Valid
0 MICD_STS 0 Mic Detect Status
0 = No Mic Accessory present (impedance is >30k ohm)
1 = Mic Accessory is present (impedance is <30k ohm)
Register D2h Mic Detect 3
REGISTER
ADDRESS
R256
(0100h)
Chip
Revision
3:0 CHIP_REV
[3:0]
Register 0100h Chip Revision
Chip revision
DESCRIPTION
REGISTER
ADDRESS
R257
(0101h)
Control
Interface
DESCRIPTION
2 AUTO_INC 1 Enables address auto-increment
Register 0101h Control Interface
0 = Disabled
1 = Enabled
REGISTER
ADDRESS
R272
(0110h)
Write
Sequencer
Ctrl (1)
DESCRIPTION
15 WSEQ_ENA 0 Write Sequencer Enable.
0 = Disabled
1 = Enabled
9 WSEQ_ABOR
T
0 Writing a 1 to this bit aborts the current sequence and returns control of the device back to the serial control interface.
8 WSEQ_START 0 Writing a 1 to this bit starts the write sequencer at the index location selected by WSEQ_START_INDEX. The sequence continues until it reaches an “End of sequence” flag. At the end of the sequence, this bit will be reset by the Write Sequencer. w PP, August 2012, Rev 3.4
292
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
6:0 WSEQ_START
_INDEX [6:0]
000_0000 Sequence Start Index. This field determines the memory location of the first command in the selected sequence. There are 127 Write Sequencer RAM addresses:
00h = WSEQ_ADDR0 (R12288)
01h = WSEQ_ADDR1 (R12292)
02h = WSEQ_ADDR2 (R12296)
….
7Fh = WSEQ_ADDR127 (R12796)
Register 0110h Write Sequencer Ctrl (1)
REGISTER
ADDRESS
R273
(0111h)
Write
Sequencer
Ctrl (2)
DESCRIPTION
8 WSEQ_BUSY 0 Sequencer Busy flag (Read Only).
0 = Sequencer idle
1 = Sequencer busy
Note: it is not possible to write to control registers via the control interface while the Sequencer is Busy.
6:0 WSEQ_CURR
ENT_INDEX
[6:0]
000_0000 Sequence Current Index. This indicates the memory location of the most recently accessed command in the write sequencer memory.
Coding is the same as WSEQ_START_INDEX.
Register 0111h Write Sequencer Ctrl (2)
REGISTER
ADDRESS
R512
(0200h)
AIF1
Clocking (1)
DESCRIPTION
4:3 AIF1CLK_SRC
[1:0]
00 AIF1CLK Source Select
00 = MCLK1
01 = MCLK2
10 = FLL1
11 = FLL2
2 AIF1CLK_INV 0 AIF1CLK Invert
0 = AIF1CLK not inverted
1 = AIF1CLK inverted
1 AIF1CLK_DIV 0 AIF1CLK Divider
0 = AIF1CLK
1 = AIF1CLK / 2
0 AIF1CLK_ENA 0 AIF1CLK Enable
0 = Disabled
1 = Enabled
Register 0200h AIF1 Clocking (1)
WM8958 w PP, August 2012, Rev 3.4
293
WM8958
REGISTER
ADDRESS
R513
(0201h)
AIF1
Clocking (2)
5:3 AIF1DAC_DIV
[2:0]
000
2:0 AIF1ADC_DIV
[2:0]
000
Pre-Production
DESCRIPTION
Selects the AIF1 input path sample rate relative to the
AIF1 output path sample rate.
This field should only be changed from default in modes where the AIF1 input path sample rate is slower than the AIF1 output path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF1 output path sample rate relative to the
AIF1 input path sample rate.
This field should only be changed from default in modes where the AIF1 output path sample rate is slower than the AIF1 input path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Register 0201h AIF1 Clocking (2)
REGISTER
ADDRESS
R516
(0204h)
AIF2
Clocking (1)
DESCRIPTION
4:3 AIF2CLK_SRC
[1:0]
00 AIF2CLK Source Select
00 = MCLK1
01 = MCLK2
10 = FLL1
11 = FLL2
2 AIF2CLK_INV 0 AIF2CLK Invert
0 = AIF2CLK not inverted
1 = AIF2CLK inverted
1 AIF2CLK_DIV 0 AIF2CLK Divider
0 = AIF2CLK
1 = AIF2CLK / 2
0 AIF2CLK_ENA 0 AIF2CLK Enable
0 = Disabled
1 = Enabled
Register 0204h AIF2 Clocking (1) w PP, August 2012, Rev 3.4
294
Pre-Production
REGISTER
ADDRESS
R517
(0205h)
AIF2
Clocking (2)
5:3 AIF2DAC_DIV
[2:0]
2:0 AIF2ADC_DIV
[2:0]
000
000
DESCRIPTION
Selects the AIF2 input path sample rate relative to the
AIF2 output path sample rate.
This field should only be changed from default in modes where the AIF2 input path sample rate is slower than the AIF2 output path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Selects the AIF2 output path sample rate relative to the
AIF2 input path sample rate.
This field should only be changed from default in modes where the AIF2 output path sample rate is slower than the AIF2 input path sample rate.
000 = Divide by 1
001 = Divide by 1.5
010 = Divide by 2
011 = Divide by 3
100 = Divide by 4
101 = Divide by 5.5
110 = Divide by 6
111 = Reserved
Register 0205h AIF2 Clocking (2)
REGISTER
ADDRESS
R520
(0208h)
Clocking (1)
DESCRIPTION
14 DSP2CLK_EN
A
0 MBC Processor Clock Enable
0 = Disabled
1 = Enabled
4 TOCLK_ENA 0 Slow Clock (TOCLK) Enable
0 = Disabled
3 AIF1DSPCLK_
ENA
2 AIF2DSPCLK_
ENA
1 SYSDSPCLK_
ENA
0
0
0
1 = Enabled
This clock is required for zero-cross timeout.
AIF1 Processing Clock Enable
0 = Disabled
1 = Enabled
AIF2 Processing Clock Enable
0 = Disabled
1 = Enabled
Digital Mixing Processor Clock Enable
0 = Disabled
1 = Enabled
0 SYSCLK_SRC 0 SYSCLK Source Select
0 = AIF1CLK
1 = AIF2CLK
Register 0208h Clocking (1)
WM8958 w PP, August 2012, Rev 3.4
295
WM8958
REGISTER
ADDRESS
R521
(0209h)
Clocking (2)
10:8 TOCLK_DIV
[2:0]
000
6:4 DBCLK_DIV
[2:0]
000
2:0 OPCLK_DIV
[2:0]
000
Pre-Production
DESCRIPTION
Slow Clock (TOCLK ) Divider
(Sets TOCLK rate relative to 256kHz.)
000 = Divide by 256 (1kHz)
001 = Divide by 512 (500Hz)
010 = Divide by 1024 (250Hz)
011 = Divide by 2048 (125Hz)
100 = Divide by 4096 (62.5Hz)
101 = Divide by 8192 (31.2Hz)
110 = Divide by 16384 (15.6Hz)
111 = Divide by 32768 (7.8Hz)
De-bounce Clock (DBCLK) Divider
(Sets DBCLK rate relative to 256kHz.)
000 = Divide by 256 (1kHz)
001 = Divide by 2048 (125Hz)
010 = Divide by 4096 (62.5Hz)
011 = Divide by 8192 (31.2Hz)
100 = Divide by 16384 (15.6Hz)
101 = Divide by 32768 (7.8Hz)
110 = Divide by 65536 (3.9Hz)
111 = Divide by 131072 (1.95Hz)
GPIO Output Clock (OPCLK) Divider
000 = SYSCLK
001 = SYSCLK / 2
010 = SYSCLK / 3
011 = SYSCLK / 4
100 = SYSCLK / 6
101 = SYSCLK / 8
110 = SYSCLK / 12
111 = SYSCLK / 16
Register 0209h Clocking (2)
REGISTER
ADDRESS
R528
(0210h)
AIF1 Rate
DESCRIPTION
0000 = 8kHz
0001 = 11.025kHz
0010 = 12kHz
0011 = 16kHz
0100 = 22.05kHz
0101 = 24kHz
0110 = 32kHz
0111 = 44.1kHz
1000 = 48kHz
1001 = 88.2kHz
1010 = 96kHz
All other codes = Reserved
Note that 88.2kHz and 96kHz modes are supported for
AIF1 input (DAC playback) only. w PP, August 2012, Rev 3.4
296
Pre-Production
REGISTER
ADDRESS
3:0 AIF1CLK_RAT
E [3:0]
DESCRIPTION
0011 Selects the AIF1CLK / fs ratio
0000 = Reserved
0001 = 128
0010 = 192
0011 = 256
0100 = 384
0101 = 512
0110 = 768
0111 = 1024
1000 = 1408
1001 = 1536
All other codes = Reserved
Register 0210h AIF1 Rate
REGISTER
ADDRESS
R529
(0211h)
AIF2 Rate
3:0 AIF2CLK_RAT
E [3:0]
DESCRIPTION
0011
0000 = 8kHz
0001 = 11.025kHz
0010 = 12kHz
0011 = 16kHz
0100 = 22.05kHz
0101 = 24kHz
0110 = 32kHz
0111 = 44.1kHz
1000 = 48kHz
1001 = 88.2kHz
1010 = 96kHz
All other codes = Reserved
Note that 88.2kHz and 96kHz modes are supported for
AIF2 input (DAC playback) only.
Selects the AIF2CLK / fs ratio
0000 = Reserved
0001 = 128
0010 = 192
0011 = 256
0100 = 384
0101 = 512
0110 = 768
0111 = 1024
1000 = 1408
1001 = 1536
All other codes = Reserved
Register 0211h AIF2 Rate
WM8958 w PP, August 2012, Rev 3.4
297
WM8958
REGISTER
ADDRESS
R530
(0212h)
Rate Status
3:0 SR_ERROR
[3:0]
0000
Pre-Production
DESCRIPTION
Sample Rate Configuration status
Indicates an error with the register settings related to sample rate configuration
0000 = No errors
0001 = Invalid sample rate
0010 = Invalid AIF divide
0011 = ADC and DAC divides both set in an interface
0100 = Invalid combination of AIF divides and samplerate
0101 = Invalid set of enables for 96kHz mode
0110 = Invalid SYSCLK rate (derived from
AIF1CLK_RATE or AIF2CLK_RATE)
0111 = Mixed ADC and DAC rates in SYSCLK AIF when AIFs are asynchronous
1000 = Invalid combination of sample rates when both
AIFs are from the same clock source
1001 = Invalid combination of mixed ADC/DAC AIFs when both from the same clock source
1010 = AIF1DAC2 (Timeslot 1) ports enabled when
SRCs connected to AIF1
Register 0212h Rate Status
REGISTER
ADDRESS
R544
(0220h)
FLL1
Control (1)
DESCRIPTION
1 FLL1_OSC_EN
A
0 FLL1 Oscillator enable
0 = Disabled
1 = Enabled
(Note that this field is required for free-running FLL1 modes only)
0 FLL1_ENA 0 FLL1 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the FLL1 enable sequence, ie. after the other FLL registers have been configured.
Register 0220h FLL1 Control (1)
REGISTER
ADDRESS
R545
(0221h)
FLL1
Control (2)
13:8 FLL1_OUTDIV
[5:0]
DESCRIPTION
00_0000 FLL1 FOUT clock divider
000000 = Reserved
000001 = Reserved
000010 = Reserved
000011 = 4
000100 = 5
000101 = 6
…
111110 = 63
111111 = 64
(FOUT = FVCO / FLL1_OUTDIV) w PP, August 2012, Rev 3.4
298
Pre-Production
REGISTER
ADDRESS
2:0 FLL1_FRATIO
[2:0]
000
DESCRIPTION
WM8958
FLL1 FVCO clock divider
000 = 1
001 = 2
010 = 4
011 = 8
1XX = 16
Register 0221h FLL1 Control (2)
REGISTER
ADDRESS
R546
(0222h)
FLL1
Control (3)
Register 0222h FLL1 Control (3)
REGISTER
ADDRESS
R547
(0223h)
FLL1
Control (4)
Register 0223h FLL1 Control (4)
_0000_000
0
FLL1 Fractional multiply for FREF
(MSB = 0.5)
000
DESCRIPTION
DESCRIPTION
FLL1 Integer multiply for FREF
(LSB = 1)
REGISTER
ADDRESS
R548
(0224h)
FLL1
Control (5)
DESCRIPTION
15 FLL1_BYP 0 FLL1 Bypass Select
0 = Disabled
1 = Enabled
When FLL1_BYP is set, the FLL1 output is derived directly from BCLK1. In this case, FLL1 can be disabled.
12:7 FLL1_FRC_NC
O_VAL [5:0]
6 FLL1_FRC_NC
O
01_1001 FLL1 Forced oscillator value
Valid range is 000000 to 111111
0x19h (011001) = 12MHz approx
(Note that this field is required for free-running FLL modes only)
0 FLL1 Forced control select
0 = Normal
1 = FLL1 oscillator controlled by FLL1_FRC_NCO_VAL
(Note that this field is required for free-running FLL modes only)
4:3 FLL1_REFCLK
_DIV [1:0]
00 FLL1 Clock Reference Divider
00 = MCLK / 1
01 = MCLK / 2
10 = MCLK / 4
11 = MCLK / 8
MCLK (or other input reference) must be divided down to <=13.5MHz.
For lower power operation, the reference clock can be divided down further if desired. w PP, August 2012, Rev 3.4
299
WM8958
REGISTER
ADDRESS
1:0 FLL1_REFCLK
_SRC [1:0]
Pre-Production
DESCRIPTION
00
Register 0224h FLL1 Control (5)
REGISTER
ADDRESS
R550
(0226h)
FLL1 EFS 1
[15:0]
Register 0226h FLL1 EFS 1
REGISTER
ADDRESS
R550
(0226h)
FLL1 EFS 1
0 FLL1_EFS_EN
A
DESCRIPTION
_0000_000
0
FLL Fractional multiply for FREF
This field sets the denominator (dividing) part of the
FLL1_THETA / FLL1_LAMBDA ratio.
Coded as LSB = 1.
0
DESCRIPTION
FLL Fractional Mode EFS enable
0 = Integer Mode
1 = Fractional Mode
This bit should be set to 1 when FLL1_THETA > 0.
Register 0227h FLL1 EFS 2
FLL1 Clock source
00 = MCLK1
01 = MCLK2
10 = LRCLK1
11 = BCLK1
REGISTER
ADDRESS
R576
(0240h)
FLL2
Control (1)
DESCRIPTION
1 FLL2_OSC_EN
A
0 FLL2 Oscillator enable
0 = Disabled
1 = Enabled
(Note that this field is required for free-running FLL2 modes only)
0 FLL2_ENA 0 FLL2 Enable
0 = Disabled
1 = Enabled
This should be set as the final step of the FLL2 enable sequence, ie. after the other FLL registers have been configured.
Register 0240h FLL2 Control (1) w PP, August 2012, Rev 3.4
300
Pre-Production
REGISTER
ADDRESS
R577
(0241h)
FLL2
Control (2)
DESCRIPTION
13:8 FLL2_OUTDIV 00_0000 FLL2 FOUT clock divider
[5:0]
000000 = Reserved
000001 = Reserved
000010 = Reserved
000011 = 4
000100 = 5
000101 = 6
…
111110 = 63
2:0 FLL2_FRATIO
[2:0]
000
111111 = 64
(FOUT = FVCO / FLL2_OUTDIV)
FLL2 FVCO clock divider
000 = 1
001 = 2
010 = 4
011 = 8
1XX = 16
Register 0241h FLL2 Control (2)
REGISTER
ADDRESS
R578
(0242h)
FLL2
Control (3)
Register 0242h FLL2 Control (3)
REGISTER
ADDRESS
R579
(0243h)
FLL2
Control (4)
Register 0243h FLL2 Control (4)
_0000_000
0
000
FLL2 Fractional multiply for FREF
(MSB = 0.5)
FLL2 Integer multiply for FREF
(LSB = 1)
DESCRIPTION
DESCRIPTION
REGISTER
ADDRESS
R580
(0244h)
FLL2
Control (5)
DESCRIPTION
15 FLL2_BYP 0 FLL2 Bypass Select
0 = Disabled
1 = Enabled
12:7 FLL2_FRC_NC
O_VAL [5:0]
When FLL2_BYP is set, the FLL2 output is derived directly from BCLK2. In this case, FLL2 can be disabled.
01_1001 FLL2 Forced oscillator value
Valid range is 000000 to 111111
0x19h (011001) = 12MHz approx
(Note that this field is required for free-running FLL modes only)
WM8958 w PP, August 2012, Rev 3.4
301
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
6 FLL2_FRC_NC
4:3 FLL2_REFCLK
_DIV [1:0]
1:0 FLL2_REFCLK
_SRC [1:0]
Register 0247h FLL2 EFS 2
O
0
00
00
FLL2 Forced control select
0 = Normal
1 = FLL2 oscillator controlled by FLL2_FRC_NCO_VAL
(Note that this field is required for free-running FLL modes only)
FLL2 Clock Reference Divider
00 = MCLK / 1
01 = MCLK / 2
10 = MCLK / 4
11 = MCLK / 8
MCLK (or other input reference) must be divided down to <=13.5MHz.
For lower power operation, the reference clock can be divided down further if desired.
FLL2 Clock source
00 = MCLK1
01 = MCLK2
10 = LRCLK2
11 = BCLK2
Register 0244h FLL2 Control (5)
REGISTER
ADDRESS
R582
(0246h)
FLL2 EFS 1
[15:0]
Register 0246h FLL2 EFS 1
REGISTER
ADDRESS
R583
(0247h)
FLL2 EFS 2
0 FLL2_EFS_EN
A
DESCRIPTION
_0000_000
0
FLL Fractional multiply for FREF
This field sets the denominator (dividing) part of the
FLL2_THETA / FLL2_LAMBDA ratio.
Coded as LSB = 1.
0
DESCRIPTION
FLL Fractional Mode EFS enable
0 = Integer Mode
1 = Fractional Mode
This bit should be set to 1 when FLL2_THETA > 0.
REGISTER
ADDRESS
R768
(0300h)
AIF1 Control
(1)
15 AIF1ADCL_SR
C
14 AIF1ADCR_SR
C
0
1
DESCRIPTION
AIF1 Left Digital Audio interface source
0 = Left ADC data is output on left channel
1 = Right ADC data is output on left channel
AIF1 Right Digital Audio interface source
0 = Left ADC data is output on right channel
1 = Right ADC data is output on right channel w PP, August 2012, Rev 3.4
302
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
13 AIF1ADC_TDM 0 AIF1 transmit (ADC) TDM Control
0 = ADCDAT1 drives logic ‘0’ when not transmitting data
1 = ADCDAT1 is tri-stated when not transmitting data
8 AIF1_BCLK_IN
V
0 BCLK1 Invert
0 = BCLK1 not inverted
1 = BCLK1 inverted
Note that AIF1_BCLK_INV selects the BCLK1 polarity in Master mode and in Slave mode.
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
00 = Right justified
01 = Left justified
10 = I2S Format
11 = DSP Mode
Register 0300h AIF1 Control (1)
REGISTER
ADDRESS
R769
(0301h)
AIF1 Control
(2)
DESCRIPTION
15 AIF1DACL_SR
C
14 AIF1DACR_SR
C
2 AIF1ADC_CO
MP
1 AIF1ADC_CO
MPMODE
0
1
11:10 AIF1DAC_BOO
ST [1:0]
00
8 AIF1_MONO 0 AIF1 DSP Mono Mode
0 = Disabled
1 = Enabled
Note that Mono Mode is only supported when
AIF1_FMT = 11.
4 AIF1DAC_CO
MP
0
AIF1 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
3 AIF1DAC_CO
MPMODE
0
AIF1 Receive Companding Enable
0 = Disabled
1 = Enabled
AIF1 Receive Companding Type
0 = µ-law
1 = A-law
0
0
AIF1 Left Receive Data Source Select
0 = Left DAC receives left interface data
1 = Left DAC receives right interface data
AIF1 Right Receive Data Source Select
0 = Right DAC receives left interface data
1 = Right DAC receives right interface data
AIF1 Transmit Companding Enable
0 = Disabled
1 = Enabled
AIF1 Transmit Companding Type
0 = µ-law
1 = A-law w
WM8958
PP, August 2012, Rev 3.4
303
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
0 AIF1_LOOPBA
CK
0 AIF1 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (ADCDAT1 data output is directly input to DACDAT1 data input).
Register 0301h AIF1 Control (2)
REGISTER
ADDRESS
R770
(0302h)
AIF1
Master/Slav e
DESCRIPTION
15 AIF1_TRI 0 AIF1 Audio Interface tri-state
0 = AIF1 pins operate normally
1 = Tri-state all AIF1 interface pins
Note that the GPIO1 pin is controlled by this register only when configured as ADCLRCLK1.
14 AIF1_MSTR 0 AIF1 Audio Interface Master Mode Select
0 = Slave mode
1 = Master mode
13 AIF1_CLK_FR
C
0 Forces BCLK1, LRCLK1 and ADCLRCLK1 to be enabled when all AIF1 audio channels are disabled.
0 = Normal
1 = BCLK1, LRCLK1 and ADCLRCLK1 always enabled in Master mode
12 AIF1_LRCLK_
FRC
0 Forces LRCLK1 and ADCLRCLK1 to be enabled when all AIF1 audio channels are disabled.
0 = Normal
1 = LRCLK1 and ADCLRCLK1 always enabled in
Master mode
Register 0302h AIF1 Master/Slave w PP, August 2012, Rev 3.4
304
Pre-Production
REGISTER
ADDRESS
R771
(0303h)
AIF1 BCLK
8:4 AIF1_BCLK_DI
V [4:0]
DESCRIPTION
0_0100 BCLK1 Rate
00000 = AIF1CLK
00001 = AIF1CLK / 1.5
00010 = AIF1CLK / 2
00011 = AIF1CLK / 3
00100 = AIF1CLK / 4
00101 = AIF1CLK / 5
00110 = AIF1CLK / 6
00111 = AIF1CLK / 8
01000 = AIF1CLK / 11
01001 = AIF1CLK / 12
01010 = AIF1CLK / 16
01011 = AIF1CLK / 22
01100 = AIF1CLK / 24
01101 = AIF1CLK / 32
01110 = AIF1CLK / 44
01111 = AIF1CLK / 48
10000 = AIF1CLK / 64
10001 = AIF1CLK / 88
10010 = AIF1CLK / 96
10011 = AIF1CLK / 128
10100 = AIF1CLK / 176
10101 = AIF1CLK / 192
10110 - 11111 = Reserved
Register 0303h AIF1 BCLK
REGISTER
ADDRESS
R772
(0304h)
AIF1ADC
LRCLK
DESCRIPTION
12 AIF1ADC_LRC
LK_INV
11 AIF1ADC_LRC
LK_DIR
10:0 AIF1ADC_RAT
E [10:0]
0
0
000_0100_
0000
Right, left and I2S modes – ADCLRCLK1 polarity
0 = normal ADCLRCLK1 polarity
1 = invert ADCLRCLK1 polarity
Note that AIF1ADC_LRCLK_INV selects the
ADCLRCLK1 polarity in Master mode and in Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK1 rising edge after
ADCLRCLK1 rising edge (mode A)
1 = MSB is available on 1st BCLK1 rising edge after
ADCLRCLK1 rising edge (mode B)
Allows ADCLRCLK1 to be enabled in Slave mode
0 = Normal
1 = ADCLRCLK1 enabled in Slave mode
ADCLRCLK1 Rate
ADCLRCLK1 clock output =
BCLK1 / AIF1ADC_RATE
Integer (LSB = 1)
Valid from 8..2047
Register 0304h AIF1ADC LRCLK w
WM8958
PP, August 2012, Rev 3.4
305
WM8958
Pre-Production
REGISTER
ADDRESS
R773
(0305h)
AIF1DAC
LRCLK
DESCRIPTION
12 AIF1DAC_LRC
LK_INV
11 AIF1DAC_LRC
LK_DIR
10:0 AIF1DAC_RAT
E [10:0]
0
0
000_0100_
0000
Right, left and I2S modes – LRCLK1 polarity
0 = normal LRCLK1 polarity
1 = invert LRCLK1 polarity
Note that AIF1DAC_LRCLK_INV selects the LRCLK1 polarity in Master mode and in Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK1 rising edge after
LRCLK1 rising edge (mode A)
1 = MSB is available on 1st BCLK1 rising edge after
LRCLK1 rising edge (mode B)
Allows LRCLK1 to be enabled in Slave mode
0 = Normal
1 = LRCLK1 enabled in Slave mode
LRCLK1 Rate
LRCLK1 clock output =
BCLK1 / AIF1DAC_RATE
Integer (LSB = 1)
Valid from 8..2047
Register 0305h AIF1DAC LRCLK
REGISTER
ADDRESS
R774
(0306h)
AIF1DAC
Data
1 AIF1DACL_DA
T_INV
0 AIF1DACR_DA
T_INV
0
0
DESCRIPTION
AIF1 Left Receive Data Invert
0 = Not inverted
1 = Inverted
AIF1 Right Receive Data Invert
0 = Not inverted
1 = Inverted
Register 0306h AIF1DAC Data
REGISTER
ADDRESS
R775
(0307h)
AIF1ADC
Data
1 AIF1ADCL_DA
T_INV
0 AIF1ADCR_DA
T_INV
0
0
DESCRIPTION
AIF1 Left Transmit Data Invert
0 = Not inverted
1 = Inverted
AIF1 Right Transmit Data Invert
0 = Not inverted
1 = Inverted
Register 0307h AIF1ADC Data w PP, August 2012, Rev 3.4
306
Pre-Production
REGISTER
ADDRESS
R784
(0310h)
AIF2 Control
(1)
DESCRIPTION
15 AIF2ADCL_SR
C
14 AIF2ADCR_SR
C
0
1
AIF2 Left Digital Audio interface source
0 = Left ADC data is output on left channel
1 = Right ADC data is output on left channel
AIF2 Right Digital Audio interface source
0 = Left ADC data is output on right channel
1 = Right ADC data is output on right channel
13 AIF2ADC_TDM 0 AIF2 transmit (ADC) TDM Enable
0 = Normal ADCDAT2 operation
1 = TDM enabled on ADCDAT2
12 AIF2ADC_TDM
_CHAN
0 AIF2 transmit (ADC) TDM Slot Select
0 = Slot 0
1 = Slot 1
8 AIF2_BCLK_IN
V
0 BCLK2 Invert
0 = BCLK2 not inverted
1 = BCLK2 inverted
Note that AIF2_BCLK_INV selects the BCLK2 polarity in Master mode and in Slave mode.
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
00 = Right justified
01 = Left justified
10 = I2S Format
11 = DSP Mode
1 AIF2TXL_ENA 1 Enable AIF2DAC (Left) input path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 output of the AIF2ADC
(Left) signal. For AIF3 output only, this bit can be set to
0.
0 AIF2TXR_ENA 1 Enable AIF2DAC (Right) input path
0 = Disabled
1 = Enabled
This bit must be set for AIF2 output of the AIF2ADC
(Left) signal. For AIF3 output only, this bit can be set to
0.
Register 0310h AIF2 Control (1)
REGISTER
ADDRESS
R785
(0311h)
AIF2 Control
(2)
15 AIF2DACL_SR
C
14 AIF2DACR_SR
C
0
1
DESCRIPTION
AIF2 Left Receive Data Source Select
0 = Left DAC receives left interface data
1 = Left DAC receives right interface data
AIF2 Right Receive Data Source Select
0 = Right DAC receives left interface data
1 = Right DAC receives right interface data
WM8958 w PP, August 2012, Rev 3.4
307
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
13 AIF2DAC_TDM 0 AIF2 receive (DAC) TDM Enable
0 = Normal DACDAT2 operation
1 = TDM enabled on DACDAT2
12 AIF2DAC_TDM
_CHAN
0 AIF2 receive (DAC) TDM Slot Select
0 = Slot 0
1 = Slot 1
11:10 AIF2DAC_BOO
ST [1:0]
00
8 AIF2_MONO 0 AIF2 DSP Mono Mode
0 = Disabled
1 = Enabled
Note that Mono Mode is only supported when
AIF2_FMT = 11.
4 AIF2DAC_CO
MP
0
AIF2 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
3 AIF2DAC_CO
MPMODE
0
AIF2 Receive Companding Enable
0 = Disabled
1 = Enabled
AIF2 Receive Companding Type
0 = µ-law
1 = A-law
2 AIF2ADC_CO
MP
0
1 AIF2ADC_CO
MPMODE
0 AIF2_LOOPBA
CK
0
0
AIF2 Transmit Companding Enable
0 = Disabled
1 = Enabled
AIF2 Transmit Companding Type
0 = µ-law
1 = A-law
AIF2 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (ADCDAT2 data output is directly input to DACDAT2 data input).
Register 0311h AIF2 Control (2)
REGISTER
ADDRESS
R786
(0312h)
AIF2
Master/Slav e
DESCRIPTION
15 AIF2_TRI 0 AIF2 Audio Interface tri-state
0 = AIF2 pins operate normally
1 = Tri-state all AIF2 interface pins
Note that pins not configured as AIF2 functions are not affected by this register.
14 AIF2_MSTR 0 AIF2 Audio Interface Master Mode Select
0 = Slave mode
1 = Master mode
13 AIF2_CLK_FR
C
0 Forces BCLK2, LRCLK2 and ADCLRCLK2 to be enabled when all AIF2 audio channels are disabled.
0 = Normal
1 = BCLK2, LRCLK2 and ADCLRCLK2 always enabled in Master mode w PP, August 2012, Rev 3.4
308
Pre-Production
REGISTER
ADDRESS
12 AIF2_LRCLK_
FRC
DESCRIPTION
0 Forces LRCLK2 and ADCLRCLK2 to be enabled when all AIF2 audio channels are disabled.
0 = Normal
1 = LRCLK2 and ADCLRCLK2 always enabled in
Master mode
Register 0312h AIF2 Master/Slave
REGISTER
ADDRESS
R787
(0313h)
AIF2 BCLK
DESCRIPTION
8:4 AIF2_BCLK_DI
V [4:0]
0_0100 BCLK2 Rate
00000 = AIF2CLK
00001 = AIF2CLK / 1.5
00010 = AIF2CLK / 2
00011 = AIF2CLK / 3
00100 = AIF2CLK / 4
00101 = AIF2CLK / 5
00110 = AIF2CLK / 6
00111 = AIF2CLK / 8
01000 = AIF2CLK / 11
01001 = AIF2CLK / 12
01010 = AIF2CLK / 16
01011 = AIF2CLK / 22
01100 = AIF2CLK / 24
01101 = AIF2CLK / 32
01110 = AIF2CLK / 44
01111 = AIF2CLK / 48
10000 = AIF2CLK / 64
10001 = AIF2CLK / 88
10010 = AIF2CLK / 96
10011 = AIF2CLK / 128
10100 = AIF2CLK / 176
10101 = AIF2CLK / 192
10110 - 11111 = Reserved
Register 0313h AIF2 BCLK
REGISTER
ADDRESS
R788
(0314h)
AIF2ADC
LRCLK
12 AIF2ADC_LRC
LK_INV
0
DESCRIPTION
11 AIF2ADC_LRC w
LK_DIR
0
WM8958
Right, left and I2S modes – ADCLRCLK2 polarity
0 = normal ADCLRCLK2 polarity
1 = invert ADCLRCLK2 polarity
Note that AIF2ADC_LRCLK_INV selects the
ADCLRCLK2 polarity in Master mode and in Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK2 rising edge after
ADCLRCLK2 rising edge (mode A)
1 = MSB is available on 1st BCLK2 rising edge after
ADCLRCLK2 rising edge (mode B)
Allows ADCLRCLK2 to be enabled in Slave mode
0 = Normal
1 = ADCLRCLK2 enabled in Slave mode
PP, August 2012, Rev 3.4
309
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
10:0 AIF2ADC_RAT
E [10:0]
000_0100_
0000
ADCLRCLK2 Rate
ADCLRCLK2 clock output =
BCLK2 / AIF2ADC_RATE
Integer (LSB = 1)
Valid from 8..2047
Register 0314h AIF2ADC LRCLK
REGISTER
ADDRESS
R789
(0315h)
AIF2DAC
LRCLK
DESCRIPTION
12 AIF2DAC_LRC
LK_INV
11 AIF2DAC_LRC
LK_DIR
10:0 AIF2DAC_RAT
E [10:0]
0
0
000_0100_
0000
Right, left and I2S modes – LRCLK2 polarity
0 = normal LRCLK2 polarity
1 = invert LRCLK2 polarity
Note that AIF2DAC_LRCLK_INV selects the LRCLK2 polarity in Master mode and in Slave mode.
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK2 rising edge after
LRCLK2 rising edge (mode A)
1 = MSB is available on 1st BCLK2 rising edge after
LRCLK2 rising edge (mode B)
Allows LRCLK2 to be enabled in Slave mode
0 = Normal
1 = LRCLK2 enabled in Slave mode
LRCLK2 Rate
LRCLK2 clock output =
BCLK2 / AIF2DAC_RATE
Integer (LSB = 1)
Valid from 8..2047
Register 0315h AIF2DAC LRCLK
REGISTER
ADDRESS
R790
(0316h)
AIF2DAC
Data
1 AIF2DACL_DA
T_INV
0 AIF2DACR_DA
T_INV
0
0
AIF2 Left Receive Data Invert
0 = Not inverted
1 = Inverted
AIF2 Right Receive Data Invert
0 = Not inverted
1 = Inverted
DESCRIPTION
Register 0316h AIF2DAC Data
REGISTER
ADDRESS
R791
(0317h)
AIF2ADC
Data
1 AIF2ADCL_DA
T_INV
0 AIF2ADCR_DA
T_INV
0
0
AIF2 Left Transmit Data Invert
0 = Not inverted
1 = Inverted
AIF2 Right Transmit Data Invert
0 = Not inverted
1 = Inverted
DESCRIPTION
Register 0317h AIF2ADC Data w PP, August 2012, Rev 3.4
310
Pre-Production
REGISTER
ADDRESS
R800
(0320h)
AIF3 Control
(1)
7 AIF3_LRCLK_I
NV
0
DESCRIPTION
Right, left and I 2 S modes – LRCLK3 polarity
0 = normal LRCLK3 polarity
1 = invert LRCLK3 polarity
DSP Mode – mode A/B select
0 = MSB is available on 2nd BCLK3 rising edge after
LRCLK3 rising edge (mode A)
1 = MSB is available on 1st BCLK3 rising edge after
LRCLK3 rising edge (mode B)
00 = 16 bits
01 = 20 bits
10 = 24 bits
11 = 32 bits
Note - 8-bit modes can be selected using the
“Companding” control bits.
Note that this controls the AIF3 Mono PCM interface path only; it does not affect AIF3 inputs/outputs routed to AIF1 or AIF2.
Register 0320h AIF3 Control (1)
REGISTER
ADDRESS
R801
(0321h)
AIF3 Control
(2)
11:10 AIF3DAC_BOO
ST [1:0]
4 AIF3DAC_CO
MP
3 AIF3DAC_CO
MPMODE
2 AIF3ADC_CO
MP
00
0
0
0
DESCRIPTION
AIF3 Input Path Boost
00 = 0dB
01 = +6dB (input must not exceed -6dBFS)
10 = +12dB (input must not exceed -12dBFS)
11 = +18dB (input must not exceed -18dBFS)
Note that this controls the AIF3 Mono PCM interface path only; it does not affect DACDAT3 input to AIF1 or
AIF2.
AIF3 Receive Companding Enable
0 = Disabled
1 = Enabled
Note that this controls the AIF3 Mono PCM interface path only; it does not affect DACDAT3 input to AIF1 or
AIF2.
AIF3 Receive Companding Type
0 = µ-law
1 = A-law
Note that this controls the AIF3 Mono PCM interface path only; it does not affect DACDAT3 input to AIF1 or
AIF2.
AIF3 Transmit Companding Enable
0 = Disabled
1 = Enabled
Note that this controls the AIF3 Mono PCM interface path only; it does not affect ADCDAT3 output from AIF1 or AIF2.
WM8958 w PP, August 2012, Rev 3.4
311
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
1 AIF3ADC_CO
MPMODE
0 AIF3_LOOPBA
CK
0
0
AIF3 Transmit Companding Type
0 = µ-law
1 = A-law
Note that this controls the AIF3 Mono PCM interface path only; it does not affect ADCDAT3 output from AIF1 or AIF2.
AIF3 Digital Loopback Function
0 = No loopback
1 = Loopback enabled (AIF3 Mono PCM data output is directly input to AIF3 Mono PCM data input).
Register 0321h AIF3 Control (2)
REGISTER
ADDRESS
R802
(0322h)
AIF3DAC
Data
0 AIF3DAC_DAT
_INV
0
DESCRIPTION
AIF3 Receive Data Invert
0 = Not inverted
1 = Inverted
Note that this controls the AIF3 Mono PCM interface path only; it does not affect DACDAT3 input to AIF1 or
AIF2.
Register 0322h AIF3DAC Data
REGISTER
ADDRESS
R803
(0323h)
AIF3ADC
Data
0 AIF3ADC_DAT
_INV
0
DESCRIPTION
AIF3 Transmit Data Invert
0 = Not inverted
1 = Inverted
Note that this controls the AIF3 Mono PCM interface path only; it does not affect ADCDAT3 output from AIF1 or AIF2.
Register 0323h AIF3ADC Data
REGISTER
ADDRESS
R1024
(0400h)
AIF1 ADC1
Left Volume
DESCRIPTION
8 AIF1ADC1_VU 0 AIF1ADC1 output path (AIF1, Timeslot 0) Volume
Update
Writing a 1 to this bit will cause the AIF1ADC1L and
AIF1ADC1R volume to be updated simultaneously
7:0 AIF1ADC1L_V
OL [7:0]
1100_0000 AIF1ADC1 (Left) output path (AIF1, Timeslot 0) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0400h AIF1 ADC1 Left Volume w PP, August 2012, Rev 3.4
312
Pre-Production
REGISTER
ADDRESS
R1025
(0401h)
AIF1 ADC1
Right
Volume
DESCRIPTION
8 AIF1ADC1_VU 0 AIF1ADC1 output path (AIF1, Timeslot 0) Volume
Update
Writing a 1 to this bit will cause the AIF1ADC1L and
AIF1ADC1R volume to be updated simultaneously
7:0 AIF1ADC1R_V
OL [7:0]
1100_0000 AIF1ADC1 (Right) output path (AIF1, Timeslot 0) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0401h AIF1 ADC1 Right Volume
REGISTER
ADDRESS
R1026
(0402h)
AIF1 DAC1
Left Volume
DESCRIPTION
8 AIF1DAC1_VU 0 AIF1DAC1 input path (AIF1, Timeslot 0) Volume
Update
Writing a 1 to this bit will cause the AIF1DAC1L and
AIF1DAC1R volume to be updated simultaneously
7:0 AIF1DAC1L_V
OL [7:0]
1100_0000 AIF1DAC1 (Left) input path (AIF1, Timeslot 0) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0402h AIF1 DAC1 Left Volume
REGISTER
ADDRESS
R1027
(0403h)
AIF1 DAC1
Right
Volume
DESCRIPTION
8 AIF1DAC1_VU 0 AIF1DAC1 input path (AIF1, Timeslot 0) Volume
Update
Writing a 1 to this bit will cause the AIF1DAC1L and
AIF1DAC1R volume to be updated simultaneously
7:0 AIF1DAC1R_V
OL [7:0]
1100_0000 AIF1DAC1 (Right) input path (AIF1, Timeslot 0) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0403h AIF1 DAC1 Right Volume
WM8958 w PP, August 2012, Rev 3.4
313
WM8958
Pre-Production
REGISTER
ADDRESS
R1028
(0404h)
AIF1 ADC2
Left Volume
DESCRIPTION
8 AIF1ADC2_VU 0 AIF1ADC2 output path (AIF1, Timeslot 1) Volume
Update
Writing a 1 to this bit will cause the AIF1ADC2L and
AIF1ADC2R volume to be updated simultaneously
7:0 AIF1ADC2L_V
OL [7:0]
1100_0000 AIF1ADC2 (Left) output path (AIF1, Timeslot 1) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0404h AIF1 ADC2 Left Volume
REGISTER
ADDRESS
R1029
(0405h)
AIF1 ADC2
Right
Volume
DESCRIPTION
8 AIF1ADC2_VU 0 AIF1ADC2 output path (AIF1, Timeslot 1) Volume
Update
Writing a 1 to this bit will cause the AIF1ADC2L and
AIF1ADC2R volume to be updated simultaneously
7:0 AIF1ADC2R_V
OL [7:0]
1100_0000 AIF1ADC2 (Right) output path (AIF1, Timeslot 1) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0405h AIF1 ADC2 Right Volume
REGISTER
ADDRESS
R1030
(0406h)
AIF1 DAC2
Left Volume
DESCRIPTION
8 AIF1DAC2_VU 0 AIF1DAC2 input path (AIF1, Timeslot 1) Volume
Update
Writing a 1 to this bit will cause the AIF1DAC2L and
AIF1DAC2R volume to be updated simultaneously
7:0 AIF1DAC2L_V
OL [7:0]
1100_0000 AIF1DAC2 (Left) input path (AIF1, Timeslot 1) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0406h AIF1 DAC2 Left Volume w PP, August 2012, Rev 3.4
314
Pre-Production
REGISTER
ADDRESS
R1031
(0407h)
AIF1 DAC2
Right
Volume
DESCRIPTION
8 AIF1DAC2_VU 0 AIF1DAC2 input path (AIF1, Timeslot 1) Volume
Update
Writing a 1 to this bit will cause the AIF1DAC2L and
AIF1DAC2R volume to be updated simultaneously
7:0 AIF1DAC2R_V
OL [7:0]
1100_0000 AIF1DAC2 (Right) input path (AIF1, Timeslot 1) Digital
Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0407h AIF1 DAC2 Right Volume
WM8958
REGISTER
ADDRESS
R1040
(0410h)
AIF1 ADC1
Filters
DESCRIPTION
15 AIF1ADC_4FS 0 Enable AIF1ADC ultrasonic mode (4FS) output, bypassing all AIF1 baseband output filtering
0 = Disabled
1 = Enabled
14:13 AIF1ADC1_HP
F_CUT [1:0]
00 AIF1ADC1 output path (AIF1, Timeslot 0) Digital HPF cut-off frequency (fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 127Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
12 AIF1ADC1L_H
PF
0
11 AIF1ADC1R_H
PF
0
AIF1ADC1 (Left) output path (AIF1, Timeslot 0) Digital
HPF Enable
0 = Disabled
1 = Enabled
AIF1ADC1 (Right) output path (AIF1, Timeslot 0) Digital
HPF Enable
0 = Disabled
1 = Enabled
Register 0410h AIF1 ADC1 Filters
REGISTER
ADDRESS
R1041
(0411h)
AIF1 ADC2
Filters
14:13 AIF1ADC2_HP
F_CUT [1:0]
12 AIF1ADC2L_H
PF
11 AIF1ADC2R_H
PF
00
0
0
DESCRIPTION
AIF1ADC2 output path (AIF1, Timeslot 1) Digital HPF cut-off frequency (fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 127Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
AIF1ADC2 (Left) output path (AIF1, Timeslot 1) Digital
HPF Enable
0 = Disabled
1 = Enabled
AIF1ADC2 (Right) output path (AIF1, Timeslot 1) Digital
HPF Enable
0 = Disabled
1 = Enabled
Register 0411h AIF1 ADC2 Filters w PP, August 2012, Rev 3.4
315
WM8958
REGISTER
ADDRESS
R1056
(0420h)
AIF1 DAC1
Filters (1)
9 AIF1DAC1_MU
TE
7 AIF1DAC1_MO
NO
5 AIF1DAC1_MU
TERATE
4 AIF1DAC1_UN
MUTE_RAMP
1
0
0
0
Pre-Production
DESCRIPTION
AIF1DAC1 input path (AIF1, Timeslot 0) Soft Mute
Control
0 = Un-mute
1 = Mute
AIF1DAC1 input path (AIF1, Timeslot 0) Mono Mix
Control
0 = Disabled
1 = Enabled
AIF1DAC1 input path (AIF1, Timeslot 0) Soft Mute
Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is 10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is 171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF1DAC1 input path (AIF1, Timeslot 0) Unmute Ramp select
0 = Disabling soft-mute (AIF1DAC1_MUTE=0) will cause the volume to change immediately to
AIF1DAC1L_VOL and AIF1DAC1R_VOL settings
1 = Disabling soft-mute (AIF1DAC1_MUTE=0) will cause the DAC volume to ramp up gradually to the
AIF1DAC1L_VOL and AIF1DAC1R_VOL settings
Register 0420h AIF1 DAC1 Filters (1)
REGISTER
ADDRESS
R1057
(0421h)
AIF1 DAC1
Filters (2)
13:9 AIF1DAC1_3D
_GAIN [4:0]
8 AIF1DAC1_3D
_ENA
DESCRIPTION
0_0000 AIF1DAC1 playback path (AIF1, Timeslot 0) 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
0 Enable 3D Stereo in AIF1DAC1 playback path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Register 0421h AIF1 DAC1 Filters (2)
REGISTER
ADDRESS
R1058
(0422h)
AIF1 DAC2
Filters (1)
9 AIF1DAC2_MU
TE
7 AIF1DAC2_MO
NO
1
0
Control
0 = Un-mute
1 = Mute
DESCRIPTION
AIF1DAC2 input path (AIF1, Timeslot 1) Soft Mute
AIF1DAC2 input path (AIF1, Timeslot 1) Mono Mix
Control
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
316
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
5 AIF1DAC2_MU
TERATE
4 AIF1DAC2_UN
MUTE_RAMP
Register 0422h AIF1 DAC2 Filters (1)
REGISTER
ADDRESS
R1059
(0423h)
AIF1 DAC2
Filters (2)
13:9 AIF1DAC2_3D
_GAIN [4:0]
8 AIF1DAC2_3D
_ENA
DESCRIPTION
0_0000 AIF1DAC2 playback path (AIF1, Timeslot 1) 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
0 Enable 3D Stereo in AIF1DAC2 playback path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Register 0423h AIF1 DAC2 Filters (2)
REGISTER
ADDRESS
R1072
(0430h)
AIF1 DAC1
Noise Gate
6:5 AIF1DAC1_NG
_HLD [1:0]
0
0
11
AIF1DAC2 input path (AIF1, Timeslot 1) Soft Mute
Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is 10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is 171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF1DAC2 input path (AIF1, Timeslot 1) Unmute Ramp select
0 = Disabling soft-mute (AIF1DAC2_MUTE=0) will cause the volume to change immediately to
AIF1DAC2L_VOL and AIF1DAC2R_VOL settings
1 = Disabling soft-mute (AIF1DAC2_MUTE=0) will cause the DAC volume to ramp up gradually to the
AIF1DAC2L_VOL and AIF1DAC2R_VOL settings
DESCRIPTION
3:1 AIF1DAC1_NG
_THR [2:0]
0 AIF1DAC1_NG
_ENA
100
0
AIF1DAC1 input path (AIF1, Timeslot 0) Noise Gate
Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
AIF1DAC1 input path (AIF1, Timeslot 0) Noise Gate
Threshold
000 = -60dB
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF1DAC1 input path (AIF1, Timeslot 0) Noise Gate
Enable
0 = Disabled
1 = Enabled
Register 0430h AIF1 DAC1 Noise Gate w
WM8958
PP, August 2012, Rev 3.4
317
WM8958
REGISTER
ADDRESS
R1073
(0431h)
AIF1 DAC2
Noise Gate
6:5 AIF1DAC2_NG
_HLD [1:0]
3:1 AIF1DAC2_NG
_THR [2:0]
0 AIF1DAC2_NG
_ENA
11
100
0
Pre-Production
DESCRIPTION
AIF1DAC2 input path (AIF1, Timeslot 1) Noise Gate
Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
AIF1DAC2 input path (AIF1, Timeslot 1) Noise Gate
Threshold
000 = -60dB
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF1DAC2 input path (AIF1, Timeslot 1) Noise Gate
Enable
0 = Disabled
1 = Enabled
Register 0431h AIF1 DAC2 Noise Gate
REGISTER
ADDRESS
R1088
(0440h)
AIF1 DRC1
(1)
15:11 AIF1DRC1_SI
G_DET_RMS
[4:0]
10:9 AIF1DRC1_SI
G_DET_PK
[1:0]
8 AIF1DRC1_NG
_ENA
7 AIF1DRC1_SI
G_DET_MODE
6 AIF1DRC1_SI
G_DET
DESCRIPTION
0_0000 AIF1 DRC1 Signal Detect RMS Threshold.
This is the RMS signal level for signal detect to be indicated when AIF1DRC1_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
11110 = -75dB
11111 = -76.5dB
00 AIF1 DRC1 Signal Detect Peak Threshold.
This is the Peak/RMS ratio, or Crest Factor, level for signal detect to be indicated when
AIF1DRC1_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
0
1
0
AIF1 DRC1 Noise Gate Enable
0 = Disabled
1 = Enabled
AIF1 DRC1 Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF1 DRC1 Signal Detect Enable
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
318
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
5 AIF1DRC1_KN
EE2_OP_ENA
3 AIF1DRC1_AN
TICLIP
0 AIF1 DRC1 KNEE2_OP Enable
0 = Disabled
1 = Enabled
4 AIF1DRC1_QR 1 AIF1 DRC1 Quick-release Enable
0 = Disabled
1 = Enabled
1 AIF1 DRC1 Anti-clip Enable
0 = Disabled
1 = Enabled
2 AIF1DAC1_DR
C_ENA
0
1 AIF1ADC1L_D
RC_ENA
0
Enable DRC in AIF1DAC1 playback path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC1 (Left) record path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
0 AIF1ADC1R_D
RC_ENA
0 Enable DRC in AIF1ADC1 (Right) record path (AIF1,
Timeslot 0)
0 = Disabled
1 = Enabled
Register 0440h AIF1 DRC1 (1)
REGISTER
ADDRESS
R1089
(0441h)
AIF1 DRC1
(2)
12:9 AIF1DRC1_AT
K [3:0]
DESCRIPTION
8:5 AIF1DRC1_DC
Y [3:0]
0100
0010
AIF1 DRC1 Gain attack rate (seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF1 DRC1 Gain decay rate (seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
WM8958 w PP, August 2012, Rev 3.4
319
WM8958
REGISTER
ADDRESS
4:2 AIF1DRC1_MI
NGAIN [2:0]
1:0 AIF1DRC1_MA
XGAIN [1:0]
001
01
Pre-Production
DESCRIPTION
AIF1 DRC1 Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved
AIF1 DRC1 Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
Register 0441h AIF1 DRC1 (2)
REGISTER
ADDRESS
R1090
(0442h)
AIF1 DRC1
(3)
15:12 AIF1DRC1_NG
_MINGAIN
[3:0]
11:10 AIF1DRC1_NG
_EXP [1:0]
9:8 AIF1DRC1_QR
_THR [1:0]
7:6 AIF1DRC1_QR
_DCY [1:0]
DESCRIPTION
0000
00
00
00
AIF1 DRC1 Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF1 DRC1 Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF1 DRC1 Quick-release threshold (crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC1 Quick-release decay rate (seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved w PP, August 2012, Rev 3.4
320
Pre-Production
REGISTER
ADDRESS
5:3 AIF1DRC1_HI_
COMP [2:0]
DESCRIPTION
2:0 AIF1DRC1_LO
_COMP [2:0]
000
000
AIF1 DRC1 Compressor slope (upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved
AIF1 DRC1 Compressor slope (lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
Register 0442h AIF1 DRC1 (3)
REGISTER
ADDRESS
R1091
(0443h)
AIF1 DRC1
(4)
DESCRIPTION
10:5 AIF1DRC1_KN
EE_IP [5:0]
4:0 AIF1DRC1_KN
EE_OP [4:0]
00_0000 AIF1 DRC1 Input signal level at the Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
0_0000 AIF1 DRC1 Output signal at the Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
Register 0443h AIF1 DRC1 (4)
REGISTER
ADDRESS
R1092
(0444h)
AIF1 DRC1
(5)
DESCRIPTION
9:5 AIF1DRC1_KN
EE2_IP [4:0]
0_0000 AIF1 DRC1 Input signal level at the Noise Gate threshold ‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when AIF1DRC1_NG_ENA = 1.
WM8958 w PP, August 2012, Rev 3.4
321
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
4:0 AIF1DRC1_KN
EE2_OP [4:0]
0_0000 AIF1 DRC1 Output signal at the Noise Gate threshold
‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when AIF1DRC1_KNEE2_OP_ENA =
1.
Register 0444h AIF1 DRC1 (5)
REGISTER
ADDRESS
R1104
(0450h)
AIF1 DRC2
(1)
DESCRIPTION
15:11 AIF1DRC2_SI
G_DET_RMS
[4:0]
10:9 AIF1DRC2_SI
G_DET_PK
[1:0]
0_0000 AIF1 DRC2 Signal Detect RMS Threshold.
This is the RMS signal level for signal detect to be indicated when AIF1DRC2_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
11110 = -75dB
11111 = -76.5dB
00 AIF1 DRC2 Signal Detect Peak Threshold.
This is the Peak/RMS ratio, or Crest Factor, level for signal detect to be indicated when
AIF1DRC2_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
8 AIF1DRC2_NG
_ENA
7 AIF1DRC2_SI
G_DET_MODE
6 AIF1DRC2_SI
G_DET
5 AIF1DRC2_KN
EE2_OP_ENA
0
1
0
0
AIF1 DRC2 Noise Gate Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF1 DRC2 Signal Detect Enable
0 = Disabled
1 = Enabled
AIF1 DRC2 KNEE2_OP Enable
0 = Disabled
1 = Enabled
4 AIF1DRC2_QR 1 AIF1 DRC2 Quick-release Enable
0 = Disabled
3 AIF1DRC2_AN
TICLIP
1
1 = Enabled
AIF1 DRC2 Anti-clip Enable
0 = Disabled
1 = Enabled
2 AIF1DAC2_DR
C_ENA
0 Enable DRC in AIF1DAC2 playback path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
322
Pre-Production
REGISTER
ADDRESS
1 AIF1ADC2L_D
RC_ENA
0 AIF1ADC2R_D
RC_ENA
0
0
Register 0450h AIF1 DRC2 (1)
REGISTER
ADDRESS
R1105
(0451h)
AIF1 DRC2
(2)
12:9 AIF1DRC2_AT
K [3:0]
8:5 AIF1DRC2_DC
Y [3:0]
4:2 AIF1DRC2_MI
NGAIN [2:0]
1:0 AIF1DRC2_MA
XGAIN [1:0]
DESCRIPTION
0100
0010
001
01
AIF1 DRC2 Gain attack rate (seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF1 DRC2 Gain decay rate (seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
AIF1 DRC2 Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved
AIF1 DRC2 Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
Register 0451h AIF1 DRC2 (2)
DESCRIPTION
Enable DRC in AIF1ADC2 (Left) record path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
Enable DRC in AIF1ADC2 (Right) record path (AIF1,
Timeslot 1)
0 = Disabled
1 = Enabled
WM8958 w PP, August 2012, Rev 3.4
323
WM8958
REGISTER
ADDRESS
R1106
(0452h)
AIF1 DRC2
(3)
15:12 AIF1DRC2_NG
_MINGAIN
[3:0]
11:10 AIF1DRC2_NG
_EXP [1:0]
9:8 AIF1DRC2_QR
_THR [1:0]
7:6 AIF1DRC2_QR
_DCY [1:0]
5:3 AIF1DRC2_HI_
COMP [2:0]
2:0 AIF1DRC2_LO
_COMP [2:0]
Pre-Production
DESCRIPTION
0000
00
00
00
000
000
AIF1 DRC2 Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF1 DRC2 Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF1 DRC2 Quick-release threshold (crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF1 DRC2 Quick-release decay rate (seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved
AIF1 DRC2 Compressor slope (upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved
AIF1 DRC2 Compressor slope (lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
Register 0452h AIF1 DRC2 (3) w PP, August 2012, Rev 3.4
324
Pre-Production
REGISTER
ADDRESS
R1107
(0453h)
AIF1 DRC2
(4)
10:5 AIF1DRC2_KN
EE_IP [5:0]
4:0 AIF1DRC2_KN
EE_OP [4:0]
DESCRIPTION
00_0000 AIF1 DRC2 Input signal level at the Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
0_0000 AIF1 DRC2 Output signal at the Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
Register 0453h AIF1 DRC2 (4)
REGISTER
ADDRESS
R1108
(0454h)
AIF1 DRC2
(5)
9:5 AIF1DRC2_KN
EE2_IP [4:0]
4:0 AIF1DRC2_KN
EE2_OP [4:0]
DESCRIPTION
0_0000 AIF1 DRC2 Input signal level at the Noise Gate threshold ‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when AIF1DRC2_NG_ENA = 1.
0_0000 AIF1 DRC2 Output signal at the Noise Gate threshold
‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when AIF1DRC2_KNEE2_OP_ENA =
1.
Register 0454h AIF1 DRC2 (5)
REGISTER
ADDRESS
R1152
(0480h)
AIF1 DAC1
EQ Gains
(1)
15:11 AIF1DAC1_EQ
_B1_GAIN [4:0]
10:6 AIF1DAC1_EQ
_B2_GAIN [4:0]
5:1 AIF1DAC1_EQ
_B3_GAIN [4:0]
0 AIF1DAC1_EQ
_ENA
Register 0480h AIF1 DAC1 EQ Gains (1)
DESCRIPTION
0_1100 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 1 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 2 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 3 Gain
-12dB to +12dB in 1dB steps
0 Enable EQ in AIF1DAC1 playback path (AIF1, Timeslot
0)
0 = Disabled
1 = Enabled w
WM8958
PP, August 2012, Rev 3.4
325
WM8958
REGISTER
ADDRESS
R1153
(0481h)
AIF1 DAC1
EQ Gains
(2)
15:11 AIF1DAC1_EQ
_B4_GAIN [4:0]
10:6 AIF1DAC1_EQ
_B5_GAIN [4:0]
0 AIF1DAC1_EQ
_MODE
Register 0481h AIF1 DAC1 EQ Gains (2)
Pre-Production
DESCRIPTION
0_1100 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 4 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 5 Gain
-12dB to +12dB in 1dB steps
0 AIF1DAC1 (AIF1, Timeslot 0) EQ Band 1 Mode
0 = Shelving filter
1 = Peak filter
REGISTER
ADDRESS
R1154
(0482h)
AIF1 DAC1
EQ Band 1
A
15:0 AIF1DAC1_EQ
_B1_A [15:0]
Register 0482h AIF1 DAC1 EQ Band 1 A
DESCRIPTION
0000_1111
_1100_101
0
EQ Band 1 Coefficient A
REGISTER
ADDRESS
R1155
(0483h)
AIF1 DAC1
EQ Band 1
B
15:0 AIF1DAC1_EQ
_B1_B [15:0]
Register 0483h AIF1 DAC1 EQ Band 1 B
DESCRIPTION
0000_0100
_0000_000
0
EQ Band 1 Coefficient B
REGISTER
ADDRESS
R1156
(0484h)
AIF1 DAC1
EQ Band 1
PG
15:0 AIF1DAC1_EQ
_B1_PG [15:0]
0000_0000
_1101_100
Register 0484h AIF1 DAC1 EQ Band 1 PG
0
DESCRIPTION
EQ Band 1 Coefficient PG
REGISTER
ADDRESS
R1157
(0485h)
AIF1 DAC1
EQ Band 2
A
15:0 AIF1DAC1_EQ
_B2_A [15:0]
Register 0485h AIF1 DAC1 EQ Band 2 A
DESCRIPTION
0001_1110
_1011_010
1
EQ Band 2 Coefficient A w PP, August 2012, Rev 3.4
326
Pre-Production
REGISTER
ADDRESS
R1158
(0486h)
AIF1 DAC1
EQ Band 2
B
15:0 AIF1DAC1_EQ
_B2_B [15:0]
Register 0486h AIF1 DAC1 EQ Band 2 B
DESCRIPTION
1111_0001
_0100_010
1
EQ Band 2 Coefficient B
REGISTER
ADDRESS
R1159
(0487h)
AIF1 DAC1
EQ Band 2
C
15:0 AIF1DAC1_EQ
_B2_C [15:0]
Register 0487h AIF1 DAC1 EQ Band 2 C
DESCRIPTION
0000_1011
_0111_010
1
EQ Band 2 Coefficient C
REGISTER
ADDRESS
R1160
(0488h)
AIF1 DAC1
EQ Band 2
PG
15:0 AIF1DAC1_EQ
_B2_PG [15:0]
0000_0001
_1100_010
Register 0488h AIF1 DAC1 EQ Band 2 PG
1
DESCRIPTION
EQ Band 2 Coefficient PG
REGISTER
ADDRESS
R1161
(0489h)
AIF1 DAC1
EQ Band 3
A
15:0 AIF1DAC1_EQ
_B3_A [15:0]
Register 0489h AIF1 DAC1 EQ Band 3 A
DESCRIPTION
0001_1100
_0101_100
0
EQ Band 3 Coefficient A
REGISTER
ADDRESS
R1162
(048Ah)
AIF1 DAC1
EQ Band 3
B
15:0 AIF1DAC1_EQ
_B3_B [15:0]
Register 048Ah AIF1 DAC1 EQ Band 3 B
DESCRIPTION
1111_0011
_0111_001
1
EQ Band 3 Coefficient B
REGISTER
ADDRESS
R1163
(048Bh)
AIF1 DAC1
EQ Band 3
C
15:0 AIF1DAC1_EQ
_B3_C [15:0]
Register 048Bh AIF1 DAC1 EQ Band 3 C
DESCRIPTION
0000_1010
_0101_010
0
EQ Band 3 Coefficient C w
WM8958
PP, August 2012, Rev 3.4
327
WM8958
Pre-Production
REGISTER
ADDRESS
R1164
(048Ch)
AIF1 DAC1
EQ Band 3
PG
15:0 AIF1DAC1_EQ
_B3_PG [15:0]
0000_0101
_0101_100
Register 048Ch AIF1 DAC1 EQ Band 3 PG
0
DESCRIPTION
EQ Band 3 Coefficient PG
REGISTER
ADDRESS
R1165
(048Dh)
AIF1 DAC1
EQ Band 4
A
15:0 AIF1DAC1_EQ
_B4_A [15:0]
Register 048Dh AIF1 DAC1 EQ Band 4 A
DESCRIPTION
0001_0110
_1000_111
0
EQ Band 4 Coefficient A
REGISTER
ADDRESS
R1166
(048Eh)
AIF1 DAC1
EQ Band 4
B
15:0 AIF1DAC1_EQ
_B4_B [15:0]
Register 048Eh AIF1 DAC1 EQ Band 4 B
DESCRIPTION
1111_1000
_0010_100
1
EQ Band 4 Coefficient B
REGISTER
ADDRESS
R1167
(048Fh)
AIF1 DAC1
EQ Band 4
C
15:0 AIF1DAC1_EQ
_B4_C [15:0]
Register 048Fh AIF1 DAC1 EQ Band 4 C
DESCRIPTION
0000_0111
_1010_110
1
EQ Band 4 Coefficient C
REGISTER
ADDRESS
R1168
(0490h)
AIF1 DAC1
EQ Band 4
PG
15:0 AIF1DAC1_EQ
_B4_PG [15:0]
0001_0001
_0000_001
Register 0490h AIF1 DAC1 EQ Band 4 PG
1
DESCRIPTION
EQ Band 4 Coefficient PG
REGISTER
ADDRESS
R1169
(0491h)
AIF1 DAC1
EQ Band 5
A
15:0 AIF1DAC1_EQ
_B5_A [15:0]
Register 0491h AIF1 DAC1 EQ Band 5 A
DESCRIPTION
0000_0101
_0110_010
0
EQ Band 5 Coefficient A w PP, August 2012, Rev 3.4
328
Pre-Production
REGISTER
ADDRESS
R1170
(0492h)
AIF1 DAC1
EQ Band 5
B
15:0 AIF1DAC1_EQ
_B5_B [15:0]
Register 0492h AIF1 DAC1 EQ Band 5 B
DESCRIPTION
0000_0101
_0101_100
1
EQ Band 5 Coefficient B
REGISTER
ADDRESS
R1171
(0493h)
AIF1 DAC1
EQ Band 5
PG
15:0 AIF1DAC1_EQ
_B5_PG [15:0]
0100_0000
_0000_000
Register 0493h AIF1 DAC1 EQ Band 5 PG
0
DESCRIPTION
EQ Band 5 Coefficient PG
REGISTER
ADDRESS
R1184
(04A0h)
AIF1 DAC2
EQ Gains
(1)
15:11 AIF1DAC2_EQ
_B1_GAIN [4:0]
10:6 AIF1DAC2_EQ
_B2_GAIN [4:0]
5:1 AIF1DAC2_EQ
_B3_GAIN [4:0]
0 AIF1DAC2_EQ
_ENA
Register 04A0h AIF1 DAC2 EQ Gains (1)
DESCRIPTION
0_1100 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 1 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 2 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 3 Gain
-12dB to +12dB in 1dB steps
0 Enable EQ in AIF1DAC2 playback path (AIF1, Timeslot
1)
0 = Disabled
1 = Enabled
REGISTER
ADDRESS
R1185
(04A1h)
AIF1 DAC2
EQ Gains
(2)
15:11 AIF1DAC2_EQ
_B4_GAIN [4:0]
10:6 AIF1DAC2_EQ
_B5_GAIN [4:0]
0 AIF1DAC2_EQ
_MODE
Register 04A1h AIF1 DAC2 EQ Gains (2)
DESCRIPTION
0_1100 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 4 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 5 Gain
-12dB to +12dB in 1dB steps
0 AIF1DAC2 (AIF1, Timeslot 1) EQ Band 1 Mode
0 = Shelving filter
1 = Peak filter
REGISTER
ADDRESS
R1186
(04A2h)
AIF1 DAC2
EQ Band 1
A
15:0 AIF1DAC2_EQ
_B1_A [15:0]
Register 04A2h AIF1 DAC2 EQ Band 1 A
DESCRIPTION
0000_1111
_1100_101
0
EQ Band 1 Coefficient A
WM8958 w PP, August 2012, Rev 3.4
329
WM8958
Pre-Production
REGISTER
ADDRESS
R1187
(04A3h)
AIF1 DAC2
EQ Band 1
B
15:0 AIF1DAC2_EQ
_B1_B [15:0]
Register 04A3h AIF1 DAC2 EQ Band 1 B
DESCRIPTION
0000_0100
_0000_000
0
EQ Band 1 Coefficient B
REGISTER
ADDRESS
R1188
(04A4h)
AIF1 DAC2
EQ Band 1
PG
15:0 AIF1DAC2_EQ
_B1_PG [15:0]
0000_0000
_1101_100
Register 04A4h AIF1 DAC2 EQ Band 1 PG
0
DESCRIPTION
EQ Band 1 Coefficient PG
REGISTER
ADDRESS
R1189
(04A5h)
AIF1 DAC2
EQ Band 2
A
15:0 AIF1DAC2_EQ
_B2_A [15:0]
Register 04A5h AIF1 DAC2 EQ Band 2 A
DESCRIPTION
0001_1110
_1011_010
1
EQ Band 2 Coefficient A
REGISTER
ADDRESS
R1190
(04A6h)
AIF1 DAC2
EQ Band 2
B
15:0 AIF1DAC2_EQ
_B2_B [15:0]
Register 04A6h AIF1 DAC2 EQ Band 2 B
DESCRIPTION
1111_0001
_0100_010
1
EQ Band 2 Coefficient B
REGISTER
ADDRESS
R1191
(04A7h)
AIF1 DAC2
EQ Band 2
C
15:0 AIF1DAC2_EQ
_B2_C [15:0]
DESCRIPTION
0000_1011
_0111_010
1
EQ Band 2 Coefficient C
Register 04A7h AIF1 DAC2 EQ Band 2 C
REGISTER
ADDRESS
R1192
(04A8h)
AIF1 DAC2
EQ Band 2
PG
15:0 AIF1DAC2_EQ
_B2_PG [15:0]
0000_0001
_1100_010
Register 04A8h AIF1 DAC2 EQ Band 2 PG
1
DESCRIPTION
EQ Band 2 Coefficient PG w PP, August 2012, Rev 3.4
330
Pre-Production
REGISTER
ADDRESS
R1193
(04A9h)
AIF1 DAC2
EQ Band 3
A
15:0 AIF1DAC2_EQ
_B3_A [15:0]
Register 04A9h AIF1 DAC2 EQ Band 3 A
DESCRIPTION
0001_1100
_0101_100
0
EQ Band 3 Coefficient A
REGISTER
ADDRESS
R1194
(04AAh)
AIF1 DAC2
EQ Band 3
B
15:0 AIF1DAC2_EQ
_B3_B [15:0]
Register 04AAh AIF1 DAC2 EQ Band 3 B
DESCRIPTION
1111_0011
_0111_001
1
EQ Band 3 Coefficient B
REGISTER
ADDRESS
R1195
(04ABh)
AIF1 DAC2
EQ Band 3
C
15:0 AIF1DAC2_EQ
_B3_C [15:0]
Register 04ABh AIF1 DAC2 EQ Band 3 C
DESCRIPTION
0000_1010
_0101_010
0
EQ Band 3 Coefficient C
REGISTER
ADDRESS
R1196
(04ACh)
AIF1 DAC2
EQ Band 3
PG
15:0 AIF1DAC2_EQ
_B3_PG [15:0]
0000_0101
_0101_100
Register 04ACh AIF1 DAC2 EQ Band 3 PG
0
DESCRIPTION
EQ Band 3 Coefficient PG
REGISTER
ADDRESS
R1197
(04ADh)
AIF1 DAC2
EQ Band 4
A
15:0 AIF1DAC2_EQ
_B4_A [15:0]
Register 04ADh AIF1 DAC2 EQ Band 4 A
DESCRIPTION
0001_0110
_1000_111
0
EQ Band 4 Coefficient A
REGISTER
ADDRESS
R1198
(04AEh)
AIF1 DAC2
EQ Band 4
B
15:0 AIF1DAC2_EQ
_B4_B [15:0]
Register 04AEh AIF1 DAC2 EQ Band 4 B
DESCRIPTION
1111_1000
_0010_100
1
EQ Band 4 Coefficient B w
WM8958
PP, August 2012, Rev 3.4
331
WM8958
Pre-Production
REGISTER
ADDRESS
R1199
(04AFh)
AIF1 DAC2
EQ Band 4
C
15:0 AIF1DAC2_EQ
_B4_C [15:0]
Register 04AFh AIF1 DAC2 EQ Band 4 C
DESCRIPTION
0000_0111
_1010_110
1
EQ Band 4 Coefficient C
REGISTER
ADDRESS
R1200
(04B0h)
AIF1 DAC2
EQ Band 4
PG
15:0 AIF1DAC2_EQ
_B4_PG [15:0]
0001_0001
_0000_001
Register 04B0h AIF1 DAC2 EQ Band 4 PG
1
DESCRIPTION
EQ Band 4 Coefficient PG
REGISTER
ADDRESS
R1201
(04B1h)
AIF1 DAC2
EQ Band 5
A
15:0 AIF1DAC2_EQ
_B5_A [15:0]
Register 04B1h AIF1 DAC2 EQ Band 5 A
DESCRIPTION
0000_0101
_0110_010
0
EQ Band 5 Coefficient A
REGISTER
ADDRESS
R1202
(04B2h)
AIF1 DAC2
EQ Band 5
B
15:0 AIF1DAC2_EQ
_B5_B [15:0]
Register 04B2h AIF1 DAC2 EQ Band 5 B
DESCRIPTION
0000_0101
_0101_100
1
EQ Band 5 Coefficient B
REGISTER
ADDRESS
R1203
(04B3h)
AIF1 DAC2
EQ Band 5
PG
15:0 AIF1DAC2_EQ
_B5_PG [15:0]
0100_0000
_0000_000
Register 04B3h AIF1 DAC2 EQ Band 5 PG
0
DESCRIPTION
EQ Band 5 Coefficient PG w PP, August 2012, Rev 3.4
332
Pre-Production
REGISTER
ADDRESS
R1280
(0500h)
AIF2 ADC
Left Volume
DESCRIPTION
8 AIF2ADC_VU 0 AIF2ADC output path Volume Update
Writing a 1 to this bit will cause the AIF2ADCL and
AIF2ADCR volume to be updated simultaneously
7:0 AIF2ADCL_VO
L [7:0]
1100_0000 AIF2ADC (Left) output path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0500h AIF2 ADC Left Volume
REGISTER
ADDRESS
R1281
(0501h)
AIF2 ADC
Right
Volume
DESCRIPTION
8 AIF2ADC_VU 0 AIF2ADC output path Volume Update
Writing a 1 to this bit will cause the AIF2ADCL and
AIF2ADCR volume to be updated simultaneously
7:0 AIF2ADCR_VO
L [7:0]
1100_0000 AIF2ADC (Right) output path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
EFh = +17.625dB
Register 0501h AIF2 ADC Right Volume
REGISTER
ADDRESS
R1282
(0502h)
AIF2 DAC
Left Volume
DESCRIPTION
8 AIF2DAC_VU 0 AIF2DAC input path Volume Update
Writing a 1 to this bit will cause the AIF2DACL and
AIF2DACR volume to be updated simultaneously
7:0 AIF2DACL_VO
L [7:0]
1100_0000 AIF2DAC (Left) input path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0502h AIF2 DAC Left Volume
REGISTER
ADDRESS
R1283
(0503h)
AIF2 DAC
Right
Volume
DESCRIPTION
8 AIF2DAC_VU 0 AIF2DAC input path Volume Update
Writing a 1 to this bit will cause the AIF2DACL and
AIF2DACR volume to be updated simultaneously
7:0 AIF2DACR_VO
L [7:0]
1100_0000 AIF2DAC (Right) input path Digital Volume
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
FFh = 0dB
Register 0503h AIF2 DAC Right Volume
WM8958 w PP, August 2012, Rev 3.4
333
WM8958
REGISTER
ADDRESS
R1296
(0510h)
AIF2 ADC
Filters
14:13 AIF2ADC_HPF
_CUT [1:0]
12 AIF2ADCL_HP
F
11 AIF2ADCR_HP
F
00
0
0
Pre-Production
DESCRIPTION
AIF2ADC output path Digital HPF Cut-Off Frequency
(fc)
00 = Hi-fi mode (fc = 4Hz at fs = 48kHz)
01 = Voice mode 1 (fc = 127Hz at fs = 8kHz)
10 = Voice mode 2 (fc = 130Hz at fs = 8kHz)
11 = Voice mode 3 (fc = 267Hz at fs = 8kHz)
AIF2ADC (Left) output path Digital HPF Enable
0 = Disabled
1 = Enabled
AIF2ADC (Right) output path Digital HPF Enable
0 = Disabled
1 = Enabled
Register 0510h AIF2 ADC Filters
REGISTER
ADDRESS
R1312
(0520h)
AIF2 DAC
Filters (1)
9 AIF2DAC_MUT
E
7 AIF2DAC_MO
NO
5 AIF2DAC_MUT
ERATE
4 AIF2DAC_UN
MUTE_RAMP
1
0
0
0
DESCRIPTION
AIF2DAC input path Soft Mute Control
0 = Un-mute
1 = Mute
AIF2DAC input path Mono Mix Control
0 = Disabled
1 = Enabled
AIF2DAC input path Soft Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is 10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is 171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
AIF2DAC input path Unmute Ramp select
0 = Disabling soft-mute (AIF2DAC_MUTE=0) will cause the volume to change immediately to AIF2DACL_VOL and AIF2DACR_VOL settings
1 = Disabling soft-mute (AIF2DAC_MUTE=0) will cause the DAC volume to ramp up gradually to the
AIF2DACL_VOL and AIF2DACR_VOL settings
Register 0520h AIF2 DAC Filters (1)
REGISTER
ADDRESS
R1313
(0521h)
AIF2 DAC
Filters (2)
13:9 AIF2DAC_3D_
GAIN [4:0]
8 AIF2DAC_3D_
ENA
DESCRIPTION
0_0000 AIF2DAC playback path 3D Stereo depth
00000 = Off
00001 = Minimum (-16dB)
…(0.915dB steps)
11111 = Maximum (+11.45dB)
0 Enable 3D Stereo in AIF2DAC playback path
0 = Disabled
1 = Enabled
Register 0521h AIF2 DAC Filters (2) w PP, August 2012, Rev 3.4
334
Pre-Production
REGISTER
ADDRESS
R1328
(0430h)
AIF2 DAC
Noise Gate
6:5 AIF2DAC_NG_
HLD [1:0]
3:1 AIF2DAC_NG_
THR [2:0]
0 AIF2DAC_NG_
ENA
11
100
0
DESCRIPTION
AIF2DAC input path Noise Gate Hold Time
(delay before noise gate is activated)
00 = 30ms
01 = 125ms
10 = 250ms
11 = 500ms
AIF2DAC input path Noise Gate Threshold
000 = -60dB
001 = -66dB
010 = -72dB
011 = -78dB
100 = -84dB
101 = -90dB
110 = -96dB
111 = -102dB
AIF2DAC input path Noise Gate Enable
0 = Disabled
1 = Enabled
WM8958
Register 0530h AIF2 DAC Noise Gate
REGISTER
ADDRESS
R1344
(0540h)
AIF2 DRC
(1)
DESCRIPTION
15:11 AIF2DRC_SIG
_DET_RMS
[4:0]
10:9 AIF2DRC_SIG
_DET_PK [1:0]
0_0000 AIF2 DRC Signal Detect RMS Threshold.
This is the RMS signal level for signal detect to be indicated when AIF2DRC_SIG_DET_MODE=1.
00000 = -30dB
00001 = -31.5dB
…. (1.5dB steps)
00
11110 = -75dB
11111 = -76.5dB
AIF2 DRC Signal Detect Peak Threshold.
This is the Peak/RMS ratio, or Crest Factor, level for signal detect to be indicated when
AIF2DRC_SIG_DET_MODE=0.
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
8 AIF2DRC_NG_
ENA
7 AIF2DRC_SIG
_DET_MODE
6 AIF2DRC_SIG
_DET
5 AIF2DRC_KNE
E2_OP_ENA
0
1
0
0
AIF2 DRC Noise Gate Enable
0 = Disabled
1 = Enabled
AIF2 DRC Signal Detect Mode
0 = Peak threshold mode
1 = RMS threshold mode
AIF2 DRC Signal Detect Enable
0 = Disabled
1 = Enabled
AIF2 DRC KNEE2_OP Enable
0 = Disabled
1 = Enabled
4 AIF2DRC_QR 1 AIF2 DRC Quick-release Enable
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
335
WM8958
REGISTER
ADDRESS
Pre-Production
DESCRIPTION
3 AIF2DRC_ANT
ICLIP
2 AIF2DAC_DRC
_ENA
1 AIF2ADCL_DR
C_ENA
0 AIF2ADCR_DR
C_ENA
1
0
0
0
AIF2 DRC Anti-clip Enable
0 = Disabled
1 = Enabled
Enable DRC in AIF2DAC playback path
0 = Disabled
1 = Enabled
Enable DRC in AIF2ADC (Left) record path
0 = Disabled
1 = Enabled
Enable DRC in AIF2ADC (Right) record path
0 = Disabled
1 = Enabled
Register 0540h AIF2 DRC (1)
REGISTER
ADDRESS
R1345
(0541h)
AIF2 DRC
(2)
12:9 AIF2DRC_ATK
[3:0]
4:2 AIF2DRC_MIN
GAIN [2:0]
1:0 AIF2DRC_MAX
GAIN [1:0]
0100
8:5 AIF2DRC_DCY
[3:0]
0010
001
01
DESCRIPTION
AIF2 DRC Gain attack rate (seconds/6dB)
0000 = Reserved
0001 = 181us
0010 = 363us
0011 = 726us
0100 = 1.45ms
0101 = 2.9ms
0110 = 5.8ms
0111 = 11.6ms
1000 = 23.2ms
1001 = 46.4ms
1010 = 92.8ms
1011 = 185.6ms
1100-1111 = Reserved
AIF2 DRC Gain decay rate (seconds/6dB)
0000 = 186ms
0001 = 372ms
0010 = 743ms
0011 = 1.49s
0100 = 2.97s
0101 = 5.94s
0110 = 11.89s
0111 = 23.78s
1000 = 47.56s
1001-1111 = Reserved
AIF2 DRC Minimum gain to attenuate audio signals
000 = 0dB
001 = -12dB (default)
010 = -18dB
011 = -24dB
100 = -36dB
101 = Reserved
11X = Reserved
AIF2 DRC Maximum gain to boost audio signals (dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 36dB
Register 0541h AIF2 DRC (2) w PP, August 2012, Rev 3.4
336
Pre-Production
REGISTER
ADDRESS
R1346
(0542h)
AIF2 DRC
(3)
15:12 AIF2DRC_NG_
MINGAIN [3:0]
11:10 AIF2DRC_NG_
EXP [1:0]
9:8 AIF2DRC_QR_
THR [1:0]
7:6 AIF2DRC_QR_
DCY [1:0]
5:3 AIF2DRC_HI_
COMP [2:0]
2:0 AIF2DRC_LO_
COMP [2:0]
DESCRIPTION
0000
00
00
00
000
000
AIF2 DRC Minimum gain to attenuate audio signals when the noise gate is active.
0000 = -36dB
0001 = -30dB
0010 = -24dB
0011 = -18dB
0100 = -12dB
0101 = -6dB
0110 = 0dB
0111 = 6dB
1000 = 12dB
1001 = 18dB
1010 = 24dB
1011 = 30dB
1100 = 36dB
1101 to 1111 = Reserved
AIF2 DRC Noise Gate slope
00 = 1 (no expansion)
01 = 2
10 = 4
11 = 8
AIF2 DRC Quick-release threshold (crest factor in dB)
00 = 12dB
01 = 18dB
10 = 24dB
11 = 30dB
AIF2 DRC Quick-release decay rate (seconds/6dB)
00 = 0.725ms
01 = 1.45ms
10 = 5.8ms
11 = Reserved
AIF2 DRC Compressor slope (upper region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 1/16
101 = 0
110 = Reserved
111 = Reserved
AIF2 DRC Compressor slope (lower region)
000 = 1 (no compression)
001 = 1/2
010 = 1/4
011 = 1/8
100 = 0
101 = Reserved
11X = Reserved
Register 0542h AIF2 DRC (3)
WM8958 w PP, August 2012, Rev 3.4
337
WM8958
REGISTER
ADDRESS
R1347
(0543h)
AIF2 DRC
(4)
10:5 AIF2DRC_KNE
E_IP [5:0]
4:0 AIF2DRC_KNE
E_OP [4:0]
Pre-Production
DESCRIPTION
00_0000 AIF2 DRC Input signal level at the Compressor ‘Knee’.
000000 = 0dB
000001 = -0.75dB
000010 = -1.5dB
… (-0.75dB steps)
111100 = -45dB
111101 = Reserved
11111X = Reserved
0_0000 AIF2 DRC Output signal at the Compressor ‘Knee’.
00000 = 0dB
00001 = -0.75dB
00010 = -1.5dB
… (-0.75dB steps)
11110 = -22.5dB
11111 = Reserved
Register 0543h AIF2 DRC (4)
REGISTER
ADDRESS
R1348
(0544h)
AIF2 DRC
(5)
DESCRIPTION
9:5 AIF2DRC_KNE
E2_IP [4:0]
4:0 AIF2DRC_KNE
E2_OP [4:0]
0_0000 AIF2 DRC Input signal level at the Noise Gate threshold
‘Knee2’.
00000 = -36dB
00001 = -37.5dB
00010 = -39dB
… (-1.5dB steps)
11110 = -81dB
11111 = -82.5dB
Only applicable when AIF2DRC_NG_ENA = 1.
0_0000 AIF2 DRC Output signal at the Noise Gate threshold
‘Knee2’.
00000 = -30dB
00001 = -31.5dB
00010 = -33dB
… (-1.5dB steps)
11110 = -75dB
11111 = -76.5dB
Only applicable when AIF2DRC_KNEE2_OP_ENA = 1.
Register 0544h AIF2 DRC (5)
REGISTER
ADDRESS
R1408
(0580h)
AIF2 EQ
Gains (1)
15:11 AIF2DAC_EQ_
B1_GAIN [4:0]
10:6 AIF2DAC_EQ_
B2_GAIN [4:0]
5:1 AIF2DAC_EQ_
B3_GAIN [4:0]
0 AIF2DAC_EQ_
ENA
DESCRIPTION
0_1100 AIF2 EQ Band 1 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF2 EQ Band 2 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF2 EQ Band 3 Gain
-12dB to +12dB in 1dB steps
0 Enable EQ in AIF2DAC playback path
0 = Disabled
1 = Enabled
Register 0580h AIF2 EQ Gains (1) w PP, August 2012, Rev 3.4
338
Pre-Production
REGISTER
ADDRESS
R1409
(0581h)
AIF2 EQ
Gains (2)
15:11 AIF2DAC_EQ_
B4_GAIN [4:0]
10:6 AIF2DAC_EQ_
B5_GAIN [4:0]
0 AIF2DAC_EQ_
MODE
DESCRIPTION
0_1100 AIF2 EQ Band 4 Gain
-12dB to +12dB in 1dB steps
0_1100 AIF2 EQ Band 5 Gain
-12dB to +12dB in 1dB steps
0 AIF2 EQ Band 1 Mode
0 = Shelving filter
1 = Peak filter
Register 0581h AIF2 EQ Gains (2)
REGISTER
ADDRESS
R1410
(0582h)
AIF2 EQ
Band 1 A
15:0 AIF2DAC_EQ_
B1_A [15:0]
Register 0582h AIF2 EQ Band 1 A
DESCRIPTION
0000_1111
_1100_101
0
EQ Band 1 Coefficient A
REGISTER
ADDRESS
R1411
(0583h)
AIF2 EQ
Band 1 B
15:0 AIF2DAC_EQ_
B1_B [15:0]
Register 0583h AIF2 EQ Band 1 B
DESCRIPTION
0000_0100
_0000_000
0
EQ Band 1 Coefficient B
REGISTER
ADDRESS
R1412
(0584h)
AIF2 EQ
Band 1 PG
15:0 AIF2DAC_EQ_
B1_PG [15:0]
Register 0584h AIF2 EQ Band 1 PG
DESCRIPTION
0000_0000
_1101_100
0
EQ Band 1 Coefficient PG
REGISTER
ADDRESS
R1413
(0585h)
AIF2 EQ
Band 2 A
15:0 AIF2DAC_EQ_
B2_A [15:0]
Register 0585h AIF2 EQ Band 2 A
DESCRIPTION
0001_1110
_1011_010
1
EQ Band 2 Coefficient A
REGISTER
ADDRESS
R1414
(0586h)
AIF2 EQ
Band 2 B
15:0 AIF2DAC_EQ_
B2_B [15:0]
Register 0586h AIF2 EQ Band 2 B
DESCRIPTION
1111_0001
_0100_010
1
EQ Band 2 Coefficient B w
WM8958
PP, August 2012, Rev 3.4
339
WM8958
Pre-Production
REGISTER
ADDRESS
R1415
(0587h)
AIF2 EQ
Band 2 C
15:0 AIF2DAC_EQ_
B2_C [15:0]
Register 0587h AIF2 EQ Band 2 C
DESCRIPTION
0000_1011
_0111_010
1
EQ Band 2 Coefficient C
REGISTER
ADDRESS
R1416
(0588h)
AIF2 EQ
Band 2 PG
15:0 AIF2DAC_EQ_
B2_PG [15:0]
Register 0588h AIF2 EQ Band 2 PG
DESCRIPTION
0000_0001
_1100_010
1
EQ Band 2 Coefficient PG
REGISTER
ADDRESS
R1417
(0589h)
AIF2 EQ
Band 3 A
15:0 AIF2DAC_EQ_
B3_A [15:0]
Register 0589h AIF2 EQ Band 3 A
DESCRIPTION
0001_1100
_0101_100
0
EQ Band 3 Coefficient A
REGISTER
ADDRESS
R1418
(058Ah)
AIF2 EQ
Band 3 B
15:0 AIF2DAC_EQ_
B3_B [15:0]
Register 058Ah AIF2 EQ Band 3 B
DESCRIPTION
1111_0011
_0111_001
1
EQ Band 3 Coefficient B
REGISTER
ADDRESS
R1419
(058Bh)
AIF2 EQ
Band 3 C
15:0 AIF2DAC_EQ_
B3_C [15:0]
Register 058Bh AIF2 EQ Band 3 C
DESCRIPTION
0000_1010
_0101_010
0
EQ Band 3 Coefficient C
REGISTER
ADDRESS
R1420
(058Ch)
AIF2 EQ
Band 3 PG
15:0 AIF2DAC_EQ_
B3_PG [15:0]
Register 058Ch AIF2 EQ Band 3 PG
DESCRIPTION
0000_0101
_0101_100
0
EQ Band 3 Coefficient PG w PP, August 2012, Rev 3.4
340
Pre-Production
REGISTER
ADDRESS
R1421
(058Dh)
AIF2 EQ
Band 4 A
15:0 AIF2DAC_EQ_
B4_A [15:0]
Register 058Dh AIF2 EQ Band 4 A
DESCRIPTION
0001_0110
_1000_111
0
EQ Band 4 Coefficient A
REGISTER
ADDRESS
R1422
(058Eh)
AIF2 EQ
Band 4 B
15:0 AIF2DAC_EQ_
B4_B [15:0]
Register 058Eh AIF2 EQ Band 4 B
DESCRIPTION
1111_1000
_0010_100
1
EQ Band 4 Coefficient B
REGISTER
ADDRESS
R1423
(058Fh)
AIF2 EQ
Band 4 C
15:0 AIF2DAC_EQ_
B4_C [15:0]
Register 058Fh AIF2 EQ Band 4 C
DESCRIPTION
0000_0111
_1010_110
1
EQ Band 4 Coefficient C
REGISTER
ADDRESS
R1424
(0590h)
AIF2 EQ
Band 4 PG
15:0 AIF2DAC_EQ_
B4_PG [15:0]
Register 0590h AIF2 EQ Band 4 PG
DESCRIPTION
0001_0001
_0000_001
1
EQ Band 4 Coefficient PG
REGISTER
ADDRESS
R1425
(0591h)
AIF2 EQ
Band 5 A
15:0 AIF2DAC_EQ_
B5_A [15:0]
Register 0591h AIF2 EQ Band 5 A
DESCRIPTION
0000_0101
_0110_010
0
EQ Band 5 Coefficient A
REGISTER
ADDRESS
R1426
(0592h)
AIF2 EQ
Band 5 B
15:0 AIF2DAC_EQ_
B5_B [15:0]
Register 0592h AIF2 EQ Band 5 B
DESCRIPTION
0000_0101
_0101_100
1
EQ Band 5 Coefficient B w
WM8958
PP, August 2012, Rev 3.4
341
WM8958
Pre-Production
REGISTER
ADDRESS
R1427
(0593h)
AIF2 EQ
Band 5 PG
15:0 AIF2DAC_EQ_
B5_PG [15:0]
Register 0593h AIF2 EQ Band 5 PG
DESCRIPTION
0100_0000
_0000_000
0
EQ Band 5 Coefficient PG
REGISTER
ADDRESS
R1536
(0600h)
DAC1 Mixer
Volumes
8:5 ADCR_DAC1_
VOL [3:0]
DESCRIPTION
3:0 ADCL_DAC1_
VOL [3:0]
0000
0000
Sidetone STR to DAC1L and DAC1R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
Sidetone STL to DAC1L and DAC1R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
Register 0600h DAC1 Mixer Volumes
REGISTER
ADDRESS
R1537
(0601h)
DAC1 Left
Mixer
Routing
5 ADCR_TO_DA
C1L
4 ADCL_TO_DA
C1L
2 AIF2DACL_TO
_DAC1L
1 AIF1DAC2L_T
O_DAC1L
0 AIF1DAC1L_T
O_DAC1L
0
0
0
0
0
DESCRIPTION
Enable Sidetone STR to DAC1L
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC1L
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to DAC1L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Left) to DAC1L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Left) to DAC1L
0 = Disabled
1 = Enabled
Register 0601h DAC1 Left Mixer Routing
REGISTER
ADDRESS
R1538
(0602h)
DAC1 Right
Mixer
Routing
5 ADCR_TO_DA
C1R
4 ADCL_TO_DA
C1R
0
0
DESCRIPTION
Enable Sidetone STR to DAC1R
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC1R
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
342
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
2 AIF2DACR_TO
_DAC1R
1 AIF1DAC2R_T
O_DAC1R
0 AIF1DAC1R_T
O_DAC1R
Register 0602h DAC1 Right Mixer Routing
REGISTER
ADDRESS
R1539
(0603h)
DAC2 Mixer
Volumes
8:5 ADCR_DAC2_
VOL [3:0]
3:0 ADCL_DAC2_
VOL [3:0]
DESCRIPTION
0000
0000
Sidetone STR to DAC2L and DAC2R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
Sidetone STL to DAC2L and DAC2R Volume
0000 = -36dB
0001 = -33dB
…. (3dB steps)
1011 = -3dB
1100 = 0dB
Register 0603h DAC2 Mixer Volumes
REGISTER
ADDRESS
R1540
(0604h)
DAC2 Left
Mixer
Routing
5 ADCR_TO_DA
C2L
4 ADCL_TO_DA
C2L
2 AIF2DACL_TO
_DAC2L
1 AIF1DAC2L_T
O_DAC2L
0 AIF1DAC1L_T
O_DAC2L
0
0
0
0
0
0
0
0
Enable AIF2 (Right) to DAC1R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Right) to DAC1R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Right) to DAC1R
0 = Disabled
1 = Enabled
DESCRIPTION
Enable Sidetone STR to DAC2L
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC2L
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to DAC2L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Left) to DAC2L
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Left) to DAC2L
0 = Disabled
1 = Enabled
Register 0604h DAC2 Left Mixer Routing
WM8958 w PP, August 2012, Rev 3.4
343
WM8958
REGISTER
ADDRESS
R1541
(0605h)
DAC2 Right
Mixer
Routing
5 ADCR_TO_DA
C2R
4 ADCL_TO_DA
C2R
2 AIF2DACR_TO
_DAC2R
1 AIF1DAC2R_T
O_DAC2R
0 AIF1DAC1R_T
O_DAC2R
Register 0605h DAC2 Right Mixer Routing
REGISTER
ADDRESS
R1542
(0606h)
AIF1 ADC1
Left Mixer
Routing
1 ADC1L_TO_AI
F1ADC1L
0 AIF2DACL_TO
_AIF1ADC1L
0
0
0
0
0
0
0
Pre-Production
DESCRIPTION
Enable Sidetone STR to DAC2R
0 = Disabled
1 = Enabled
Enable Sidetone STL to DAC2R
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to DAC2R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 1, Right) to DAC2R
0 = Disabled
1 = Enabled
Enable AIF1 (Timeslot 0, Right) to DAC2R
0 = Disabled
1 = Enabled
Enable ADCL / DMIC1 (Left) to AIF1 (Timeslot 0, Left) output
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to AIF1 (Timeslot 0, Left) output
0 = Disabled
1 = Enabled
DESCRIPTION
Register 0606h AIF1 ADC1 Left Mixer Routing
REGISTER
ADDRESS
R1543
(0607h)
AIF1 ADC1
Right Mixer
Routing
1 ADC1R_TO_AI
F1ADC1R
0 AIF2DACR_TO
_AIF1ADC1R
0
0
DESCRIPTION
Enable ADCR / DMIC1 (Right) to AIF1 (Timeslot 0,
Right) output
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to AIF1 (Timeslot 0, Right) output
0 = Disabled
1 = Enabled
Register 0607h AIF1 ADC1 Right Mixer Routing
REGISTER
ADDRESS
R1544
(0608h)
AIF1 ADC2
Left Mixer
Routing
1 ADC2L_TO_AI
F1ADC2L
0 AIF2DACL_TO
_AIF1ADC2L
0
0
DESCRIPTION
Enable DMIC2 (Left) to AIF1 (Timeslot 1, Left) output
0 = Disabled
1 = Enabled
Enable AIF2 (Left) to AIF1 (Timeslot 1, Left) output
0 = Disabled
1 = Enabled
Register 0608h AIF1 ADC2 Left Mixer Routing w PP, August 2012, Rev 3.4
344
Pre-Production
REGISTER
ADDRESS
R1545
(0609h)
AIF1 ADC2
Right mixer
Routing
1 ADC2R_TO_AI
F1ADC2R
0 AIF2DACR_TO
_AIF1ADC2R
0
0
DESCRIPTION
Enable DMIC2 (Right) to AIF1 (Timeslot 1, Right) output
0 = Disabled
1 = Enabled
Enable AIF2 (Right) to AIF1 (Timeslot 1, Right) output
0 = Disabled
1 = Enabled
Register 0609h AIF1 ADC2 Right mixer Routing
REGISTER
ADDRESS
R1552
(0610h)
DAC1 Left
Volume
DESCRIPTION
9 DAC1L_MUTE 1 DAC1L Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
8 DAC1_VU 0 DAC1L and DAC1R Volume Update
Writing a 1 to this bit will cause the DAC1L and DAC1R volume to be updated simultaneously
[7:0] 00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
Register 0610h DAC1 Left Volume
REGISTER
ADDRESS
R1553
(0611h)
DAC1 Right
Volume
DESCRIPTION
9 DAC1R_MUTE 1 DAC1R Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
8 DAC1_VU 0 DAC1L and DAC1R Volume Update
Writing a 1 to this bit will cause the DAC1L and DAC1R volume to be updated simultaneously
[7:0]
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
Register 0611h DAC1 Right Volume
WM8958 w PP, August 2012, Rev 3.4
345
WM8958
Pre-Production
REGISTER
ADDRESS
R1554
(0612h)
DAC2 Left
Volume
DESCRIPTION
9 DAC2L_MUTE 1 DAC2L Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
8 DAC2_VU 0 DAC2L and DAC2R Volume Update
Writing a 1 to this bit will cause the DAC2L and DAC2R volume to be updated simultaneously
[7:0]
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
Register 0612h DAC2 Left Volume
REGISTER
ADDRESS
R1555
(0613h)
DAC2 Right
Volume
DESCRIPTION
9 DAC2R_MUTE 1 DAC2R Soft Mute Control
0 = DAC Un-mute
1 = DAC Mute
8 DAC2_VU 0 DAC2L and DAC2R Volume Update
Writing a 1 to this bit will cause the DAC2L and DAC2R volume to be updated simultaneously
[7:0]
00h = MUTE
01h = -71.625dB
… (0.375dB steps)
C0h = 0dB
… (0.375dB steps)
E0h = 12dB
FFh = 12dB
Register 0613h DAC2 Right Volume
REGISTER
ADDRESS
R1556
(0614h)
DAC
Softmute
1 DAC_SOFTMU
TEMODE
0
DESCRIPTION
0 DAC_MUTERA
TE
0
DAC Unmute Ramp select
0 = Disabling soft-mute (DAC [1/2] [L/R]_MUTE=0) will cause the DAC volume to change immediately to DAC
[1/2] [L/R]_VOL settings
1 = Disabling soft-mute (DAC [1/2] [L/R]_MUTE=0) will cause the DAC volume to ramp up gradually to the DAC
[1/2] [L/R]_VOL settings
DAC Soft Mute Ramp Rate
0 = Fast ramp (fs/2, maximum ramp time is 10.7ms at fs=48k)
1 = Slow ramp (fs/32, maximum ramp time is 171ms at fs=48k)
(Note: ramp rate scales with sample rate.)
Register 0614h DAC Softmute w PP, August 2012, Rev 3.4
346
Pre-Production
REGISTER
ADDRESS
R1568
(0620h)
Oversamplin g
DESCRIPTION
1 ADC_OSR128 1 ADC / Digital Microphone Oversample Rate Select
0 = Low Power
1 = High Performance
0 DAC_OSR128 0 DAC Oversample Rate Select
0 = Low Power
1 = High Performance
Register 0620h Oversampling
REGISTER
ADDRESS
R1569
(0621h)
Sidetone
9:7 ST_HPF_CUT
[2:0]
000
DESCRIPTION
Sidetone HPF cut-off frequency (relative to 44.1kHz sample rate)
000 = 2.7kHz
001 = 1.35kHz
010 = 675Hz
011 = 370Hz
100 = 180Hz
101 = 90Hz
110 = 45Hz
111 = Reserved
Note - the cut-off frequencies scale with the Digital
Mixing (SYSCLK) clocking rate. The quoted figures apply to 44.1kHz sample rate.
0 = Disabled
1 = Enabled
1 STR_SEL 0 Select source for sidetone STR path
0 = ADCR / DMICDAT1 (Right)
1 = DMICDAT2 (Right)
0 = ADCL / DMICDAT1 (Left)
1 = DMICDAT2 (Left)
Register 0621h Sidetone
REGISTER
ADDRESS
R1792
(0700h)
GPIO 1
DESCRIPTION
15 GP1_DIR 1 GPIO1 Pin Direction
0 = Output
1 = Input
14 GP1_PU 0 GPIO1 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP1_PD 0 GPIO1 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP1_POL 0 GPIO1 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP1_OP_CFG 0 GPIO1 Output Configuration
0 = CMOS
1 = Open Drain w
WM8958
PP, August 2012, Rev 3.4
347
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
0 = Disabled
1 = Enabled
6 GP1_LVL 0 GPIO1 level. Write to this bit to set a GPIO output.
Read from this bit to read GPIO input level.
For output functions only, when GP1_POL is set, the register contains the opposite logic level to the external pin.
00h = ADCLRCLK1
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 0700h GPIO 1
REGISTER
ADDRESS
R1793
(0701h) Pull
Control
(MCLK2)
DESCRIPTION
14 MCLK2_PU 0 MCLK2 Pull-up enable
0 = Disabled
1 = Enabled
13 MCLK2_PD 1 MCLK2 Pull-down enable
0 = Disabled
1 = Enabled
Register 0701h Pull Control (MCLK2) w PP, August 2012, Rev 3.4
348
Pre-Production
REGISTER
ADDRESS
R1794
(0702h) Pull
Control
(BCLK2)
DESCRIPTION
14 BCLK2_PU 0 BCLK2 Pull-up enable
0 = Disabled
1 = Enabled
13 BCLK2_PD 1 BCLK2 Pull-down enable
0 = Disabled
1 = Enabled
Register 0702h Pull Control (BCLK2)
REGISTER
ADDRESS
R1795
(0703h) Pull
Control
(DACLRCLK
2)
14 DACLRCLK2_
PU
13 DACLRCLK2_
PD
0
1
DESCRIPTION
DACLRCLK2 Pull-up enable
0 = Disabled
1 = Enabled
DACLRCLK2 Pull-down enable
0 = Disabled
1 = Enabled
Register 0703h Pull Control (DACLRCLK2)
REGISTER
ADDRESS
R1796
(0704h) Pull
Control
(DACDAT2)
DESCRIPTION
14 DACDAT2_PU 0 DACDAT2 Pull-up enable
0 = Disabled
1 = Enabled
13 DACDAT2_PD 1 DACDAT2 Pull-down enable
0 = Disabled
1 = Enabled
Register 0704h Pull Control (DACDAT2)
REGISTER
ADDRESS
R1797
(0705h)
GPIO 6
DESCRIPTION
15 GP6_DIR 1 GPIO6 Pin Direction
0 = Output
1 = Input
14 GP6_PU 0 GPIO6 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP6_PD 1 GPIO6 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP6_POL 0 GPIO6 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP6_OP_CFG 0 GPIO6 Output Configuration
0 = CMOS
1 = Open Drain
0 = Disabled
1 = Enabled w
WM8958
PP, August 2012, Rev 3.4
349
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
6 GP6_LVL 0 GPIO6 level. Write to this bit to set a GPIO output.
Read from this bit to read GPIO input level.
For output functions only, when GP6_POL is set, the register contains the opposite logic level to the external pin.
00h = ADCLRCLK2
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 0705h GPIO 6
REGISTER
ADDRESS
R1799
(0707h)
GPIO 8
DESCRIPTION
15 GP8_DIR 1 GPIO8 Pin Direction
0 = Output
1 = Input
14 GP8_PU 0 GPIO8 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP8_PD 1 GPIO8 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP8_POL 0 GPIO8 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP8_OP_CFG 0 GPIO8 Output Configuration
0 = CMOS
1 = Open Drain
0 = Disabled
1 = Enabled
6 GP8_LVL 0 GPIO8 level. Write to this bit to set a GPIO output.
PP, August 2012, Rev 3.4
w
350
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
Read from this bit to read GPIO input level.
For output functions only, when GP8_POL is set, the register contains the opposite logic level to the external pin.
00h = DACDAT3
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 0707h GPIO 8
REGISTER
ADDRESS
R1800
(0708h)
GPIO 9
DESCRIPTION
15 GP9_DIR 1 GPIO9 Pin Direction
0 = Output
1 = Input
14 GP9_PU 0 GPIO9 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP9_PD 1 GPIO9 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP9_POL 0 GPIO9 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP9_OP_CFG 0 GPIO9 Output Configuration
0 = CMOS
1 = Open Drain
0 = Disabled
1 = Enabled
WM8958 w PP, August 2012, Rev 3.4
351
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
6 GP9_LVL 0 GPIO9 level. Write to this bit to set a GPIO output.
Read from this bit to read GPIO input level.
For output functions only, when GP9_POL is set, the register contains the opposite logic level to the external pin.
00h = ADCDAT3
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 0708h GPIO 9
REGISTER
ADDRESS
R1801
(0709h)
GPIO 10
DESCRIPTION
15 GP10_DIR 1 GPIO10 Pin Direction
0 = Output
1 = Input
14 GP10_PU 0 GPIO10 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP10_PD 1 GPIO10 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP10_POL 0 GPIO10 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP10_OP_CF
G
0 GPIO10 Output Configuration
0 = CMOS
1 = Open Drain
8 GP10_DB 1 GPIO10 Input De-bounce
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
352
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
6 GP10_LVL 0 GPIO10 level. Write to this bit to set a GPIO output.
Read from this bit to read GPIO input level.
For output functions only, when GP10_POL is set, the register contains the opposite logic level to the external pin.
00h = LRCLK3
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 0709h GPIO 10
REGISTER
ADDRESS
R1802
(070Ah)
GPIO 11
DESCRIPTION
15 GP11_DIR 1 GPIO11 Pin Direction
0 = Output
1 = Input
14 GP11_PU 0 GPIO11 Pull-Up Enable
0 = Disabled
1 = Enabled
13 GP11_PD 1 GPIO11 Pull-Down Enable
0 = Disabled
1 = Enabled
10 GP11_POL 0 GPIO11 Polarity Select
0 = Non-inverted (Active High)
1 = Inverted (Active Low)
9 GP11_OP_CF
G
0 GPIO11 Output Configuration
0 = CMOS
1 = Open Drain
8 GP11_DB 1 GPIO11 Input De-bounce
0 = Disabled
1 = Enabled w
WM8958
PP, August 2012, Rev 3.4
353
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
6 GP11_LVL 0 GPIO11 level. Write to this bit to set a GPIO output.
Read from this bit to read GPIO input level.
For output functions only, when GP11_POL is set, the register contains the opposite logic level to the external pin.
00h = BCLK3
01h = GPIO
02h = Reserved
03h = IRQ
04h = Temperature (Shutdown) status
05h = MICDET status
06h = Reserved
07h = Reserved
08h = Reserved
09h = FLL1 Lock
0Ah = FLL2 Lock
0Bh = SRC1 Lock
0Ch = SRC2 Lock
0Dh = AIF1 DRC1 Signal Detect
0Eh = AIF1 DRC2 Signal Detect
0Fh = AIF2 DRC Signal Detect
10h = Write Sequencer Status
11h = FIFO Error
12h = OPCLK Clock output
13h = Temperature (Warning) status
14h = DC Servo Done
15h = FLL1 Clock output
16h = FLL2 Clock output
17h to 1Fh = Reserved
Register 070Ah GPIO 11
REGISTER
ADDRESS
R1824
(0720h) Pull
Control (1)
DESCRIPTION
11 DMICDAT2_P
U
10 DMICDAT2_P
D
9 DMICDAT1_P
U
8 DMICDAT1_P
D
0
0
0
0
DMICDAT2 Pull-Up enable
0 = Disabled
1 = Enabled
DMICDAT2 Pull-Down enable
0 = Disabled
1 = Enabled
DMICDAT1 Pull-Up enable
0 = Disabled
1 = Enabled
DMICDAT1 Pull-Down enable
0 = Disabled
1 = Enabled
7 MCLK1_PU 0 MCLK1 Pull-up enable
0 = Disabled
1 = Enabled
6 MCLK1_PD 0 MCLK1 Pull-down enable
0 = Disabled
1 = Enabled w PP, August 2012, Rev 3.4
354
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
5 DACDAT1_PU 0 DACDAT1 Pull-up enable
0 = Disabled
1 = Enabled
4 DACDAT1_PD 0 DACDAT1 Pull-down enable
0 = Disabled
1 = Enabled
3 DACLRCLK1_
PU
0 LRCLK1 Pull-up enable
0 = Disabled
1 = Enabled
2 DACLRCLK1_
PD
0 LRCLK1 Pull-down enable
0 = Disabled
1 = Enabled
1 BCLK1_PU 0 BCLK1 Pull-up enable
0 = Disabled
1 = Enabled
0 BCLK1_PD 0 BCLK1 Pull-down enable
0 = Disabled
1 = Enabled
Register 0720h Pull Control (1)
REGISTER
ADDRESS
R1825
(0721h) Pull
Control (2)
DESCRIPTION
8 ADDR_PD 1 ADDR Pull-down enable
0 = Disabled
1 = Enabled
6 LDO2ENA_PD 1 LDO2ENA Pull-down enable
0 = Disabled
1 = Enabled
4 LDO1ENA_PD 1 LDO1ENA Pull-down enable
0 = Disabled
1 = Enabled
1 SPKMODE_PU 1 SPKMODE Pull-up enable
0 = Disabled
1 = Enabled
Register 0721h Pull Control (2)
REGISTER
ADDRESS
R1840
(0730h)
Interrupt
Status 1
DESCRIPTION
10 GP11_EINT 0 GPIO11 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
9 GP10_EINT 0 GPIO10 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
8 GP9_EINT 0 GPIO9 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
7 GP8_EINT 0 GPIO8 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written. w
WM8958
PP, August 2012, Rev 3.4
355
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
5 GP6_EINT 0 GPIO6 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
0 GP1_EINT 0 GPIO1 Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
Register 0730h Interrupt Status 1
REGISTER
ADDRESS
R1841
(0731h)
Interrupt
Status 2
DESCRIPTION
15 TEMP_WARN_
EINT
14 DCS_DONE_E
INT
13 WSEQ_DONE
_EINT
12 FIFOS_ERR_E
INT
11 AIF2DRC_SIG
_DET_EINT
10 AIF1DRC2_SI
G_DET_EINT
9 AIF1DRC1_SI
G_DET_EINT
8 SRC2_LOCK_
EINT
0
0
0
0
0
0
0
0
Temperature Warning Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
DC Servo Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
Write Sequencer Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
Digital Core FIFO Error Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF2 DRC Activity Detect Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF1 DRC2 (Timeslot 1) Activity Detect Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
AIF1 DRC1 (Timeslot 0) Activity Detect Interrupt
(Rising edge triggered)
Note: Cleared when a ‘1’ is written.
SRC2 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
7 SRC1_LOCK_
EINT
6 FLL2_LOCK_E
INT
0
0
SRC1 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
FLL2 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
5 FLL1_LOCK_E
INT
0 TEMP_SHUT_
EINT
0 FLL1 Lock Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
1 MICD_EINT 0 Microphone Detection Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
0 Temperature Shutdown Interrupt
(Rising and falling edge triggered)
Note: Cleared when a ‘1’ is written.
Register 0731h Interrupt Status 2 w PP, August 2012, Rev 3.4
356
Pre-Production
REGISTER
ADDRESS
R1842
(0732h)
Interrupt
Raw Status
2
15 TEMP_WARN_
STS
14 DCS_DONE_S
TS
13 WSEQ_DONE
_STS
12 FIFOS_ERR_S
TS
11 AIF2DRC_SIG
_DET_STS
10 AIF1DRC2_SI
G_DET_STS
9 AIF1DRC1_SI
G_DET_STS
8 SRC2_LOCK_
STS
7 SRC1_LOCK_
STS
6 FLL2_LOCK_S
TS
5 FLL1_LOCK_S
TS
0 TEMP_SHUT_
STS
0
0
0
0
0
0
0
0
0
0
0
0
DESCRIPTION
Temperature Warning status
0 = Temperature is below warning level
1 = Temperature is above warning level
DC Servo status
0 = DC Servo not complete
1 = DC Servo complete
Write Sequencer status
0 = Sequencer Busy (sequence in progress)
1 = Sequencer Idle
Digital Core FIFO Error status
0 = Normal operation
1 = FIFO Error
AIF2 DRC Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
AIF1 DRC2 (Timeslot 1) Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
AIF1 DRC1 (Timeslot 0) Signal Detect status
0 = Signal threshold not exceeded
1 = Signal threshold exceeded
SRC2 Lock status
0 = Not locked
1 = Locked
SRC1 Lock status
0 = Not locked
1 = Locked
FLL2 Lock status
0 = Not locked
1 = Locked
FLL1 Lock status
0 = Not locked
1 = Locked
Temperature Shutdown status
0 = Temperature is below shutdown level
1 = Temperature is above shutdown level
Register 0732h Interrupt Raw Status 2
REGISTER
ADDRESS
R1848
(0738h)
Interrupt
Status 1
Mask
DESCRIPTION
10 IM_GP11_EIN
T
9 IM_GP10_EIN
T
1
1
GPIO11 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
GPIO10 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
8 IM_GP9_EINT 1 GPIO9 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
7 IM_GP8_EINT 1 GPIO8 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt. w
WM8958
PP, August 2012, Rev 3.4
357
WM8958
Pre-Production
REGISTER
ADDRESS
DESCRIPTION
5 IM_GP6_EINT 1 GPIO6 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
0 IM_GP1_EINT 1 GPIO1 Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Register 0738h Interrupt Status 1 Mask
REGISTER
ADDRESS
R1849
(0739h)
Interrupt
Status 2
Mask
15 IM_TEMP_WA
RN_EINT
14 IM_DCS_DON
E_EINT
1
1
DESCRIPTION
13 IM_WSEQ_DO
NE_EINT
12 IM_FIFOS_ER
R_EINT
11 IM_AIF2DRC_
SIG_DET_EIN
T
10 IM_AIF1DRC2
_SIG_DET_EI
NT
9 IM_AIF1DRC1
_SIG_DET_EI
NT
8 IM_SRC2_LOC
K_EINT
7 IM_SRC1_LOC
K_EINT
6 IM_FLL2_LOC
K_EINT
5 IM_FLL1_LOC
K_EINT
1 IM_MICD_EIN
T
0 IM_TEMP_SH
UT_EINT
1
1
1
1
1
1
1
1
1
1
1
Temperature Warning Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
DC Servo Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Write Sequencer Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Digital Core FIFO Error Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF2 DRC Activity Detect Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF1 DRC2 (Timeslot 1) Activity Detect Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
AIF1 DRC1 (Timeslot 0) Activity Detect Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
SRC2 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
SRC1 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
FLL2 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
FLL1 Lock Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Microphone Detection Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Temperature Shutdown Interrupt mask.
0 = Do not mask interrupt.
1 = Mask interrupt.
Register 0739h Interrupt Status 2 Mask w PP, August 2012, Rev 3.4
358
Pre-Production
REGISTER
ADDRESS
R1856
(0740h)
Interrupt
Control
Register 0740h Interrupt Control
DESCRIPTION
0 = Do not mask interrupt.
1 = Mask interrupt.
REGISTER
ADDRESS
R1864
(0748h) IRQ
Debounce
5 TEMP_WARN_
DB
0 TEMP_SHUT_
DB
1
1
DESCRIPTION
Temperature Warning de-bounce
0 = Disabled
1 = Enabled
Thermal shutdown de-bounce
0 = Disabled
1 = Enabled
Register 0748h IRQ Debounce
REGISTER
ADDRESS
R2304
(0900h)
DSP2_Progr am
DESCRIPTION
0 DSP2_ENA 0 DSP2 Audio Processor Enable.
0 = Disabled
1 = Enabled
This bit must be set before the MBC is enabled. It must remain set whenever the MBC is enabled.
Register 0900h DSP2_Program
REGISTER
ADDRESS
R2305
(0901h)
DSP2_Confi g
DESCRIPTION
00 = AIF1DAC1 input path (AIF1, Timeslot 0)
01 = AIF1DAC2 input path (AIF1, Timeslot 1)
10 = AIF2DAC input path
11 = Reserved
0 MBC_ENA 0 MBC Enable
0 = Disabled
1 = Enabled
Register 0901h DSP2_Config
REGISTER
ADDRESS
R2573
(0A0Dh)
DSP2_Exec
Control
2
1
DSP2_STOP
DSP2_RUNR
0
0
DESCRIPTION
Stop the DSP2 audio processor
Writing a 1 to this bit will cause the DSP2 processor to stop processing audio data
Start the DSP2 audio processor
Writing a 1 to this bit will cause the DSP2 processor to start processing audio data
Register 0A0Dh DSP2_ExecControl w
WM8958
PP, August 2012, Rev 3.4
359
WM8958
APPLICATIONS INFORMATION
Pre-Production
RECOMMENDED EXTERNAL COMPONENTS
AUDIO INPUT PATHS
The WM8958 provides 8 analogue audio inputs. Each of these inputs is referenced to the internal DC reference, VMID. A DC blocking capacitor is required for each input pin used in the target application.
The choice of capacitor is determined by the filter that is formed between that capacitor and the input impedance of the input pin. The circuit is illustrated in Figure 84. w
Figure 84 Audio Input Path DC Blocking Capacitor
If the input impedance is known, and the cut-off frequency is known, then the minimum capacitor value may be derived easily. However, it can be seen from the representation in Figure 84 that the input impedance is not fixed in all applications but can vary with gain and boost amplifier settings.
The PGA input resistance for every gain setting is detailed in Table 153.
IN1L_VOL[4:0],
IN2L_VOL[4:0],
IN1R_VOL[4:0],
IN2R_VOL[4:0]
VOLUME
(dB)
INPUT RESISTANCE
(kΩ)
SINGLE-ENDED
MODE
DIFFERENTIAL
MODE
00000 -16.5 58 52.5
00001 -15.0 56.9 50.6
00010 -13.5 55.6 48.6
00011 -12.0 54.1 46.4
00100 -10.5 52.5 44.1
00101 -9.0 50.7 41.5
00110 -7.5 48.6 38.9
00111 -6.0 46.5 36.2
01000 -4.5 44.1 33.4
01001 -3.0 41.6 30.6
01010 -1.5 38.9 27.8
01100 +1.5 33.4 22.5
01101 +3.0 30.6 20.0
01110 +4.5 27.8 17.7
01111 +6.0 25.1 15.6
10000 +7.5 22.5 13.6
10001 +9.0 20.1 11.9
10010 +10.5 17.8 10.3
10011 +12.0 15.6 8.9
10100 +13.5 13.7 7.6
10101 +15.0 11.9 6.5
PP, August 2012, Rev 3.4
360
Pre-Production w
WM8958
IN1L_VOL[4:0],
IN2L_VOL[4:0],
IN1R_VOL[4:0],
IN2R_VOL[4:0]
VOLUME
(dB)
INPUT RESISTANCE
(kΩ)
SINGLE-ENDED
MODE
DIFFERENTIAL
MODE
10110 +16.5 10.3 5.6
10111 +18.0 8.9 4.8
11000 +19.5 7.7 4.1
11001 +21.0 6.6 3.5
11010 +22.5 5.6 2.9
11011 +24.0 4.8 2.5
11100 +25.5 4.1 2.1
11101 +27.0 3.5 1.8
11110 +28.5 2.9 1.5
11111 +30.0 2.5 1.3
Table 153 PGA Input Pin Resistance
The appropriate input capacitor may be selected using the PGA input resistance data provided in
Table 153, depending on the required PGA gain setting(s).
The choice of capacitor for a 20Hz cut-off frequency is shown in Table 154 for a selection of typical input impedance conditions.
INPUT IMPEDANCE MINIMUM CAPACITANCE
FOR 20HZ PASS BAND
2k 4
15k 0.5
30k 0.27
60k 0.13
Table 154 Audio Input DC Blocking Capacitors
Using the figures in Table 154, it follows that a 1 F capacitance for all input connections will give good results in most cases. Tantalum electrolytic capacitors are particularly suitable as they offer high stability in a small package size.
Ceramic equivalents are a cost effective alternative to the superior tantalum packages, but care must be taken to ensure the desired capacitance is maintained at the AVDD1 operating voltage. Also, ceramic capacitors may show microphonic effects, where vibrations and mechanical conditions give rise to electrical signals. This is particularly problematic for microphone input paths where a large signal gain is required.
A single capacitor is required for a line input or single-ended microphone connection. In the case of a differential microphone connection, a DC blocking capacitor is required on both input pins.
HEADPHONE OUTPUT PATH
The headphone output on WM8958 is ground referenced and therefore does not require the large, expensive capacitors necessary for VMID reference solutions. For best audio performance, it is recommended to connect a zobel network to the audio output pins. This network should comprise of a
100nF capacitor and 20ohm resistor in series with each other (see “Analogue Outputs” section).
These components have the effect of dampening high frequency oscillations or instabilities that can arise outside the audio band under certain conditions. Possible sources of these instabilities include the inductive load of a headphone coil or an active load in the form of an external line amplifier.
PP, August 2012, Rev 3.4
361
WM8958
Pre-Production
EARPIECE DRIVER OUTPUT PATH
The earpiece driver on HPOUT2P and HPOUT2N is designed as a 32ohm BTL speaker driver. The outputs are referenced to the internal DC reference VMID, but direct connection to the speaker is possible because of the BTL configuration. There is no requirement for DC blocking capacitors.
LINE OUTPUT PATHS
The WM8958 provides four line outputs (LINEOUT1P, LINEOUT1N, LINEOUT2P and LINEOUT2N).
Each of these outputs is referenced to the internal DC reference, VMID. In any case where a line output is used in a single-ended configuration (i.e. referenced to AGND), a DC blocking capacitor will be required in order to remove the DC bias. In the case where a pair of line outputs is configured as a
BTL differential pair, then the DC blocking capacitor should be omitted.
The choice of capacitor is determined from the filter that is formed between the capacitor and the load impedance – see Figure 85.
Figure 85 Line Output Path Components
LOAD IMPEDANCE MINIMUM CAPACITANCE
FOR 20HZ PASS BAND
10k 0.8
47k 0.17
Table 155 Line Output Frequency Cut-Off
Using the figures in Table 155, it follows that that a 1 F capacitance would be a suitable choice for a line load. Tantalum electrolytic capacitors are again particularly suitable but ceramic equivalents are a cost effective alternative. Care must be taken to ensure the desired capacitance is maintained at the appropriate operating voltage. w PP, August 2012, Rev 3.4
362
Pre-Production
WM8958
POWER SUPPLY DECOUPLING
Electrical coupling exists particularly in digital logic systems where switching in one sub-system causes fluctuations on the power supply. This effect occurs because the inductance of the power supply acts in opposition to the changes in current flow that are caused by the logic switching. The resultant variations (or ‘spikes’) in the power supply voltage can cause malfunctions and unintentional behavior in other components. A decoupling (or ‘bypass’) capacitor can be used as an energy storage component which will provide power to the decoupled circuit for the duration of these power supply variations, protecting it from malfunctions that could otherwise arise.
Coupling also occurs in a lower frequency form when ripple is present on the power supply rail caused by changes in the load current or by limitations of the power supply regulation method. In audio components such as the WM8958, these variations can alter the performance of the signal path, leading to degradation in signal quality. A decoupling (or ‘bypass’) capacitor can be used to filter these effects, by presenting the ripple voltage with a low impedance path that does not affect the circuit to be decoupled.
These coupling effects are addressed by placing a capacitor between the supply rail and the corresponding ground reference. In the case of systems comprising multiple power supply rails, decoupling should be provided on each rail.
The recommended power supply decoupling capacitors for WM8958 are listed below in Table 156.
POWER SUPPLY
LDO1VDD, DBVDD1, DBVDD2, DBVDD3,
AVDD2
0.1
DECOUPLING CAPACITOR
F ceramic (see Note)
SPKVDD1, SPKVDD2 4.7
F ceramic
AVDD1 4.7
F ceramic
DCVDD 2.2
F ceramic
CPVDD 4.7
F ceramic
VMIDC 4.7
F ceramic
VREFC 1.0
F ceramic
Table 156 Power Supply Decoupling Capacitors
Note: 0.1
F is required with 4.7F a guide to the total required power rail capacitance, including that at the regulator output.
All decoupling capacitors should be placed as close as possible to the WM8958 device. The connection between AGND, the AVDD1 decoupling capacitor and the main system ground should be made at a single point as close as possible to the AGND ball of the WM8958.
The VMID capacitor is not, technically, a decoupling capacitor. However, it does serve a similar purpose in filtering noise on the VMID reference. The connection between AGND, the VMID decoupling capacitor and the main system ground should be made at a single point as close as possible to the AGND ball of the WM8958.
Due to the wide tolerance of many types of ceramic capacitors, care must be taken to ensure that the selected components provide the required capacitance across the required temperature and voltage ranges in the intended application. For most applications, the use of ceramic capacitors with capacitor dielectric X5R is recommended. w PP, August 2012, Rev 3.4
363
WM8958
Pre-Production
CHARGE PUMP COMPONENTS
A fly-back capacitor is required between the CPCA and CPCB pins. The required capacitance is
2.2µF at 2V.
A decoupling capacitor is required on CPVOUTP and CPVOUTN; the recommended value is 2.2µF at
2V.
The positioning of the Charge Pump capacitors is important, particularly the fly-back capacitor. These capacitors should be placed as close as possible to the WM8958.
Due to the wide tolerance of many types of ceramic capacitors, care must be taken to ensure that the selected components provide the required capacitance across the required temperature and voltage ranges in the intended application. For most applications, the use of ceramic capacitors with capacitor dielectric X5R is recommended.
MICROPHONE BIAS CIRCUIT
The WM8958 is designed to interface easily with up to four analogue microphones. These may be connected in single-ended or differential configurations, as illustrated in Figure 86. The single-ended method allows greater capability for the connection of multiple audio sources simultaneously, whilst the differential method provides better performance due to its rejection of common-mode noise.
In either configuration, the analogue microphone requires a bias current (electret condenser microphones) or voltage supply (silicon microphones), which can be provided by MICBIAS1 or
MICBIAS2.
A current-limiting resistor is required when using an electret condenser microphone (ECM). The resistance should be chosen according to the minimum operating impedance of the microphone and
MICBIAS voltage so that the maximum bias current of the WM8958 is not exceeded. Wolfson recommends a 2.2k
current limiting resistor as it provides compatibility with a wide range of microphone models.
MIC
AGND
C
IN1LN,
IN2LN,
IN1RN,
IN2RN
IN1LP,
IN2LP,
IN1RP,
IN2RP
2k2
VMID
Line Input
MICBIAS1/2
-
+
PGA
To input mixers
AGND
2k2
MIC
C
IN1LN,
IN2LN,
IN1RN,
IN2RN
C
IN1LP,
IN2LP,
IN1RP,
IN2RP
2k2
VMID
Line Input
MICBIAS1/2
-
+
PGA
To input mixers
Figure 86 Single-Ended and Differential Analogue Microphone Connections
The WM8958 also supports up to four digital microphone inputs. The MICBIAS1 generator is suitable for use as a low noise supply for digital microphones, as shown in Figure 87. w PP, August 2012, Rev 3.4
364
Pre-Production
WM8958
Figure 87 Digital Microphone Connection
The MICBIAS generators can each operate as a voltage regulator or in bypass mode. See “Analogue
Input Signal Path” for details of the MICBIAS generators.
In Regulator mode, the MICBIAS regulators are designed to operate without external decoupling capacitors. It is important that parasitic capacitances on the MICBIAS1 or MICBIAS2 pins do not exceed the specified limit in Regulator mode (see “Electrical Characteristics”).
If the capacitive load on MICBIAS1 or MICBIAS2 exceeds the specified limit (eg. due to a decoupling capacitor or long PCB trace), then the respective generator must be configured in Bypass mode.
The maximum output current is noted in the “Electrical Characteristics”. This limit must be observed on each MICBIAS output, especially if more than one microphone is connected to a single MICBIAS pin. Note that the maximum output current differs between Regulator mode and Bypass mode. The
MICBIAS output voltage can be adjusted using register control in Regulator mode. w PP, August 2012, Rev 3.4
365
WM8958
Pre-Production
EXTERNAL ACCESSORY DETECTION COMPONENTS
The accessory detection circuit measures the impedance of an external load connected to the
MICDET pin.
This function uses the MICBIAS2 output as a reference, as shown in Figure 88. Note that the
WM8958 will automatically enable MICBIAS2 when required in order to perform the detection function.
The WM8958 can detect the presence of a typical microphone and up to 7 push-buttons, using the components shown. When the microphone detection circuit is enabled, then each of the push-buttons shown will cause a different bit within the MICD_LVL register to be set.
The microphone detect function is specifically designed to detect a video accessory (typical 75 ) load if required. A measured external impedance of 75 will cause the MICD_LVL [4] bit to be set.
Figure 88 External Accessory Detect Connection w PP, August 2012, Rev 3.4
366
Pre-Production
WM8958
CLASS D SPEAKER CONNECTIONS
The WM8958 incorporates two Class D/AB speaker drivers. By default, the speaker drivers operate in
Class D mode, which offers high amplifier efficiency at large signal levels. As the Class D output is a pulse width modulated signal, the choice of speakers and tracking of signals is critical for ensuring good performance and reducing EMI in this mode.
The efficiency of the speaker drivers is affected by the series resistance between the WM8958 and the speaker (e.g. PCB track loss and inductor ESR) as shown in Figure 89. This resistance should be as low as possible to maximise efficiency.
Figure 89 Speaker Connection Losses
The Class D output requires external filtering in order to recreate the audio signal. This may be implemented using a 2 nd order LC or 1 st order RC filter, or else may be achieved by using a loudspeaker whose internal inductance provides the required filter response. An LC or RC filter should be used if the loudspeaker characteristics are unknown or unsuitable, or if the length of the loudspeaker connection is likely to lead to EMI problems.
In applications where it is necessary to provide Class D filter components, a 2 nd order LC filter is the recommended solution as it provides more attenuation at higher frequencies and minimises power dissipated in the filter when compared to a first order RC filter (lower ESR). This maximises both rejection of unwanted switching frequencies and overall speaker efficiency. A suitable implementation is illustrated in Figure 90. w
Figure 90 Class D Output Filter Components
PP, August 2012, Rev 3.4
367
WM8958
Pre-Production
A simple equivalent circuit of a loudspeaker consists of a serially connected resistor and inductor, as shown in Figure 91. This circuit provides a low pass filter for the speaker output. If the loudspeaker characteristics are suitable, then the loudspeaker itself can be used in place of the filter components described earlier. This is known as ‘filterless’ operation.
Figure 91 Speaker Equivalent Circuit for Filterless Operation
For filterless Class D operation, it is important to ensure that a speaker with suitable inductance is chosen. For example, if we know the speaker impedance is 8Ω and the desired cut-off frequency is
20kHz, then the optimum speaker inductance may be calculated as:
8 loudspeakers typically have an inductance in the range 20H to 100H, however, it should be noted that a loudspeaker inductance will not be constant across the relevant frequencies for Class D operation (up to and beyond the Class D switching frequency). Care should be taken to ensure that the cut-off frequency of the loudspeaker’s filtering is low enough to suppress the high frequency energy of the Class D switching and, in so doing, to prevent speaker damage. The Class D outputs of the WM8958 operate at much higher frequencies than is recommended for most speakers and it must be ensured that the cut-off frequency is low enough to protect the speaker.
RECOMMENDED EXTERNAL COMPONENTS DIAGRAM
Figure 92 provides a summary of recommended external components for WM8958. Note that this diagram does not include any components that are specific to the end application e.g. it does not include filtering on the speaker outputs (assume filterless class D operation), RF decoupling, or RF filtering for pins which connect to the external world i.e. headphone or speaker outputs. w PP, August 2012, Rev 3.4
368
Pre-Production
WM8958
Figure 92 Recommended External Components Diagram w PP, August 2012, Rev 3.4
369
WM8958
Pre-Production
DIGITAL AUDIO INTERFACE CLOCKING CONFIGURATIONS
The WM8958 provides 3 digital audio interfaces and supports many different clocking configurations.
The asynchronous sample rate converter enables more than one digital audio interface to be supported simultaneously, even when there is no synchronisation between these interfaces. In a typical application, this enables audio mixing between a multimedia applications processor and a baseband voice call processor, for example.
The AIF1 and AIF2 audio interfaces can be configured in Master or Slave modes, and can also support defined combinations of mixed sample rates. In all applications, it is important that the system clocking configuration is correctly designed. Incorrect clock configurations will lead to audible clicks arising from dropped or repeated audio samples; this is caused by the inherent tolerances of multiple asynchronous system clocks.
To ensure reliable clocking of the audio interface functions, it is a requirement that, for each audio interface, the external interface clocks (eg. BCLK, LRCLK) are derived from the same clock source as the respective AIF clock (AIFnCLK).
In AIF Master mode, the external BCLK and LRCLK signals are generated by the WM8958 and synchronisation of these signals with AIFnCLK is guaranteed. In this case, clocking of the AIF is derived from the MCLK1 or MCLK2 inputs, either directly or via one of the Frequency Locked Loop
(FLL) circuits.
In AIF Slave mode, the external BCLK and LRCLK signals are generated by another device, as inputs to the WM8958. In this case, it must be ensured that the respective AIF clock is generated from a source that is synchronised to the external BCLK and LRCLK inputs. In a typical Slave mode application, the BCLK input is selected as the clock reference, using the FLL to perform frequency shifting. It is also possible to use the MCLK1 or MCLK2 inputs, but only if the selected clock is synchronised externally to the BCLK and LRCLK inputs.
The valid AIF clocking configurations are listed in Table 157 for AIF Master and AIF Slave modes.
AUDIO INTERFACE MODE
AIF Master Mode
AIF Slave Mode
CLOCKING CONFIGURATION
AIFnCLK_SRC selects FLL1 or FLL2 as AIFnCLK source;
FLLn_REFCLK_SRC selects MCLK1 or MCLK2 as FLLn source.
AIFnCLK_SRC selects MCLK1 or MCLK2 as AIFnCLK source.
AIFnCLK_SRC selects FLL1 or FLL2 as AIFnCLK source;
FLLn_REFCLK_SRC selects BCLKn as FLLn source.
AIFnCLK_SRC selects MCLK1 or MCLK2 as AIFnCLK source, provided MCLK is externally synchronised to the BCLKn input.
AIFnCLK_SRC selects FLL1 or FLL2 as AIFnCLK source;
FLLn_REFCLK_SRC selects MCLK1 or MCLK2 as FLLn source, provided MCLK is externally synchronised to the BCLKn input.
Table 157 Audio Interface Clocking Confgurations
In each case, the AIFnCLK frequency must be a valid ratio to the LRCLKn frequency; the supported clocking ratios are defined by the AIFnCLK_RATE register.
The valid AIF clocking configurations are illustrated in Figure 93 to Figure 97 below. Note that, where
MCLK1 is illustrated as the clock source, it is equally possible to select MCLK2 as the clock source.
Similarly, in cases where FLL1 is illustrated, it is equally possible to select the FLL2. w PP, August 2012, Rev 3.4
370
Pre-Production
WM8958
Figure 93 AIF Master Mode, using MCLK as reference
WM8958
FLL1_REFCLK_SRC
FLL1
AIFnCLK
AIFn
(Master Mode)
AIFnCLK_SRC
BCLKn
LRCLKn
DACDATn
ADCDATn
Processor
Oscillator
Figure 94 AIF Master Mode, using MCLK and FLL as reference
WM8958
FLL1_REFCLK_SRC
FLL1
AIFnCLK_SRC
AIFnCLK
AIFn
(Slave Mode)
Figure 95 AIF Slave Mode, using BCLK and FLL as reference
BCLKn
LRCLKn
DACDATn
ADCDATn
Processor w PP, August 2012, Rev 3.4
371
WM8958
Pre-Production
Figure 96 AIF Slave Mode, using MCLK as reference
Figure 97 AIF Slave Mode, using MCLK and FLL as reference w PP, August 2012, Rev 3.4
372
Pre-Production
WM8958
PCB LAYOUT CONSIDERATIONS
Poor PCB layout will degrade the performance and be a contributory factor in EMI, ground bounce and resistive voltage losses. All external components should be placed as close to the WM8958 device as possible, with current loop areas kept as small as possible. Specific factors relating to
Class D loudspeaker connection are detailed below.
CLASS D LOUDSPEAKER CONNECTION
Long, exposed PCB tracks or connection wires will emit EMI. The distance between the WM8958 and the loudspeaker should therefore be kept as short as possible. Where speakers are connected to the
PCB via a cable form, it is recommended that a shielded twisted pair cable is used. The shield should be connected to the main system, with care taken to ensure ground loops are avoided.
Further reduction in EMI can be achieved using PCB ground (or VDD) planes and also by using passive LC components to filter the Class D switching waveform. When passive filtering is used, low
ESR components should be chosen in order to minimise the series resistance between the WM8958 and the speaker, maximising the power efficiency.
LC passive filtering will usually be effective at reducing EMI at frequencies up to around 30MHz. To reduce emissions at higher frequencies, ferrite beads can also be used. These should be positioned as close to the device as possible.
These techniques for EMI reduction are illustrated in Figure 98.
SPKP
SPKN
EMI
Long, exposed tracks emit EMI
SPKP
SPKN Short connection wires will reduce EMI emission
SPKP
SPKN
Shielding using PCB ground (or VDD) planes will reduce EMI emission
SPKP
LOW ESR
SPKN
LOW ESR
LC filtering will reduce EMI emission up to around 30MHz
SPKP
SPKN w
Figure 98 EMI Reduction Techniques
Ferrite beads will reduce EMI emission at frequencies above 30MHz.
PP, August 2012, Rev 3.4
373
WM8958
PACKAGE DIMENSIONS
Pre-Production
B: 72 BALL W-CSP PACKAGE 4.516 X 4.258 X 0.698 mm BODY, 0.50 mm BALL PITCH
DETAIL 1 g A2
2
A
9 8 7 6 5 4 3 2 1
A
B
C
4
A1
CORNER
D
E
F
G e
5
E1
H
DETAIL 2 e ddd
M
Z A B
D1
BOTTOM VIEW
2 X
2 X aaa B aaa A f1
6
D
TOP VIEW
SOLDER BALL
DM119.A
A
E
B bbb Z h f2
1
Z ccc Z
A1
DETAIL 2
Symbols
E
E1 e f1
A
A1
A2
D
D1 f2
MIN
0.658
0.206
0.418
4.491
4.233
0.246
0.367
Dimensions (mm)
NOM
0.698
0.242
0.434
4.516
4.00 BSC
4.258
3.50 BSC
0.50 BSC
MAX
0.738
0.278
0.450
4.541
4.283
NOTE
5
8
9 g h aaa
0.264
0.022
0.314
0.025
0.060
0.364
bbb ccc ddd
0.030
0.015
NOTES:
1. PRIMARY DATUM -Z- AND SEATING PLANE ARE DEFINED BY THE SPHERICAL CROWNS OF THE SOLDER BALLS.
2. THIS DIMENSION INCLUDES STAND-OFF HEIGHT ‘A1’ AND BACKSIDE COATING.
3. A1 CORNER IS IDENTIFIED BY INK/LASER MARK ON TOP PACKAGE.
4. BILATERAL TOLERANCE ZONE IS APPLIED TO EACH SIDE OF THE PACKAGE BODY.
5. ‘e’ REPRESENTS THE BASIC SOLDER BALL GRID PITCH.
6. THIS DRAWING IS SUBJECT TO CHANGE WITHOUT NOTICE.
7. FOLLOWS JEDEC DESIGN GUIDE MO-211-C.
8. f1 = NOMINAL DISTANCE OF BALL CENTRE TO DIE EDGE X AXIS (AS PER POD) – APPLICABLE TO ALL CORNERS OF DIE.
9. f2 = NOMINAL DISTANCE OF DIE CENTRE TO DIE EDGE IN Y AXIS (AS PER POD) – APPLICABLE TO ALL CORNERS OF DIE.
w PP, August 2012, Rev 3.4
374
Pre-Production
WM8958
IMPORTANT NOTICE
Wolfson Microelectronics plc (“Wolfson”) products and services are sold subject to Wolfson’s terms and conditions of sale, delivery and payment supplied at the time of order acknowledgement.
Wolfson warrants performance of its products to the specifications in effect at the date of shipment. Wolfson reserves the right to make changes to its products and specifications or to discontinue any product or service without notice. Customers should therefore obtain the latest version of relevant information from Wolfson to verify that the information is current.
Testing and other quality control techniques are utilised to the extent Wolfson deems necessary to support its warranty.
Specific testing of all parameters of each device is not necessarily performed unless required by law or regulation.
In order to minimise risks associated with customer applications, the customer must use adequate design and operating safeguards to minimise inherent or procedural hazards. Wolfson is not liable for applications assistance or customer product design. The customer is solely responsible for its selection and use of Wolfson products. Wolfson is not liable for such selection or use nor for use of any circuitry other than circuitry entirely embodied in a Wolfson product.
Wolfson’s products are not intended for use in life support systems, appliances, nuclear systems or systems where malfunction can reasonably be expected to result in personal injury, death or severe property or environmental damage.
Any use of products by the customer for such purposes is at the customer’s own risk.
Wolfson does not grant any licence (express or implied) under any patent right, copyright, mask work right or other intellectual property right of Wolfson covering or relating to any combination, machine, or process in which its products or services might be or are used. Any provision or publication of any third party’s products or services does not constitute
Wolfson’s approval, licence, warranty or endorsement thereof. Any third party trade marks contained in this document belong to the respective third party owner.
Reproduction of information from Wolfson datasheets is permissible only if reproduction is without alteration and is accompanied by all associated copyright, proprietary and other notices (including this notice) and conditions. Wolfson is not liable for any unauthorised alteration of such information or for any reliance placed thereon.
Any representations made, warranties given, and/or liabilities accepted by any person which differ from those contained in this datasheet or in Wolfson’s standard terms and conditions of sale, delivery and payment are made, given and/or accepted at that person’s own risk. Wolfson is not liable for any such representations, warranties or liabilities or for any reliance placed thereon by any person.
ADDRESS:
Wolfson Microelectronics plc
26 Westfield Road
Edinburgh
EH11 2QB
United Kingdom
Tel :: +44 (0)131 272 7000
Fax :: +44 (0)131 272 7001
Email :: [email protected]
w PP, August 2012, Rev 3.4
375
WM8958
Pre-Production
REVISION HISTORY
DATE REV
16/11/10 2.2
4/1/11
24/1/11
2.2
2.2
28/1/11 2.2
24/03/11 3.0
DESCRIPTION OF CHANGES
DRC Signal Detect registers DRC_SIG_DET_RMS, DRC_SIG_DET_PK and
DRC_SIG_DET_MODE updated.
Additional details provided on pull-up / pull-down functions.
Added notes that DRC and MBC must not be enabled simultaneously on the same playback path.
Added notes that the Output Path HPF should be enabled when DRC is used on a record (ADC) path.
Power domains and Ground references listed for each input/output.
MICD_BIAS_STARTTIME and MICD_RATE descriptions updated, and associated text / recommended settings.
Noted that DRC Anti-Clip and Quick Release features should not be used at the same time.
VMID soft-start descriptions updated, including requirement to reset soft-start circuit before re-enabling VMID.
Speaker driver performance graphs added.
SPKAB_REF_SEL added to ‘Registers By Address’ section.
Added note that LDOs are not suitable for external loads.
Noted RF suppression on analogue inputs.
Noise Gate function defined for AIF1 and AIF2 input paths.
DAC Volume registers updated to support values up to +12dB.
EQ Band 1 now configurable as Shelf or Peak filter.
DSP2CLK_SRC register deleted.
MBC Control sequences updated.
EFS modes described for FLL1 and FLL2.
Decoupling capacitor removed from DMIC connection drawing.
Additional register writes added to the MBC enable sequence.
Pin description list re-sorted by Name, in order to draw attention to any multiple pins with a common name.
Updates noting that Ultrasonic (4FS) mode uses ADCLRCLK (not LRCLK).
GPIO1/GPIO6 must be configured for AIF1/AIF2 respectively.
Input Path drawing updated, showing VMID as PGA reference.
Accessory detection / impedance sensing added to Introduction section and on front page.
Updated electrical characteristics to reflect 4 ohm mono mode THD performance.
Restriction on MICBIAS capacitance clarified - 50pF limit is only applicable in Normal
(regulator) mode.
Applications Information (MICBIAS) enhanced to incorporate Digital Microphone connections.
Interrupts section updated to improve clarity.
Corrections to the FLL Example settings
LDO2 output voltage updated (1.1V to 1.3V)
Updated speaker inductive load conditions in electrical characteristics to 22uH.
Updated LDO2 output voltage in electrical characteristics.
Added max/min limits to electrical characteristics.
CHANGED BY
PH
PH
PH
KOL
KOL w PP, August 2012, Rev 3.4
376
Pre-Production
DATE REV
14/02/11 3.0
04/04/11 3.1
18/05/11 3.3
24/11/11 3.3
03/01/12 3.3
25/05/12 3.4
WM8958
DESCRIPTION OF CHANGES
Updated speaker inductive load in electrical characteristics to 22uH.
Updated LDO2 output voltage in electrical characteristics.
Notes added requiring VMID_BUF_ENA is enabled for direct signal paths from input pins to Input Mixers, Output Mixers or Speaker Mixers. Descriptions of affected register bits updated.
Reel order quantity updated
MICBIAS modes clarified as “Regulator” mode and “Bypass” mode.
Ultrasonic (4FS) mode deleted on AIF2.
External Accessory Detect description & characteristics updated; Recommended
External Components added in Applications Information section.
Block diagram updated (input digital mixing paths and MICBIAS references)
Updated ADC Path characteristics - input is -1dBV, not -1dBFS.
Clarification of DAC_OSR128 modes in DAC playback path Electrical Characteristics.
Input PGA Mute behaviour description updated.
Noted that HPF is required when using DRC Signal Activity Detect.
Updates to FLL Input Frequency range.
Minimum headphone load resistance updated.
Clarifications and formatting updates to Electrical Characteristics and Recommended
Operating Conditions.
Noted phase inversion in ‘Direct Voice’ paths.
Clarification to the usage of the INPUTS_CLAMP register.
PSRR specifications added for LDO1 and LDO2.
Drop-out voltage specification added for LDO1.
RMS Limiter function added within the MBC description.
TSHUT_ENA default corrected in Power Management section (default is 1).
Absolute Maximum Ratings updated: (AVDD1 domain) added to Voltage range analogue inputs.
Maximum MICBIASn load capacitance noted in Electrical Characteristics.
Specifications added for LINEOUTFB and HPOUT1FB ground noise rejection.
System clocking updated - DBCLK is derived independently of TOCLK_ENA.
Additional details in Absolute Maximum Ratings.
Clarification of Line Output discharge functions and associated Electrical
Characteristics.
Package diagram changed to DM119.A
CHANGED BY
PH
PH
PH
PH
JMacD w PP, August 2012, Rev 3.4
377
advertisement
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Related manuals
advertisement
Table of contents
- 1 WM8958
- 1 Multi-Channel Audio Hub CODEC for Smartphones
- 1 DESCRIPTION
- 1 FEATURES
- 1 APPLICATIONS
- 2 TABLE OF CONTENTS
- 7 BLOCK DIAGRAM
- 8 PIN CONFIGURATION
- 8 ORDERING INFORMATION
- 9 PIN DESCRIPTION
- 12 ABSOLUTE MAXIMUM RATINGS
- 13 RECOMMENDED OPERATING CONDITIONS
- 14 THERMAL PERFORMANCE
- 15 ELECTRICAL CHARACTERISTICS
- 15 INPUT SIGNAL LEVEL
- 16 INPUT PIN RESISTANCE
- 18 PROGRAMMABLE GAINS
- 19 OUTPUT DRIVER CHARACTERISTICS
- 20 ADC INPUT PATH PERFORMANCE
- 21 DAC OUTPUT PATH PERFORMANCE
- 24 BYPASS PATH PERFORMANCE
- 26 MULTI-PATH CROSSTALK
- 28 DIGITAL INPUT / OUTPUT
- 28 DIGITAL FILTER CHARACTERISTICS
- 29 MICROPHONE BIAS CHARACTERISTICS
- 30 MISCELLANEOUS CHARACTERISTICS
- 31 TERMINOLOGY
- 32 TYPICAL PERFORMANCE
- 32 TYPICAL POWER CONSUMPTION
- 33 TYPICAL SIGNAL LATENCY
- 34 SPEAKER DRIVER PERFORMANCE
- 35 SIGNAL TIMING REQUIREMENTS
- 35 SYSTEM CLOCKS & FREQUENCY LOCKED LOOP (FLL)
- 36 AUDIO INTERFACE TIMING
- 40 CONTROL INTERFACE TIMING
- 41 DEVICE DESCRIPTION
- 41 INTRODUCTION
- 43 ANALOGUE INPUT SIGNAL PATH
- 56 DIGITAL MICROPHONE INTERFACE
- 59 ANALOGUE TO DIGITAL CONVERTER (ADC)
- 61 DIGITAL CORE ARCHITECTURE
- 63 DIGITAL MIXING
- 70 MULTIBAND COMPRESSOR (MBC)
- 75 DYNAMIC RANGE CONTROL (DRC)
- 89 RETUNE MOBILE PARAMETRIC EQUALIZER (EQ)
- 94 3D STEREO EXPANSION
- 95 DIGITAL VOLUME AND FILTER CONTROL
- 109 DIGITAL TO ANALOGUE CONVERTER (DAC)
- 116 ANALOGUE OUTPUT SIGNAL PATH
- 138 CHARGE PUMP
- 140 DC SERVO
- 144 ANALOGUE OUTPUTS
- 151 EXTERNAL ACCESSORY DETECTION
- 156 GENERAL PURPOSE INPUT/OUTPUT
- 165 INTERRUPTS
- 170 DIGITAL AUDIO INTERFACE
- 178 DIGITAL AUDIO INTERFACE CONTROL
- 200 CLOCKING AND SAMPLE RATES
- 221 SAMPLE RATE CONVERSION
- 224 CONTROL INTERFACE
- 227 CONTROL WRITE SEQUENCER
- 237 POP SUPPRESSION CONTROL
- 239 LDO REGULATORS
- 241 REFERENCE VOLTAGES AND MASTER BIAS
- 243 POWER MANAGEMENT
- 248 THERMAL SHUTDOWN
- 249 POWER ON RESET
- 251 QUICK START-UP AND SHUTDOWN
- 252 SOFTWARE RESET AND DEVICE ID
- 253 REGISTER MAP
- 360 APPLICATIONS INFORMATION
- 360 RECOMMENDED EXTERNAL COMPONENTS
- 370 DIGITAL AUDIO INTERFACE CLOCKING CONFIGURATIONS
- 373 PCB LAYOUT CONSIDERATIONS
- 374 PACKAGE DIMENSIONS
- 375 IMPORTANT NOTICE
- 375 ADDRESS:
- 376 REVISION HISTORY