- No category
advertisement
EV10AS150B
High Linearity ADC 10-bit 2.6 Gsps with 1:4 DMUX
5 GHz Full Power Bandwidth
Datasheet
Features
•
ADC 10-bit Resolution
•
Up to 2.6 Gsps Sampling Rate
•
Selectable 1:4 or 1:2 Demultiplexed Digital LVDS Outputs
•
True Single Core Architecture (No Calibration Required)
•
External Interleaving Possible Via 3-Wire Serial Interface
– Gain Adjust
– Offset Adjust
– Sampling Delay Adjust
•
Full Scale Analog Input Voltage Span 500 mVpp
•
100
Ω Differential Analog Input and Clock Input
•
Differential Digital Outputs, LVDS Logic Compatibility
•
Low Latency Pipeline Delay
•
Test Mode for Output Data Registering (BIST)
•
Power Management (Nap, Sleep Mode)
•
EBGA317 (Enhanced Ball Grid Array) Package
Performance
•
Single Tone Performance in 1 st
Nyquist (–1 dBFS)
– ENOB = 8.0 bit, SFDR = –57 dBFS at 2.6 Gsps, Fin = 495 MHz
– ENOB = 7.9 bit, SFDR = –57 dBFS at 2.6 Gsps, Fin = 1295 MHz
•
Single Tone Performance in 2 nd
Nyquist (–3 dBFS):
– ENOB = 7.9 bit, SFDR = –59 dBFS at 2.6 Gsps, Fin = 2595 MHz
•
5 GHz Full Power Input Bandwidth (–3 dB)
•
±0.5 dB Band Flatness from 10 MHz to 2.0 GHz
•
Input VSWR = 1.25:1 from DC to 3 GHz
•
Bit Error Rate: 10
–12
at 2.6 Gsps
Screening
•
Temperature Range
– Commercial “C” Grade: Tamb > 0
° C ; T
J
< 90
° C
– Industrial “V” Grade: Tamb > –40
° C ; T
J
< 110
° C
Applications
•
Direct Broadband RF Down Conversion
•
Wide Band Communications Receiver
•
High Speed Instrumentation
•
High Speed Data Acquisition Systems
e2v semiconductors SAS 2012
Visit our website: www.e2v.com
for the latest version of the datasheet
1076D–BDC–10/12
EV10AS150B
1.
Block Diagram
The EV10AS150B combines a 10-bit 2.6 Gsps fully bipolar analog-to-digital converter chip, driving a fully bipolar DMUX chip with selectable Demultiplexing ratio (1:2) or (1:4). The 5 GHz full power input bandwidth of the ADC allows the direct digitization of up to 1 GHz broadband signals in the high IF region, in either L_Band or S_Band.
The 1:4 demultiplexed digital outputs are LVDS logic compatible, which allows easy interface with standard FPGAs or DSPs. The EV10AS150B operates at up to 2.6 Gsps in DMUX 1:4 and up to 2.0 Gsps in
1:2 DMUX ratio (The speed limitation with 1:2 DMUX ratio is mainly dictated by external data flow exchange capability at 2 × 1 Gsps with available FPGAs).
The EV10AS150B ADC+DMUX combo device is packaged in a 25 × 35 mm Enhanced Ball Grid Array
EBGA317. This Package is based on multiple layers which allows the design of low impedance continuous ground and power supplies planes, and the design of 50
Ω controlled impedance lines (100Ω differential impedance). This package has the same Thermal Coefficient of Expansion (TCE) as FR4 application boards, thus featuring excellent long term reliability when submitted to repeated thermal cycles.
Figure 1-1.
Functional Block Diagram
DRR (ADC Reset)
CLK
CLKN
3 WSI RESET
3 WSI
SLDN
SDATA
SCLK
3-wire Serial
Interface
(3WSI)
8-bit
DAC
Sampling delay
Gain
8-bit
DAC
Offset
8-bit
DAC
VIN
VINN
100
Ω
T/H
100
Ω
Timing
Circuitry
& SDA
Tunable
Delay line
CLKDACTRL
ASYNCRST
(Dmux Reset)
Tunable
Delay line
Demultiplexer
1:2 or 1:4
2
2
20
2
20
20
2
20
2
Port A
AOR/AORN
DRA/DRAN
Port B
BOR/BORN
DRB/DRBN
Port C
COR/CORN
DRC/DRCN
Port D
DOR/DORN
DRD/DRDN
DR/DRN
BIST (Pattern Generator)
SLEEP (DMUX)
STAGG (Latency)
RS (Ratio Sel)
DRTYPE (Dual Data Rate)
2
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.
Specifications
This section describes the device specifications in terms of:
• Absolute max ratings
• Recommended conditions of use
• Electrical operating characteristics
• Timings
2.1
Absolute Maximum Ratings
Absolute maximum ratings are limiting values (referenced to GND = 0V), to be applied individually, while other parameters are within specified operating conditions. Long exposure to maximum rating may affect device reliability.
Maximum ratings on I/Os are defined with device powered ON.
All integrated circuits have to be handled with appropriate care to avoid damages due to ESD. Damage caused by inappropriate handling or storage could range from performance degradation to complete failure.
Table 2-1.
Absolute Maximum Ratings
Parameter
Analog 4.9V Power Supply voltage
Analog 3.25V Power Supply voltage
Digital 3.3V Power Supply Voltage
Output 2.5V Power Supply voltage
Minimum Analog input peak voltage
(with differential input)
Maximum Analog input peak voltage
(with differential input)
Maximum difference between V
IN
and V
(with differential input)
Symbol
V
CCA5
V
CCA3
V
CCD
V
PLUSD
V
IN
or V
INN
V
IN
or V
INN
|V
IN
– V
INN
|
Minimum Analog input peak voltage
(with single ended input)
Maximum Analog input peak voltage
(with single ended input)
V
IN
with V
INN
= 50
Ω to GND or
V
INN
with V
IN
= 50
Ω to GND
V
IN
with V
INN
= 50
Ω to GND or
V
INN
with V
IN
= 50
Ω to GND
Maximum amplitude on V
IN
or V
INN
(with single ended input)
Minimum Clock input peak voltage
(with differential clock)
Maximum Clock input peak voltage
(with differential clock)
Maximum difference between V
CLK
and V
CLKN
(with differential clock)
3WSI input voltage
|V
IN
| or |V
INN
|
V
CLK or V
CLKN
V
CLK
or V
CLKN
|V
CLK
– V
CLKN
|
SDATA, SLDN, SCLK, RESET
Value
GND to 6.0
GND to 3.6
GND to 3.6
GND to 3.0
2.0
4.0
2.0
(4 Vpp = +13 dBm in 100
Ω)
2.0
4.0
(2 Vpp = +10 dBm in 50
Ω)
1.5
4.0
1.5
(3 Vpp)
–0.3 to V
CCA3
+ 0.3
Unit
V
V
V
V
V
V
V
V
V
V
V
V
V
V
1076D–BDC–10/12
3
e2v semiconductors SAS 2012
EV10AS150B
Table 2-1.
Absolute Maximum Ratings (Continued)
Parameter
ADC Reset Voltage
DMUX function input voltage
DMUX Asynchronous Reset
DMUX Control Voltage
Maximum input voltage on DIODE
Maximum input current on DIODE
Max Junction Temperature
Storage temperature
ESD protection (HBM)
Symbol
DRR
RS, DRTYPE, SLEEP, STAGG, BIST
ASYNCRST
CLKDACTRL
DIODE ADC
DIODE ADC
T
J
Tstg
Value
–0.3 to V
CCA3
+ 0.3
–0.3 to V
CCD
+ 0.3
–0.3 to V
CCD
+ 0.3
–0.3 to V
CCD
+ 0.3
700
1
135
–55 to 150
≥ 500 on ADC inputs
500 on DMUX outputs
Notes:
.
.
2.1.1
Analog Input Max Ratings in differential configuration
Internal DC Common mode bias for differential analog inputs is: +3V (90.9% of V
CCA3
).
Input impedance on V
IN
and V
INN
is: 55
Ω // 550Ω = 50Ω.
Max rating is ±1V = 2 Vpp on each single ended input, corresponding to 4 Vpp in differential.
Unit
V
V
V
V mV mA
° C
° C
V
• Vinput max = 3V + 1V = 4V
• Vinput min = 3V – 1V = 2V
• With V
IN
= +4V and V
INN
= +2V => V
IN
– V
INN
= +2V
• With V
INN
= +4V and V
IN
= +2V => V
IN
– V
INN
= –2V
Figure 2-1.
Analog Input Max Ratings in Differential Configuration
Differential analog input signals:
Max rating differential inputs = ± 2 Vpeak = 4 Vpeak to peak
Maximum voltage on VIN or on VINN : +4 Volt:
=> ESD protections connected between VIN and VINN to +3.25V
4V max
VIN
VINN
DC blocking
50
Ω Line
VIN
DC blocking
50
Ω Line
2V min
VINN
Max (VIN - VINN) = 4 Volt peak-peak
EV10AS150B
V
CCA3
= +3.25V
+0.7V max
+ 10 mA max
55
Ω
ESD
V
CCA5
= +4.9V
V
CCA3
= +3.25V
+3V DC common mode
+0.7V max
+ 10 mA max
55
Ω
± 2Vpeak max diff
ESD
VINN
ESD
550
Ω
550
Ω
ESD
GND
4
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.1.2
Analog Input Max Ratings in Single Ended Configuration
Internal DC common mode bias for differential analog inputs is +3V (90.9% of V
CCA3
).
Input impedance on V
IN
(and V
INN
) is: 55
Ω // 550Ω = 50Ω
Max rating is ±1V on one single ended signal, corresponding to 2 Vpp on 50
Ω
Limiting parameter for maximum rating is not V
IN
– V
INN
differential input voltage but V
IN
peak voltage value which shall not exceed +4V to avoid turning on the ESD protection tied to V
CCA3
= +3.25V.
Vinput max = 3V (CM) + 1V (ESD) = 4V (If +4 Vp is applied, the ESD protections becomes forward biased: (+4V – 3.25V = +0.75V).
Vinput min = 3V – 1V = 2V
The ESD diode can sustain up to 10 mA forward biasing without damage, but incoming signal will be clipped at +4V = +3V (internal DC common mode) + 1V.
Figure 2-2.
Analog Input Max Ratings in Single Ended Configuration
Single-ended analog input signal:
Max rating differential inputs = ±1Vpeak = 2 Vpeak to peak
Maximum voltage on VIN: +4 Volt
Minimum voltage on VIN: +2 Volt
=> ESD protections connected between VIN and VINN to +3.25V:
+0.7 Volt max, + 10 mA max
VIN
4V max
VIN
DC blocking
50
Ω Line
EV10AS150B
V
CCA3
= +3.25V
+0.7V max
+ 10 mA max
55
Ω
ESD
V
CCA5
= +4.9V
+3V DC
± 2Vpeak max diff
V common mode
CCA3
55
Ω
= +3.25V
+0.7V max
+ 10 mA max
ESD
VINN
2V min
ESD
550
Ω 550
Ω
ESD
50
Ω
DC blocking
VINN
50
Ω Line
GND
GND = 0V
1076D–BDC–10/12
5
e2v semiconductors SAS 2012
EV10AS150B
2.2
Recommended Conditions of Use
Table 2-2.
Recommended Conditions of Use
Symbol Parameter
Power supplies
Analog 3.25V Power Supply voltage
Analog 4.9V Power Supply voltage
Digital 3.3V Power Supply voltage
Output 2.5V Power Supply voltage
Analog Input
Recommended Configuration
Clock Input
Recommended Configuration
V
CCA3
V
CCA5
V
CCD
V
PLUSD
Clock input power level
External clock Duty cycle
Control functions input level
ADC 3WSI Inputs
ADC Reset
P
CLK
P
CLKN
DCYC
DMUX Control Inputs
Operating Temperature
SDATA, SLDN, SCLK, RESET
DRR
SLEEP, STAGG, ASYNCRST,
BIST, RS, DRTYPE,
CLKDACTRL
T
J
Comments
No specific power supply sequencing required during power
ON / OFF
100
Ω differential clock
Recommended
3.25
4.75V to 4.9V
3.3
2.5
Differential Analog input
Differential Clock input
+1 dBm / 50
Ω
(1Vpp in 100
Ω)
50
0 to V
CCA3
0 to V
CCA3
0 to V
CCD
Best performances at highest temperatures
Unit
dBm
%
V
V
V
V
V
V
V
Table 2-3.
3WSI Register
Recommended Configuration for Optimum Dynamic Performances
Recommended value
State register SDA OFF
comment
For optimum SNR in 2 nd
Nyquist zone if not used in interleaving mode
Clock duty Cycle adjust register D4….D0
35/65%
Clock adjust register D4 … D0 +30 ps
Gain adjust D7...D0
Minimum gain
For optimum SFDR and THD
For optimum SFDR and THD
For optimum SNR and SFDR
SNR can be increased by around 1 dB if V
CCA5
is decreased from 4.9V to 4.75V.
SFDR in the 2 nd
Nyquist is optimum near V
CCA
= 4.9V. Decreasing V
CCA the 2 nd
Nyquist by around 4 dB.
at 4.75V will degrade SFDR in
6
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.3
Electrical Characteristics for Supplies, Inputs and Outputs
Unless otherwise specified:
Values are given over temperature and power supplies range.
Table 2-4.
Electrical Characteristics for Supplies, Inputs and Outputs
Test Level Symbol Parameter
Power requirements
Power Supply voltages
Analog 4.9V
Analog 3.25V
Digital 3.3V
Output 2.5V
Power Supply current in 1:2 DMUX
Analog V
CCA5
= 4.9V
Analog V
CCA3
= 3.25V
Digital V
CCD
= 3.3V
Output V
PLUSD
= 2.5V
Power Supply current in 1:4 DMUX
Analog V
CCA5
= 4.9V
Analog V
CCA3
= 3.25V
Digital V
CCD
= 3.3V
Output V
PLUSD
= 2.5V
Power Supply current in NAP and SLEEP mode
Analog V
CCA5
= 4.9V
Analog V
CCA3
= 3.25V
Digital V
CCD
= 3.3V
Output V
PLUSD
= 2.5V
Power dissipation
- 1:2 DMUX
- 1:4 DMUX
- NAP & SLEEP mode (1:4 or 1:2)
Analog inputs
Common mode compatibility for analog inputs
Analog inputs common voltage (internal)
Full-Scale input Voltage Range
Differential mode
Full-Scale input Voltage Range
Single ended mode with other input connected to ground through 50
Ω resistors
Analog Input power Level
(in 100
Ω differential termination)
Analog Input power Level
(in 50
Ω single ended termination)
Analog input capacitance (die)
Input leakage current
1
1
1
1
1
4
1
1
1
4
4
4
V
CCA5
V
CCA3
V
CCD
V
PLUSD
I
VCCA5
I
VCCA3
I
VCCD
I
VPLUSD
I
VCCA5
I
VCCA3
I
VCCD
I
VPLUSD
I
VCCA5
I
VCCA3
I
VCCD
I
VPLUSD
P
D
V
CM
V
IN
V
INN
V
IN
or V
INN
P
IN, INN
P
IN or P
INN
C
IN
I
IN
Min
4.75
3.15
3.15
2.4
–250
Typ
4.9
3.25
3.3
2.5
150
850
350
470
150
850
400
490
110
560
130
400
5.8
6.0
3.8
–5
0.5
–2
0.3
10
Max
5.05
3.35
3.45
2.6
190
990
430
620
190
990
470
650
150
800
150
550
7.4
7.6
5.4
AC or DC (AC recommended)
3.0
–125
–125
125
125
250
Unit
W
W
W mA mA mA mA mA mA mA mA mA mA mA mA
V
V
V
V
V mV mV dBm
Vpp dBm pF
µA
1076D–BDC–10/12
7
e2v semiconductors SAS 2012
EV10AS150B
Table 2-4.
Parameter
Input resistance
- Single-ended
- Differential
Clock inputs
Electrical Characteristics for Supplies, Inputs and Outputs (Continued)
Test Level Symbol Min
4 R
IN
R
IN, INN
47
94
Logic common mode compatibility for clock inputs
4
4
Typ
50
100
Max
53
106
Low phase noise sinewave at 2.6 GHz (> 155 dBc/Hz) or low jitter LVDS/LVPECL (<100 fs rms)
120
Clock inputs internal DC common mode voltage 4 V
CM
3.0
90.9%
Clock input voltage on each single ended input (2.6 GHz sinewave)
Clock input voltage into 100
Ω differential clock input
(2.6 GHz sinewave)
Clock input power level (2.6 GHz low phase noise sinewave input) in 50
Ω
Sinewave or Square wave
Clock signal (minimum) slew-rate
Square Wave differential Clock input voltage on 100
Ω input
(LVDS & LVPECL compatible)
Clock input capacitance (die)
Clock input Resistance
- Single-ended
- Differential
Digital Data Outputs
Logic compatibility
50
Ω transmission lines, 100Ω (2 × 50Ω) differential termination
- Logic low
- Logic high
- Differential output
- Common mode
Control Function Inputs
ASYNCRST (DEMUX Reset) input voltages and currents
- Logic Low
- Logic High (reset mode)
4
4
4
4
4
4
4
1
1
V
CLK or V
CLKN
|V
CLK
–V
CLKN
|
P
CLK, CLKN
SR
CLK, CLKN
VSQ
CLK-CLKN
C
CLK
R
CLK
R
CLK, CLKN
VOL
VOH
VODIFF
VOCM
V
IL
I
IL
V
IH
I
IH
±0.158
0.632
–3
5
0.25
47
94
1.25
250
1.125
0
–300
1.8
±0.25
1
+1
8
0.8
0.3
50
100
LVDS
1.075
1.375
350
1.25
±0.5
2
+7
2
53
106
1.25
450
1.375
1.4
V
CCD
50
Logic low
Logic high
Threshold
4
V
IL
V
IH
V
TH
0
1.5
1.4
1.1
V
CCA3
Unit
fs rms
V
% of
V
CCA3
V
Vpp dBm
GV/s
Vpp pF
V
µA
V
µA
V
V
V
Ω
Ω
Ω
Ω
V
V mV
V
8
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Table 2-4.
Electrical Characteristics for Supplies, Inputs and Outputs (Continued)
Test Level Symbol Min Parameter
RS, BIST, STAGG, SLEEP, DRTYPE (Control Input Voltages)
Resistor to ground
Voltage level
Input low current
Resistor to ground
Voltage level
Input high current
SDATA, SLDN, SCLK, RESET
Logic low
Logic high
CLKDACTRL input voltage
1
4
4
R
IL
V
IL
I
IL
R
IH
V
IH
I
IH
V
IL
V
IH
0
–500
10k
2.0
0
2.3
1/3 × V
CCD
Typ Max
10
0.5
Infinite
10
1.0
V
CCA3
2/3 × V
CCD
Unit
Ω
V
µA
Ω
V
µA
V
V
V
Notes: 1. The DC analog common mode voltage is provided by ADC.
2. Measured with SDA OFF(ADC Jitter = 120 fs rms). ADC Jitter with SDA = ON is 150 fs rms, and 170 fs rms with SDA = ON and fully tuned.
3. DRR logic programmable with 3WSI interface. Default value is DRR active low.
for control signal settings.
2.4
Converter Characteristics
Unless otherwise specified:
Recommended conditions of use (refer to Table 2-2 and Table 2-3 on page 6
). Values are specified for
Tj
≥ 65°C and over specified power supplies range.
Typical values are given at ambient (Tj ~65°C) with typical power supplies.
These conditions apply for all tables from Table 2-5 to
Table 2-5.
DC Converter Characteristics
Test Level Parameter
DC Accuracy
Resolution
Differential non-linearity
Differential non-linearity
Gain center value
Gain error drift
Input offset voltage
1
4
4
4
1
4
4
1
4
Symbol
N
DNLrms
DNL+
DNL-
INL-
INL+
G
G(T)
OFFSET
Min
–1.0
0.95
Typ
–3.0
2.3
1
50
10
0.22
0.75
0.6
Max
1.05
Unit
bit
LSB
LSB
LSB
LSB
LSB ppm/
° C mV –20 20
Notes: 1. Histogram testing at Fs = 2.6 Gsps Fin = 100 MHz satured.
DNLrms is the deviation from ideal ADC rms quantification noise. For reference, a DNL of 0.22 LSB rms has the same order of magnitude as the 10 Bit rms quantification noise : 1 LSB/ SQRT(12) = 1 LSB /
3.46 = 0.288 LSB rms.
1076D–BDC–10/12
9
e2v semiconductors SAS 2012
EV10AS150B
2. ADC gain with programmed default value.
This ADC Gain can be fine tuned to “1” by monitoring of the gain adjust function through the 3WSI serial interface.
3. ADC offset with programmed default values.
The ADC Offset can be compensated through the 3WSI up to ± 20 mV by an 8 Bit embedded control
).
Table 2-6.
Dynamic Converter Characteristics
Test
Level Parameter
AC Characteristics
Full Power Input Bandwidth (–3 dB)
Gain Flatness (10 MHz-1.7 GHz)
• Systematic roll-off
• AC gain variation
Gain Flatness (1.7 GHz-2.9 GHz)
• Systematic roll-off
• AC gain variation
Gain Flatness (2.9 GHz-4.4 GHz)
• Systematic roll-off
• AC gain variation
Input Voltage Standing Wave Ratio
(DC
–
3 GHz)
(3 GHz – 5 GHz)
4
4
4
Symbol
FPBW
GF
VSWR
Min Typ
5
–0.5
±0.1
–
0.5
±0.2
–
1.0
±0.3
Max
1.25:1
1.50:1
Unit
GHz dB
Table 2-7.
Dynamic Converter Characteristics
Parameter
Effective Number of Bits
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Test
Level
AC Performance with differential analog input
Ain = –1 dBFS, +1 dBm differential clock (1 Vpp in 100
Ω), 50% external duty cycle, binary output mode
Signal to Noise and Distortion Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
4
4
1
1
4
SINAD
44.0
44.0
50.0
49.5
48.0
48.0
49.0
4
4
1
1
4
Symbol
ENOB
Min
7.0
7.0
Typ
8.0
7.9
7.7
7.7
7.8
Max Unit
dBFS
Bits
10
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Table 2-7.
Dynamic Converter Characteristics (Continued)
Parameter
Signal to Noise Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Total Harmonic Distortion (10 harmonics)
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Test
Level
1
1
4
4
4
1
1
4
4
4
Symbol
SNR
|THD|
Min
45.0
45.0
45.0
45.0
Typ
52.0
52.0
51.0
49.0
50.0
54.0
54.0
51.0
53.0
56.0
Spurious Free Dynamic Range
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Two-Tone 3 rd order Intermodulation distortion
IMD3 (
–
): (2f1-f2, 2f2-f1) at
–
7 dBFS each tone, Fs = 2.6 Gsps
Fin1 = 790 MHz, Fin2 = 800 MHz
Fin1 = 1550 MHz, Fin2 = 1560 MHz
Fin1 = 2640 MHz, Fin2 = 2650 MHz
1
1
4
4
4
1
1
4
4
4
8.1
8.1
8.0
7.9
7.9
57.0
57.0
54.5
57.0
60.0
59
59
59
Signal independent Spurious level
(Fclk/4 with 1:4 DMUX Ratio)
(
–
1 dBFS Analog Input)
Differential driven analog input (100
Ω)
(ADC full scale input power:
–
5 dBm)
(0.5Vpp in 100
Ω Differential analog input)
4 Fclk/4
–
–
78
83
Ain =
–
3 dBFS, + 1 dBm differential clock (1 Vpp in 100
Ω), 50% external duty cycle, binary output mode
Signal to Noise and Distortion Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
1
1
4
4
4
SINAD
45.0
45.0
50.5
51.0
50.0
49.0
49.0
Effective Number of Bits
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
4
|SFDR|
IMD3(
–
)
ENOB
49.0
49.0
7.2
7.2
Max Unit
dBFS dBFS dBFS dBc dBFS dBm dBFS
Bits
1076D–BDC–10/12
11
e2v semiconductors SAS 2012
EV10AS150B
Table 2-7.
Dynamic Converter Characteristics (Continued)
Parameter
Signal to Noise Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Test
Level
1
1
4
4
4
Symbol
SNR
Min
46.0
46.0
Typ
52.0
52.0
52.0
50.5
50.5
Total Harmonic Distortion (10 harmonics)
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Effective Number of Bits
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
1
1
4
4
4
Spurious Free Dynamic Range
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
1
1
4
4
4
|SFDR|
50.0
50.0
60.0
61.0
60.0
59.0
59.0
Ain = –13 dBFS, +1 dBm differential clock (1 Vpp in 100
Ω), 50% external duty cycle, binary output mode
Signal to Noise and Distortion Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
4
4
1
1
4
SINAD
47.0
47.0
51.5
51.5
51.5
51.0
50.5
4
4
1
1
4
|THD|
ENOB
49.0
49.0
7.5
7.5
56.0
55.0
54.5
54.0
55.0
8.2
8.2
8.2
8.2
8.1
Max Unit
dBFS dBFS dBFS dBFS
Bits
12
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
6.5
6.0
5.5
5.0
7.5
7.0
Table 2-7.
Dynamic Converter Characteristics (Continued)
Test
Level Symbol Parameter
Signal to Noise Ratio
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Total Harmonic Distortion (10 harmonics)
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
Spurious Free Dynamic Range
Fs = 2.6 Gsps Fin = 495 MHz
Fs = 2.6 Gsps Fin = 1295 MHz
Fs = 2.6 Gsps Fin = 1595 MHz
Fs = 2.6 Gsps Fin = 2295 MHz
Fs = 2.6 Gsps Fin = 2595 MHz
1
1
4
4
4
1
1
4
4
4
1
1
4
4
4
SNR
|THD|
|SFDR|
Min
50.5
50.5
48.0
48.0
50.0
50.0
Typ
57.0
57.0
57.0
56.5
57.0
53.0
53.0
53.0
53.0
52.0
58.0
58.0
58.0
57.5
58.0
Max Unit
dBFS dBFS dBFS
Note: Dynamic performances near full scale may vary by up to typically #0.6 bit ENOB (compared to performances specified in
Table 2-7 ) around Fclk 0.85 Gsps, 1.6 Gsps, 1.8 Gsps, 2.1 Gsps & 2.35 Gsps as
described in Table 2-2 and Table 2-3
.
Figure 2-3.
ENOB performance versus Fclock
ENOB versus Fclk & Gain - Fin=2595 MHz @-1dBFs
Recommended conditions of use and typical supplies - Tj = 65°C
Min Gain
Typ gain
Max Gain
8.0
Fclock (Gsps)
1076D–BDC–10/12
13
e2v semiconductors SAS 2012
EV10AS150B
2.4.1
2.4.2
When a single tone sinewave at level of around -6 dBFs is applied, a SNR reduction may be observed by typically #3-5 dB resulting in ENOB decrease (#-0.8 bit). (No impact on THD/SFDR). In time domain the
"glitches" that may be observed in this condition are 32 LSB maximum. Their potential occurrence is reduced with minimum gain and minimum VCCA5 and higher temperature. Low Fin (1st Nyquist) and/or low Fclk Frequencies are also more favourable cases.
ADC Converter Characteristics with Single Ended Analog Input
For optimum dynamic performances, it is strongly recommended to use a differential Analog input.
In single ended input at low levels, dynamic performances are similar to those specified in
.
However, for high input levels, THD & SFDR are reduced by around 1dB compared to
.
Sensitivity of Dynamic Performances to Power Supplies
The three figures below illustrate the impact of VCCA3 and VCCA5 power supplies on dynamic performances near full scale depending on Nyquist zone (1st or 2nd).
Measurements conditions are the recommended conditions of use defined in
ambient temperature (Tj ~ 65°C).
Figure 2-4.
ENOB versus VCCA5 & VCCA3
ENOB @2.6Gsps versus Vcca5 & Vcca3
Recommended conditions of use - Tj = 70°C
7.95
7.90
7.85
7.80
7.75
7.70
7.65
4.70
4.75
4.80
4.85
4.90
Vcca5 (V)
4.95
5.00
5.05
5.10
Vcca3 typ, Fin=2.595 GHz
@-1dBFs
Vcca3 min, Fin=2.595 GHz
@-1dBFs
Vcca3 max, Fin=2.595 GHz
@-1dBFs
Vcca3 typ, Fin=1.295 GHz
@-1dBFs
Vcca3 min, Fin=1.295 GHz
@-1dBFs
Vcca3 max, Fin=1.295 GHz
@-1dBFs
14
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 2-5.
SNR versus VCCA5 & VCCA3
SNR @2.6Gsps versus Vcca5 & Vcca3
Recommended conditions of use - Tj = 70°C
52.5
52.0
51.5
51.0
50.5
50.0
49.5
49.0
4.70
4.75
4.80
4.85
4.90
Vcca5 (V)
4.95
5.00
5.05
5.10
Vcca3 typ, Fin=2.595 GHz
@-1dBFs
Vcca3 min, Fin=2.595 GHz
@-1dBFs
Vcca3 max, Fin=2.595 GHz
@-1dBFs
Vcca3 typ, Fin=1.295 GHz
@-1dBFs
Vcca3 min, Fin=1.295 GHz
@-1dBFs
Vcca3 max, Fin=1.295 GHz
@-1dBFs
Figure 2-6.
THD versus VCCA5 & VCCA3
THD @2.6Gsps versus Vcca5 & Vcca3
Recommended conditions of use - Tj = 70°C
-56
-55
-54
-53
-52
4.70
4.75
4.80
4.85
4.90
4.95
Vcca5 (V)
5.00
5.05
5.10
Vcca3 typ, Fin=2.595 GHz
@-1dBFs
Vcca3 min, Fin=2.595 GHz
@-1dBFs
Vcca3 max, Fin=2.595 GHz
@-1dBFs
Vcca3 typ, Fin=1.295 GHz
@-1dBFs
Vcca3 min, Fin=1.295 GHz
@-1dBFs
Vcca3 max, Fin=1.295 GHz
@-1dBFs
1076D–BDC–10/12
15
e2v semiconductors SAS 2012
EV10AS150B
2.4.3
Sensitivity of dynamic performances to temperature
The three following figures illustrate the impact of junction temperature on dynamic performances in 1st
& 2nd Nyquist zone and versus Analog input level.
Measurements conditions are the recommended conditions of use defined in
&
Figure 2-7.
ENOB versus junction temperature in 1st & 2nd Nyquist
ENOB @2.6Gsps Fin = 1.295 GHz
ENOB @2.6Gsps Fin = 2.595 GHz
9.0
9.0
8.5
8.5
8.0
@-1dBFs
@-3dBFs
@-7dBFs
@-13dBFs
8.0
7.5
7.5
7.0
7.0
6.5
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
6.5
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
Figure 2-8.
SNR versus junction temperature in 1st & 2nd Nyquist
SNR @2.6Gsps Fin = 1.295 GHz
54.0
53.0
52.0
51.0
50.0
49.0
48.0
47.0
46.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
@-1dBFs
@-3dBFs
@-7dBFs
@-13dBFs
SNR @2.6Gsps Fin = 2.595 GHz
54.0
53.0
52.0
51.0
50.0
49.0
48.0
47.0
46.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
Figure 2-9.
SFDR versus junction temperature in 1st & 2nd Nyquist
SFDR @2.6Gsps Fin = 1.295 GHz
-69.0
-69.0
SFDR @2.6Gsps Fin = 2.595 GHz
-64.0
-59.0
-54.0
-64.0
@-1dBFs
@-3dBFs
@-7dBFs
@-13dBFs
-59.0
-54.0
-49.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
-49.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
16
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.4.4
Figure 2-10. THD versus junction temperature in 1st & 2nd Nyquist
THD @2.6Gsps Fin = 1.295 GHz
-68.0
-68.0
THD @2.6Gsps Fin = 2.595 GHz
-63.0
-58.0
@-1dBFs
@-3dBFs
@-7dBFs
@-13dBFs
-63.0
-58.0
-53.0
-53.0
-48.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
-48.0
-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130
Tj (°C)
Examples of FFT Spectrum
Following spectrums (32 Kpts FFT) are given for the following conditions:
Unless otherwise specified:
Recommended conditions of use (refer to
Typical supplies
Ambient temperature with heatspreader (Tj ~65 °C)
Figure 2-11. FFT 2.6 Gsps Fin = 1295 MHz @–1 dBFS
2.6Gsps Fin = 1.295 GHz -1dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
1076D–BDC–10/12
17
e2v semiconductors SAS 2012
EV10AS150B
Figure 2-12. FFT 2.6 Gsps Fin = 1295 MHz @-3 dBFS
2.6 Gsps Fin = 1.295 GHz -3dBFs
-40
-50
-60
-70
-80
-90
-100
0
0
-10
-20
-30
100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
Figure 2-13. FFT 2.6 Gsps Fin = 1295 MHz @-7 dBFS
2.6Gsps Fin = 1.295 GHz -7dBFs
-30
-40
-50
-60
-70
-80
-90
-100
0
0
-10
-20
100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
18
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 2-14. FFT 2.6 Gsps Fin = 1295 MHz @–13 dBFS
2.6Gsps Fin = 1.295GHz -13dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
Figure 2-15. FFT 2.6 Gsps Fin = 2595 MHz @–1 dBFS (Carrier at end of the 2nd Nyquist Zone is
folded back to baseband)
2.6Gsps Fin = 2.595 GHz -3dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
1076D–BDC–10/12
19
e2v semiconductors SAS 2012
EV10AS150B
Figure 2-16. FFT 2.6 Gsps Fin = 2595 MHz @–3 dBFS (Carrier at end of the 2nd Nyquist Zone is
folded back to baseband)
2.6Gsps Fin = 2.595 GHz -3dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
Figure 2-17. FFT 2.6 Gsps Fin = 2595 MHz @–7 dBFS (Carrier at end of the 2nd Nyquist Zone is
folded back to baseband)
2.6Gsps Fin = 2.595GHz -7dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
20
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 2-18. FFT 2.6 Gsps Fin = 2595 MHz @–13 dBFS (Carrier at end of the 2nd Nyquist Zone is
folded back to baseband)
2.6Gsps Fin = 2.595 GHz -13dBFs
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
0 100 200 300 400 500 600
F (MHz)
700 800 900 1000 1100 1200 1300
2.5
Transient and Switching Characteristics
Unless otherwise specified:
Recommended conditions of use (refer to Table 2-2 and Table 2-3 on page 6
). Values are specified for
Tj
≥ 65°C and over specified power supplies range.
Typical values are given at ambient (Tj ~65°C) with typical power supplies.
These conditions apply for all tables from Table 2-8 to
Table 2-8.
Transient Characteristics
Parameter Test Level Symbol Min Typ Max Unit
TRANSIENT PERFORMANCE
ADC settling time (± 2%)
Overvoltage recovery time
ADC step response
Rise/Fall Time (10–90%)
4
4
4
4
BER
TS
ORT
10
–12
70
350
400
85
Error/sample ps ps ps
Note: 1. Measured with SDA OFF
ADC Bit Error Rate is related to internal regeneration latches indecision (for analog inputs very close to latches threshold), which may produce large amplitude output errors. The probability of error is measured at 2.6 Gsps maximum operating frequency.
1076D–BDC–10/12
21
e2v semiconductors SAS 2012
EV10AS150B
Table 2-9.
Switching Characteristics
Parameter
Switching Performance and Characteristics
Maximum Clock Frequency
DRTYPE = DR/2, 1:2 mode
DRTYPE = DR, 1:2 mode
DRTYPE = DR/2 1:4 mode
DRTYPE = DR 1:4 mode
Minimum clock frequency
Minimum Clock pulse width (High)
Minimum Clock pulse width (Low)
External clock Duty cycle
Aperture Jitter added by the ADC
Output Rise/Fall time for Data (20% – 80%)
)
Output Rise/Fall time for Output Clock (20% – 80%)
Digital Data Output propagation delay
Data Ready Clock Output propagation delay
Differential propagation delay (Output Data vs. Data clock)
Tskew (40 digital output data)
Output Data Pipeline delay (Latency)
Synchronized 1:2 mode on Port A
Synchronized 1:2 mode on Port B
Synchronized 1:4 mode on Port A
Synchronized 1:4 mode on Port B
Synchronized 1:4 mode on Port C
Synchronized 1:4 mode on Port D
Test Level
4
4
4
4
4
4
4
4
4
4
4
4
1
1
4
4
Symbol
Fs MAX
Fs MIN
TC1
TC2
DCYC
TA
Jitter
TR/TF
TR/TF
TOD
TDR
TOD – TDR
Tskew
Min
2.0
2.0
2.6
2.6
0.192
0.192
45
300
0
Typ
50
350
120
120/120
120/120
2.5
2.35
150
± 15
5.5
4.5
Max
500
1.0
1.0
55
400
180/180
180/180
200
± 35
Unit
Gsps
4 PD 7.5
6.5
5.5
4.5
Clock
Cycles
Staggered 1:2 mode or 1:4 mode
ASYNCRST minimum pulse width
DRR minimum pulse width
5
5
RSTPW
DRRPW
4.5
3
3.5
ns ns
Notes: 1. ADC performance are given for optimum value of 50% external clock duty cycle.
2. ADC Aperture delay and Aperture jitter measured with SDA = OFF. (Default setting at Reset).
3. Rise time and fall time are defined for 100
Ω differentially terminated output load with 2nH and 2 pF termination parasitics.
2.6
Timing Diagram
2.6.1
Aperture Delay
The analog input is sampled on the rising edge of the differential clock input (CLK, CLKN) after TA (aperture delay) of +350 ps typical. Aperture delay (TA) is measured at package input balls with the assumption that the external trace length of analog input and clock input are well matched (6.6 ps/mm of mismatch with
ε r
= 4).
Msps ns ns
% ps fs rms ps ps ns ns ps ps
22
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.6.2
2.6.3
Latency (Simultaneous Mode and Staggered Mode)
In simultaneous output mode with 1:4 DMUX Ratio, the digitized digital output data N, N+1,N+2, N+3 respectively on port A, B, C and D are aligned (on the latest data available N+3 on port D). The data N on port A is available after 7.5 Clock cycles pipeline delay, plus an additional propagation delay TOD
(due to Output Buffers + Package propagation delay). Due to data alignment, the pipeline delay is decreased of one clock cycle for each port from port A to D leading to:
• 7.5 clock cycles for port A
• 6.5 clock cycles for port B
• 5.5 clock cycles for port C
• 4.5 clock cycles for port D
In simultaneous output mode with 1:2 DMUX ratio, the latency becomes respectively 5.5 and 4.5 Clock cycles on port A and port B, with same TOD propagation delay.
In staggered output mode, the latency of the 4 Digital output ports A,B,C,D is the same, since data are presented on output port as soon as available.
In staggered mode for 1:4 DMUX ratio and 1:2 DMUX ratio latency is only 4.5 Clock cycles for port A, B,
C and D (see Timing Diagram).
The output propagation delays TOD1, TOD2, TOD3, and TOD4 of the 4 outputs ports in staggered mode can be considered as identical to TOD.
Data Ready Positioning Versus Output Data (DR/2 Mode and DR Mode)
The Data Ready output clock signal (DR, DRN) is synchronized with ADC (CLK, CLKN) differential clock falling edges to be synchronous with Digital output data (since digital data are output on falling edge of sampling Clock after a latency of 7.5 Clock cycles).
In 1:4 DMUX Ratio, the (DR, DRN) signal is shifted by 2 clock cycles in order to be located at center of data pulse.
In 1:2 DMUX Ratio, the (DR, DRN) signal is shifted by 1 clock cycle to be located at center of data pulse.
Furthermore, the output propagation delay (TDR) of the Data Ready signal and the output propagation delay of the digital data (TOD) are matching very closely, and track each other over full operating temperature range.
Therefore the Data signals and Data clock signals are synchronized at Package output, with the differential Data Ready output clock pulse rising edge being centered within Data pulse, in either dual Data rate mode (DR/2) or DR mode.
In dual data rate (DR/2 mode), the Data clock switches at the same rate as the digital data, and therefore both the rising and falling edges of (DR, DRN) data clock are located at the center of the data pulse over temperature (with max TOD-TDR = 200 ps).
In DR mode, the Data clock switches at twice the rate of the digital data, with the differential Data Ready pulse rising edge being centered within Data pulse and differential falling edges being synchronous with
Data transitions.
1076D–BDC–10/12
23
e2v semiconductors SAS 2012
EV10AS150B
2.6.4
2.6.5
Differential Timing Values TOD-TDR Versus Absolute Timing Values TOD and TDR
The absolute values for TOD and TDR are given for information only, and are corresponding to the digital output data propagation delay and to the Data Ready output propagation delay, related to ADC output buffers throughput delay and package propagation times.
TOD and TDR are measured at Package I/Os level, (Input/Outputs Balls), taking out the board extra propagation delays of the 50
Ω/100Ω controlled impedance lines.
Assuming the application board trace lengths are matched for digital data and data ready lines (within skew limit), one has only to consider the time difference between differential digital data outputs and differential Data clock signals TOD-TDR in simultaneous mode.
In staggered mode, the differential delays are (TOD1-TDR1), (TOD2-TDR2), (TOD3-TDR3), (TOD4-
TDR4). See
Therefore the absolute delay values TOD and TDR are not actually of interest: only the time difference
TOD-TDR has to be actually considered.
The measurement of the relative time difference is easy with matches probes, whereas absolute timings are very difficult to measure.
If the propagation time delays (trace lengths for digital data outputs and Data Ready outputs are well matched, together with ideal TOD = TDR, we shall measure ideally TOD-TDR = 0 at application board outputs (FPGA or DSP incoming signals).
Alignment Between Data Ready and Data Including Skew Management
Real TOD-TDR excluding skew between different data is 150 ps typical, and 200 ps max.
In simultaneous mode, one common Data Ready pulse (DR, DRN) is output for all 4 differential Output
Ports A,B,C,D. Therefore the skews of the 4 differential data ports have to be as low as possible: the skews of the 4 differential output ports due to ADC Package and internal ADC Buffers is less than ±
35 ps max (measured at Package output Balls).
The external skews due to track length differences of the external 100
Ω controlled impedance lines), shall be kept as low as possible.
For example, (considering 3.3 ps/mm propagation time in vacuum), the signal propagation time in a different medium of dielectric constant
ε r
= 4 (at 10 GHz), is yielding to SQRT(
ε r
) × 3.3 ps/mm =
6.6 ps/mm: a 3 mm skew in length between the 40 differential data will result in a 3 mm × 6.6 ps/mm =
~ 20 ps skew, to be added to the ±35 ps skew due to the ADC (Package outputs).
The total skew (ADC + board) will be in this case 70 ps + 20 ps = 90 ps = ± 45 ps in actual skew at FPGA incomings.
Since TOD-TDR is 150 ps typical, and 200 ps max, TOD-TDR shall be added to the total data skew,
(ADC and board), leading to:
(TOD-TDR) + Tskew(total) = (200 ps) + (± 45 ps) = 245 ps maximum uncertainty on positioning of differential Data Ready signal rising edge (DR, DRN), pulse within Data pulse.
In staggered mode, the 4 (four) out-of-range bit function for the 4 (four) Ports A,B,C,D are respectively re-allocated to the Data Ready function, available for each output Port, since latency is different, namely:
Port A (AOR, AORN); Port B (BOR, BORN); Port C (COR, CORN); Port D (DOR, DORN) is respectively replaced by: (DRA, DRAN), (DRB, DRBN), (DRC, DRCN), (DRD, DRDN).
The output propagation delays of the 4 Data Ready pulses (TDR1, TDR2, TDR3, TDR4) are identical to the TDR output propagation delay in simultaneous mode, and are matching the Digital Data propagation delays of the 4 Output Ports over temperature.
24
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.6.6
The relative differential timing values (TOD1-TDR1), (TOD2-TDR2), (TOD3-TDR3), (TOD4-TDR4) are identical to the TOD - TDR differential timing of the simultaneous mode. Skew for port A, B, C, D are considered for 10 bit data in staggered instead of 44 bit of data in simultaneous. Skew between 10 data is slightly better in staggered mode. To simplify calculation skew values of previous section shall be applied.
Minimum Available Time Width Between Data and Data Ready (TD1, TD2)
At 2.6 GHz sampling rate, the time difference between zero crossing point of change of differential data and differential Data Ready output clock rising edge (centered within data pulse) is defined by TD1. The time difference between differential data clock rising edge and next point of change of the differential data output is defined by TD2.
The order of magnitude of time difference TD1-TD2 is identical to TOD-TDR:
TOD-TDR is frequency independent, whereas TD1 and TD2 are sampling frequency dependent:
For example at 1:4 DMUX ratio, the data pulse width at maximum operating frequency of 2.6 GHz is
4 × 384.6 ps = 1.538 ns. Assuming TOD = TDR, the rising edge of differential data clock is ideally located at center of data pulse, with TD1 = TD2 = 770 ps. With a maximum difference of TOD-TDR of
200 ps over temperature, and a total of ±45 ps output Data skew (ADC + Board) and 120 ps for rise and fall times, this is yielding to a minimum available time width for TD1 or TD2 pulses of: 770 ps –200 ps –
45 ps –120 ps = 425 ps (with 50/50 duty cycle).
e2v semiconductors SAS 2012
1076D–BDC–10/12
25
EV10AS150B
2.6.7
Timing Diagram in Simultaneous and Staggered Mode
Figure 2-19. Timing Diagram Simultaneous mode, 1:4 DMUX Ratio
ADC
VIN
ADC CLK
(internal)
N
TA = +350 ps
ADC Pipeline Delay: 4 clock cycles
ADC+1:4 DMUX Pipeline Delay: 7.5 clock cycles on Port A
4 clock cycles: ADC Pipeline Delay
Digital Outputs
( internal )
ADC Data Clock
( internal Fclock/2)
N-4
N
N+1 N+2
N+3 N+4 N+5 N+6
DMUX Even Latches
DMUX Odd Latches
-2
3.5 clock cycles
N
2
1:4 DMUX Pipeline Delay on Port A
4
6
-1
1 3
5
7
Port Select A
Port Select B
Port Select C
Port Select D
N - 1
N - 2
N - 3
N - 4
N
N + 4
N + 1
N + 2
N + 3
N + 5
N + 6
N + 7
DMUX
(internal)
(A0, A0N)...(A9, A9N)
Digital outputs : Port A
(B0, B0N)...(B9, B9N)
Digital outputs : Port B
(C0, C0N)...(C9, C9N)
Digital outputs : Port C
N - 4
N - 3
N
N + 1
N + 2
N+4
N + 5
N + 6
N - 2
(D0, D0N)...(D9, D9N)
Digital outputs : Port D
Clock/4 synchronised on falling edges of Clock
Clock/8 synchronized with output Data on falling edges of Clock
Clock/8 shifted
+ 2 clock cycles
2 Tclock
ADC : 4 cycles
Digital Outputs
Port A, Port B, Port C, Port D
N - 1
2 Tclock
DMUX : 3.5 cycles
N + 3
7.5 clock cycles (ADC+DMUX)
N -8
N -7
N -6
N -5
TOD
N -4
N -3
N -2
N -1
N + 7
Tskew
TD1 TD2
Tskew
Port A : N,
Port B : N+1
Port C : N+2
Port D :
N+3
TDR
(DR, DRN) Data Ready
(in DR/2 mode)
26
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 2-20. Timing Diagram Staggered mode, 1:4 DMUX Ratio
TA = +350ps
N
N+1
N+2 N+3
A B C D
Pipeline Delay Port A, B ,C ,D : 4.5 clock cycles
VIN
ADC CLK
ADC Pipeline Delay: 4 clock cycles
ADC
(internal)
ADC Digital
Outputs
ADC Data Clock
(Fclock / 2)
N-4
N
N+1 N+2 N+3
N+4 N+5 N+6
DMUX Even Latches
-2
N
2
4 6
DMUX Odd Latches
-1
1 3
5
7
ADC Fclock / 4
DMUX
(internal)
(A0, A0N)...(A9, A9N)
Port Select A
Port Select B
Port Select C
(B0, B0N)...(B9, B9N)
Port Select D
Latches Port A
Latches Port B
(C0, C0N)...(C9, C9N)
Latches : Port C
N - 1
N
N + 4
N - 4
N - 2
N - 3
N + 1
N + 2
N + 5
N + 6
N - 4
N + 3
Port A B C D DMUX Latency : 0.5 clock
N
N+4
N + 7
N - 3
N - 2
N + 1
N + 2
N + 5
N + 6
(D0, D0N)...(D9, D9N)
Port A (N) : 4.5 Clock Cycles
(DRA, DRAN) in DR/2 mode
Latches : Port D
4.5 ADC+DMUX
Data for N
N - 1
TOD1
TDR1
N + 3
N
N + 7
Tskew
TD
1 TD2
N+4
Tskew
Port B (N+1) : 4.5 Clock Cycles
(DRB, DRBN) in DR/2 mode
N+1
4.5 ADC+DMUX
Data for N+1
TOD2
N+1
N+5
TDR2
Port C (N+2 ) : 4.5 Clock Cycles
(DRC, DRCN) in DR/2 mode
N+2
4.5 ADC+DMUX
Data for N+2
TOD3
TDR3
N+2
Port D (N+3 ): 4.5 Clock Cycles
(DRD, DRDN) in DR/2 mode
4.5 ADC+DMUX
TOD4
N+3
Data for N+3
N+3
Digital Outputs & Data Ready Signals for Port A, Port B, Port C, Port D:
Digital Data: after TOD1, TOD2, TOD3, TOD4 ; Data Ready Signals: after TDR1, TDR2, TDR3, TDR4
N+6
Tskew
TD1 TD2
N+7
1076D–BDC–10/12
27
e2v semiconductors SAS 2012
EV10AS150B
2.7
Digital Output Data Coding
Table 2-10.
Digital Output Data Coding Table
Differential analog input
> 250.25 mV
250.25 mV
249.75 mV
125.25 mV
124.75 mV
0.25 mV
–0.25 mV
–124.75 mV
–124.25 mV
–249.75 mV
–250.25 mV
< –250.25 mV
Voltage level
>Top end of full scale + ½ LSB
Top end of full scale + ½ LSB
Top end of full scale – ½ LSB
3
/
4 full scale + ½ LSB
3
/
4 full scale – ½ LSB
Mid scale + ½ LSB
Mid scale – ½ LSB
1
/
4 full scale + ½ LSB
1
/
4 full scale – ½ LSB
Bottom end of full scale + ½ LSB
Bottom end of full scale – ½ LSB
< Bottom end of full scale – ½ LSB
Natural Binary
MSB……….LSB OR
0 0 0 0 0 0 0 0 0 0 1
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 1 0
0 0 1 1 1 1 1 1 1 1 0
0 1 0 0 0 0 0 0 0 0 0
0 1 1 1 1 1 1 1 1 1 0
1 0 0 0 0 0 0 0 0 0 0
1 0 1 1 1 1 1 1 1 1 0
1 1 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1 1 1 1
Digital output
Binary 2’s Complement
MSB……....LSB OR
1 0 0 0 0 0 0 0 0 0 1
1 0 0 0 0 0 0 0 0 0 0
1 0 0 0 0 0 0 0 0 1 0
1 0 1 1 1 1 1 1 1 1 0
1 1 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 0
0 0 0 0 0 0 0 0 0 0 0
0 0 1 1 1 1 1 1 1 1 0
0 1 0 0 0 0 0 0 0 0 0
0 1 1 1 1 1 1 1 1 0 0
0 1 1 1 1 1 1 1 1 1 0
0 1 1 1 1 1 1 1 1 1 1
Gray Coding
MSB……..LSB OR
0 1 1 1 1 1 1 1 1 1 1
0 1 1 1 1 1 1 1 1 1 0
0 1 1 1 1 1 1 1 1 0 0
0 1 0 1 1 1 1 1 1 1 0
0 0 0 1 1 1 1 1 1 1 0
0 0 1 1 1 1 1 1 1 1 0
1 0 1 1 1 1 1 1 1 1 0
1 0 0 1 1 1 1 1 1 1 0
1 1 0 1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1 1 1 1
A9 = B9 = C9 = D9 = MSB
A0 = B0 = C0 = D0 = LSB
Note: 1. Be aware that code 0x000 is obtained for positive full scale analog input and code 0x3FF for negative full scale.
2. Refer to Table 4-9 on page 51
for selection between natural binary, binary two’s complement or Gray coding.
2.8
Explanation of Test Levels
Table 2-11.
Test Levels
1
2
100% production tested at +25
° C
(for C temperature range
)
100% production tested at +25
° C
, and sample tested at specified temperatures for V temperature
3
4
5
Sample tested only at specified temperature
Parameter is guaranteed by design and characterization testing (thermal steady-state conditions at specified temperature)
Parameter is a typical value only guaranteed by design only
Only MIN and MAX values are guaranteed (typical values are issuing from characterization results).
Notes: 1. Unless otherwise specified.
2. Refer to
Section 7. ”Ordering Information” on page 72
.
28
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
2.9
Definition of Terms
Table 2-12.
Definition of Terms
(Fs max)
(Fs min)
Maximum Sampling
Frequency
Minimum Sampling frequency
Performances are guaranteed up to Fs max
Performances are guaranteed for clock frequency higher than Fs min
(BER)
(FPBW)
(SSBW)
(SINAD)
(SNR)
(THD)
(SFDR)
Bit Error Rate
Full power input bandwidth
Small Signal Input bandwidth
Signal to noise and distortion ratio
Signal to noise ratio
Total harmonic distortion
Spurious free dynamic range
Probability to exceed a specified error threshold for a sample at maximum specified sampling rate. An error code is a code that differs by more than ± 32 LSB from the correct code.
Analog input frequency at which the fundamental component in the digitally reconstructed output waveform has fallen by 3 dB with respect to its low frequency value (determined by FFT analysis) for input at Full Scale
–1 dB (–1 dBFS).
Analog input frequency at which the fundamental component in the digitally reconstructed output waveform has fallen by 3 dB with respect to its low frequency value (determined by FFT analysis) for input at Full Scale
–10 dB (–10 dBFS).
Ratio expressed in dB of the RMS signal amplitude, set to 1 dB below Full Scale (–1 dBFS), to the RMS sum of all other spectral components, including the harmonics except DC.
Ratio expressed in dB of the RMS signal amplitude, set to 1 dB below Full Scale, to the RMS sum of all other spectral components excluding the ten first harmonics.
Ratio expressed in dB of the RMS sum of the first ten harmonic components, to the RMS input signal amplitude. It may be reported in dBFS (i.e, related to converter Full Scale), or in dBc (i.e, related to input signal level).
Ratio expressed in dB of the RMS signal amplitude to the RMS value of the highest spectral component
(peak spurious spectral component). The peak spurious component may or may not be a harmonic. It may be reported in dBFS (i.e., related to converter Full Scale), or in dBc (i.e, related to input signal level).
(ENOB)
(DNL)
(INL)
(TA)
(JITTER)
(TS)
(ORT)
(TOD)
(TDR)
(TD1)
Effective Number Of Bits
Differential non linearity
Integral non linearity
Aperture delay
Aperture uncertainty
Settling time
Overvoltage recovery time
Digital data Output delay
Data ready output delay
Time delay from Data transition to Data Ready
ENOB =
Where A is the actual input amplitude and FS is the full scale range of the ADC under test
6.02
The Differential Non Linearity for an output code i is the difference between the measured step size of code i and the ideal LSB step size. DNL (i) is expressed in LSBs. DNL is the maximum value of all DNL (i). DNL error specification of less than 1 LSB guarantees that there are no missing output codes and that the transfer function is monotonic.
The Integral Non Linearity for an output code i is the difference between the measured input voltage at which the transition occurs and the ideal value of this transition.
INL (i) is expressed in LSBs, and is the maximum value of all |INL (i)|.
Delay between the rising edge of the differential clock inputs (CLK, CLKN) (zero crossing point), and the time at which (V
IN,
V
INN
) is sampled.
Sample to sample variation in aperture delay. The voltage error due to jitter depends on the slew rate of the signal at the sampling point.
Time delay to achieve 0.2% accuracy at the converter output when a 80% Full Scale step function is applied to the differential analog input.
Time to recover 0.2% accuracy at the output, after a 150% full scale step applied on the input is reduced to midscale.
Delay from the rising edge of the differential clock inputs (CLK, CLKN) (zero crossing point) to the next point of change in the differential output data (zero crossing) with specified load.
Delay from the falling edge of the differential clock inputs (CLK, CLKN) (zero crossing point) to the next point of change in the differential output data (zero crossing) with specified load.
Time delay between Data transition to output clock (Data Ready). If output clock is in the middle of the Data,
TD1 = Tdata/2
1076D–BDC–10/12
29
e2v semiconductors SAS 2012
EV10AS150B
Table 2-12.
Definition of Terms (Continued)
(TD2)
TD1-TD2
(TC)
(TPD)
(TR)
(TF)
(PSRR)
(IMD)
Time delay from Data
Ready to Data
Encoding clock period
Pipeline Delay
Rise time
Fall time
Power supply rejection ratio
Intermodulation
Distortion
Time delay between output clock (Data Ready) to Data transition. If output clock is in the middle of the Data,
TD2 = Tdata/2.
The difference TD1-TD2 gives an information if the output clock is centered on the output data. If output clock is in the middle of the data TD1 = TD2 = Tdata/2.
TC1 = Minimum clock pulse width (high) TC = TC1 + TC2
TC2 = Minimum clock pulse width (low)
Number of clock cycles between the sampling edge of an input data and the associated output data being made available, (not taking in account the TOD).
Time delay for the output DATA signals to rise from 20% to 80% of delta between low level and high level.
Time delay for the output DATA signals to fall from 20% to 80% of delta between low level and high level.
Ratio of input offset variation to a change in power supply voltage.
(NPR)
(VSWR)
Noise Power Ratio
Voltage Standing Wave
Ratio
The two tones intermodulation distortion (IMD) rejection is the ratio of either input tone to the worst third order intermodulation products.
The NPR is measured to characterize the ADC performance in response to broad bandwidth signals. When applying a notch-filtered broadband white-noise signal as the input to the ADC under test, the Noise Power
Ratio is defined as the ratio of the average out-of-notch to the average in-notch power spectral density magnitudes for the FFT spectrum of the ADC output sample test.
The VSWR corresponds to the ADC input insertion loss due to input power reflection. For example a VSWR of 1.2 corresponds to a 20 dB return loss (that is, 99% power transmitted and 1% reflected).
30
1076D–BDC–10/12 e2v semiconductors SAS 2012
3.
Pin Description
3.1
Pinout View
Figure 3-1.
EBGA 317 Pinout Table (View from Bottom of the Package)
EV10AS150B
Dam and Fill
(not an exposed pad)
Dam and Fill
(not an exposed pad)
Note: Area in dashed line corresponds to dam & fill (not an exposed pad) e2v semiconductors SAS 2012
1076D–BDC–10/12
31
EV10AS150B
3.2
Pin Description Table
Table 3-1.
Pin Description Table
Signal Name
POWER SUPPLIES
Pin Number
V
CCA5
V
CCA3
V
PLUSD
V
CCD
A24, A26, A27,
B24, B26, B27,
C24, C26, C27,
D24, D26, D27,
E24, E26, F25, L25,
L26, M27, R21,
T21, U21
A25, B22, B25,
C20, C22, C25,
D20, D22, D25,
E20; E22, E25,
F20, F22, F24, K25,
K26, L27, M25,
M26, N26, N27,
R20, T20
C4, C5, C6, C7, C9,
C11, C13, C14,
C15, C16, C19, D5,
D6, D7, D9, D11,
D13, D19, E3, E19,
F19, J3, J4, L3, L4,
N3, N4, R3, R4,
R19, T6, T7, T9,
T11, T13, T14, T15,
T19, U4, U5, U6,
U7, U9, U11, U13,
U14, U15
C3, C8, C10, C12,
D3, D4, D8, D10,
D12, D16, E4, E17,
G3, G4, K3, K4,
R16, T3, T4, T5,
T8, T10, T12, T16,
T17, U3, U8, U10,
U12
Description
Analog 4.9V
Power Supply (ADC)
Analog 3.25V
Power Supply (ADC)
Output 2.5V
Power Supply (ADC and
DMUX)
Digital 3.3V Power Supply
(DMUX)
SUB D14, D15, R17
Substrate connect to Board
Ground Plane (DGND)
DGND
A19, A20, B19,
B20, C17, C18,
D17, D18, E18, F3,
F4, F18, H3, H4,
M3, M4, P3, P4,
R18, T18, U16,
U17, U19, U20
Digital Ground, connect to
Board Ground Plane
Dir.
Equivalent Simplified Schematics
32
1076D–BDC–10/12 e2v semiconductors SAS 2012
Table 3-1.
Signal Name
AGND
Pin Description Table (Continued)
Description Pin Number
B21, B23, C21,
C23, D21, D23,
E21, E23, F21, F23,
F26, F27, G25,
G26, G27, H25,
H26, J25, J26, K27,
N25, P25, R22,
R23, R24, R25,
R26, R27, T22, t23,
T24, T25, T26, T27,
U22, U23, U24,
U25, U26, U27,
V21, V23, V24,
V26, V27, W22,
W25, W26, W27
Analog Ground, connect to
Board Ground Plane
NO CONNECT
DNC B18
Do Not connect (Leave this pin floating)
NC
W1, W18, W17,
W19, V19, W20,
V20, A1, V18, P26,
V22, V25
Pins can be left no connect or grounded.
ANALOG INPUTS
Dir.
Equivalent Simplified Schematics
V
IN
V
INN
H27
J27
ADC In-Phase Analog input
ADC Inverted-Phase
Analog input
(On chip 2 × 50
Ω terminated, 3V biased)
I
VCCA5
VCCA3
VIN
VINN
ESD
ESD
55 55
40p
ESD
ESD
550 550
AGND
EV10AS150B
6 mA
80
6 mA
1076D–BDC–10/12
33
e2v semiconductors SAS 2012
EV10AS150B
Table 3-1.
Signal Name
CLOCK INPUTS
Pin Description Table (Continued)
Pin Number Description
CLK
CLKN
W24
W23
Dir.
Equivalent Simplified Schematics
ADC Clock Differential
Inputs
(On chip 50
Ω terminated,
3V biased)
I
VCCAS
VCCA3
CLK
CLKN
ESD
ESD
55 55
60p
ESD ESD
550 550
AGND
RESET INPUT
DRR
ASYNCRST
P27
B17
VCCA3
2k 2k 15k
ESD
10k
ADC Data Ready reset
(active High or Low depending on bit D8 of state register at address
0110)
I
DRR
ESD
AGND
10k
10k
VCCD (3.3V ± 5%)
ESD
12K
DMUX Asynchronous
Reset
• Leave floating or connect to V
CCD
for RESET mode
• Connect to Ground for normal mode
I
ASYNCRST
ESD
12K 500
500
12K 4K 4K
DGND (0V)
SUBST (0V)
3K 3K
87
μΑ
34
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Table 3-1.
Pin Description Table (Continued)
Signal Name
DIGITAL OUTPUTS
Pin Number Description
A0…A9
A0N…A9N
AOR/DRAN
AORN/DRA
B0…B9
B0N…B9N
BOR/DRBN
BORN/DRB
B16, B15, B14,
B13, B12, B11,
B10, B9, B8, B7
A16, A15, A14,
A13, A12, A11,
A10, A9, A8, A7
B6
A6
B5, B4, B3, B2, C2,
D2, E2, F2, G2, H2
A5, A4, A3, A2, B1,
C1, D1, E1, F1, G1
J2
H1
In-Phase Digital Outputs
Port A
LVDS compatible
A0 = LSB, A9 = MSB
Inverted-Phase Digital
Outputs Port A
LVDS compatible
A0N = LSB, A9N = MSB
Port A Out of Range in simultaneous mode or
Port A Data Ready Output
Clock in staggered mode
In-Phase Digital Outputs
Port B
LVDS compatible
B0 = LSB, B9 = MSB
Inverted-Phase Digital
Outputs Port B
LVDS compatible
B0N = LSB, B9N = MSB
Port B Out of Range in simultaneous mode or
Port B Output Clock in staggered mode
Dir.
Equivalent Simplified Schematics
O
O
O
O
O
O
590
100
590
820
μΑ
VPLUSD (2.5V ± 5%)
490 490
1.8K
1.8K
DGND (0V)
55
ESD ESD
ESD ESD
SUBST (0V) out outn
1076D–BDC–10/12
35
e2v semiconductors SAS 2012
EV10AS150B
Table 3-1.
Signal Name
C0…C9
C0N…C9N
COR/DRCN
CORN/DRC
D0…D9
D0N…D9N
DOR/DRDN
DORN/DRD
DR
DRN
Pin Description Table (Continued)
Pin Number
M2, N2, P2, R2, T2,
U2, V1, V2, V3, V4
L1, M1, N1, P1, R1,
T1, U1, W2, W3,
W4
V5
W5
V6, V7, V8, V9,
V10, V11, V12,
V13, V14, V15
W6, W7, W8, W9,
W10, W11, W12,
W13, W14, W15
V16
W16
J1
K2
Description
In-Phase Digital Outputs
Port C
LVDS compatible
C0 = LSB, C9 = MSB
Inverted-Phase Digital
Outputs Port C
LVDS compatible
C0N = LSB, C9N = MSB
Port C Out of Range in simultaneous mode
Or
Port C Data Ready Output
Clock in staggered mode
In-Phase Digital Outputs
Port D
LVDS compatible
D0 = LSB, D9 = MSB
Inverted-Phase Digital
Outputs Port D
LVDS compatible
D0N = LSB, D9N = MSB
Port D Out of Range in simultaneous mode or
Port D Data Ready Output
Clock in staggered mode
Differential Data Ready
Output Clock
LVDS compatible
CONTROL INPUT FUNCTIONS
RESET
SDATA
SLDN
E27
A23
A21
Reset for fast programming of 3WSI to default values.
Data input of 3WSI
(3 Wires Serial Interface)
Load Enable input of 3WSI
Dir.
O
O
O
O
O
O
O
I
I
I
Equivalent Simplified Schematics
590
VCCA3
100
590
820
μΑ
VPLUSD (2.5V ± 5%)
490
1.8K
1.8K
DGND (0V)
10k
8k
490
ESD
8k
55
ESD ESD
ESD ESD
SUBST (0V)
19.6k
out outn
SCLK A22 Clock input of 3WSI I
SCLK
SDATA
Reset
SLDN
ESD
AGND
14k
20k
20k
36
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Table 3-1.
Signal Name
Pin Description Table (Continued)
Pin Number Description
SLEEP A18
Dir.
Equivalent Simplified Schematics
8K
VCCD (3.3V ± 5%)
1.2K
1.2K
ESD
DMUX SLEEP mode
Enable
Leave floating or connect to V
CCD
for normal mode
Connect to Ground for
SLEEP mode
I
10
Ω = 0
10 K
Ω = 1
RS
DRTYPE
BIST
SLEEP
STAGG
ESD
16K
10K
14K 10K
DGND (0V)
STAGG
DRTYPE
RS
BIST
A17
K1
L2
V17
SUBST (0V)
DMUX Staggered mode
Enable
Leave floating or connect to V
CCD
for normal mode
Connect to Ground for
STAGG mode
DMUX Output Clock mode selection
Connect to Ground for
DR/2 type
Leave floating or connect to V
CCD
for DR type
DMUX Ratio mode selection
Connect to Ground for 1:2 ratio
Leave floating or connect to V
CCD
for 1:4 ratio
DMUX BIST mode
Leave floating or connect to V
CCD
for normal mode
Connect to Ground for
BIST mode
I
I
I
I
10
Ω = 0
10 K
Ω = 1
RS
DRTYPE
BIST
SLEEP
STAGG
ESD
ESD
SUBST (0V)
8K
VCCD (3.3V ± 5%)
1.2K
1.2K
16K
DGND (0V)
10K
14K 10K
1076D–BDC–10/12
37
e2v semiconductors SAS 2012
EV10AS150B
Table 3-1.
Signal Name
Pin Description Table (Continued)
Pin Number Description
CLKDACTRL U18
DMUX clock Delay control
(from 1/3 × V
CCD
to 2/3 ×
V
CCD
)
I
Dir.
Equivalent Simplified Schematics
VCCD (3.3V ± 5%)
ESD
10.0 K 2.3 K 2.3 K
4K 4K
CLKDACTRL
3K 3K
ESD
200 µA
10.0 K
3K 3K
DGND (0V)
SUBST (0V)
CONTROL OUTPUT FUNCTIONS
DIOD E
DIODE ADC W21
ADC Die Junction
Temperature monitoring
I/O
2 protection diodes
1 mA
AGND
Protection
Diodes
38
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.
Theory of Operation
4.1
Overview
The EV10AS150B is a 10-bit 2.6 Gsps ADC combined with a 1:4 demultiplexer (DMUX) allowing to lower the output Data stream (10-bit data and one out-of-range bit) by a selectable factor of 2 or 4. (Please note that for maximal operating speed, the DMUX has to be used in 1:4 mode).
The ADC works in fully differential mode from analog input up to digital outputs. The ADC differential
Analog Inputs and Clock Inputs are on chip 100
Ω terminated.
The ADC analog input can be driven in either single ended or differential configuration without significant impact on dynamic performances (SFDR), but slight degradation on THD since the weighting of second and third harmonics are similar in single-ended.
The output Data clock and Digital output data are LVDS logic compatible, and should be 100
Ω differentially terminated (2nH // 2 pF maximum output loading parasitics).
The main functions of the EV10AS150B ADC are digitally controlled via on-chip DACs controlled by 3wire serial interface (3WSI):
• Sampling Delay Adjust, with a tuning range of ~ 120 ps, with a coarse and a fine tuning available (8bit resolution on the fine tuning of ~30 ps).
• Offset Control, ± 20 mV, 8-bit resolution.
• Gain Control, ±0.5 dB, 8-bit resolution
• Delay adjustment between the logical clock and the T/H clock, 0 to 30 ps (5-bit resolution).
• Internal clock duty cycle adjust from 60% to 70% at 2.6 Gsps (5-bit resolution).
• Selectable binary or the Gray coding mode.
• Selectable inversion of MSB for binary two's complement.
• Reset to program quickly all registers to default values.
• NAP (ADC part) mode to save power when device is not used.
The Sampling Delay Adjust function (controlled through the 3WSI) may be used to fine-tune the ADC aperture delay from 0 to 120 ps. The SDA function is very useful when interleaving multiple ADCs.
The output demultiplexing ratio 1:4 or 1:2 can be selected by the means of RS digital control input.
The data outputs are available at the output of the EV10AS150B with two different latency modes:
• Staggered (Low latency): data output on ports A, B, C and D are shifted from one clock cycle between two successive ports.
• Simultaneous: data output on all ports A, B, C and D are aligned.
A Built-In Test (BIST) is provided for quick debug or acquisition setup: activation of checker board like pattern generator.
The ADC junction temperature monitoring is made possible through the DIODE ADC input by sensing the voltage drop across a diode implemented on the ADC close to chip hot point.
The EV10AS150B is delivered in an Enhanced Ball Grid Array (EBGA), suitable for applications subjected to large thermal variations (thanks to its TCE which is similar to FR4 material TCE).
1076D–BDC–10/12
39
e2v semiconductors SAS 2012
EV10AS150B
4.2
Functional Pin Table
Table 4-1.
Name
V
CCA5
V
CCA3
V
CCD
V
PLUSD
AGND
DGND
CLK, CLKN
V
IN
,
V
INN
DRR
ASYNCRST
DR/DRN
A0… A9
A0N…A9N
Functional Pin Table
Function
Analog 4.9V Power Supply
Analog 3.25V Power supply
Digital 3.3V Power supply
Output 2.5V Power Supply
Analog Ground
Digital Ground
ADC clock input signal
ADC Analog Input signal
ADC Data Ready reset
DMUX Asynchronous Reset
Data Ready output clock
Digital Output Data Port A
AOR/DRAN,
AORN/DRA
Out-of-range bit Port A or Data Ready output clock in staggered mode Port A
B0… B9
B0N…B9N
Digital Output data Port B
BOR/DRBN,
BORN/DRB
Out-of-range bit Port B or Data Ready output clock in staggered mode Port B
C0… C9
C0N…C9N
Digital Output data Port C
COR/DRCN,
CORN/DRC
Out-of-range bit Port C or Data Ready output clock in staggered mode Port C
D0… D9
D0N…D9N
Digital Output data Port D
DOR/DRDN,
DORN/DRD
RS
CLKDACTRL
DIODE ADC
Out-of-range bit Port D or Data Ready output clock in staggered mode Port D
DMUX Ratio Selection signal (1:2 or 1:4)
Control signal input for fine tuning of internal ADC Data Clock Delay line.
Diode for die Junction Temperature monitoring (ADC)
Name
BIST
SLEEP
DRTYPE
STAGG
SCLK
SDATA
SLDN
RESET
2
VIN, VINN
CLK, CLKN
DRR
ASYNCRST
SDATA
SLDN
SCLK
RESET
2
CLKDACTRL
VCCA3
3.25V
VCCA5
4.9V
VCCD
3.3V
VPLUSD
2.5V
EV10AS150B
20
20
2
20
2
20
2
2
2
[A0…A9]
[A0N…A9N]
AOR/DRAN,
AORN/DRA
[B0…B9]
[B0N…B9N]
BOR/DRBN,
BORN/DRB
[C0…C9]
[C0N…C9N]
COR/DRCN,
CORN/DRC
[D0…D9]
[D0N…D9N]
DOR/DRDN,
DORN/DRD
DR, DRN
SLEEP
STAGG
RS
BIST
DRTYPE
DIODE ADC
AGND
Function
DGND
Built In output checker board
Pattern generation
DMUX Sleep mode input pin selection signal
Data Ready Output Clock Type selection
(Dual Data Rate DR/2 or DR)
Staggered mode selection for Data outputs (low latency)
3 Wire Serial Interface (3WSI)
Clock input
3 Wire Serial Interface (3WSI) serial Input Data
3 Wire Serial Interface (3WSI) Load
Enable signal (Active low)
3WSI Reset input (force registers default settings).
40
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.3
RESETs and ADC Synchronization
There are three reset signals available on the device to start the device properly:
• DRR (Data Ready Reset) is used to reset and synchronize the Data Ready Output Clock of the ADC.
DRR ensures that the first edge of the ADC output Data clock after DRR reset pulse is always a rising edge. This ensures that the first Data for N sampled after ADC reset corresponds to the first acquisition in the DMUX.
• ASYNCRST is an asynchronous reset used to synchronize the DMUX so that the data are outputted on the right order (Port A, then Port B, then Port C and then Port D). It is also necessary to have a deterministic sequence in BIST mode.
• RESET is a signal used to reset (load default settings) the 3 Wire Serial Interface (3WSI), see
DRR, RESET and ASYNCRST are mandatory for proper device initialization.
Please refer to the paragraph 5.5 for more information on how to implement these reset functions.
DRR is active high or low depending on bit D8 of state register at address 0110 while ASYNCRST is active high.
It is mandatory to apply ASYNCRST while DRR is active.
2 sequences are recommended depending on the applications.
• Sequence 1: to be used in the case where multiple ADCs synchronization is not needed
1.
Perform a RESET on ADC (DRR active low or high depending on 3WSI settings). Maintain DRR active up to step 4.
2.
While DRR is active, perform an asynchronous reset of the DEMUX (ASYNCRST high).
3.
Release ASYNCRST (ASYNCRST low) (minimum pulse width is 3ns)
4.
Release DRR (minimum DRR pulse width is 3.5ns)
5.
3WSI interface can then be programmed if needed.
During this sequence 1, the clock can be running or not running. But pipeline delay up to ADC outputs will not be deterministic.
• Sequence 2: to be used in the case where multiple ADCs synchronization is needed
1.
Apply a running clock in order to initialize ADC clock path
2.
Stop ADC clock at low level
3.
Perform a RESET on ADC (DRR active low or high depending on 3WSI settings).
4.
Maintain DRR active up to step 7.
5.
While DRR is active, perform an asynchronous reset of the DEMUX (ASYNCRST high).
6.
Release ASYNCRST (ASYNCRST low) (minimum pulse width is 3ns)
7.
Release DRR (minimum DRR pulse width is 3.5ns)
8.
Restart ADC clock
9.
3WSI interface can then be programmed if needed.
With this sequence 2, all ADCs will start on the first rising edge of the clock and they will be synchronized.
1076D–BDC–10/12
41
e2v semiconductors SAS 2012
EV10AS150B
Figure 4-1.
Asynchronous Reset Timing Diagram, 1:2 Mode, Simultaneous Mode (Principle of
Operation)
TA = 350 ps
N VIN
CLK
3.5 ns min
DRR
ASYNCRST
3 ns min
A9…A0
B9…B0
DR
(DR mode)
DR
(DR/2 mode)
TOD + 5.5 cycles
N - 2 N N + 2 N + 4
N - 1 N + 1 N + 3 N + 5
Figure 4-2.
Asynchronous Reset Timing Diagram, 1:4 mode, simultaneous mode (Principle of operation)
TA = 350 ps
VIN
CLK
N
3.5 ns min
DRR
3 ns min
ASYNCRST
A9…A0
B9…B0
C9…C0
D9…D0
DR
(DR mode)
DR
(DR/2 mode)
TOD + 7.5 cycles
N - 4
N - 3
N - 2
N - 1
N
N + 1
N + 2
N + 3
42
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.4
Control Signal Settings (DMUX)
The SLEEP, RS, STAGG, BIST and DRTYPE control signals use the same input buffer.
SLEEP, STAGG, BIST are activated on Logic Low (10
Ω Grounded), and deactivated on Logic High (10
K
Ω to Ground, or tied to V
CCD
= 3.3V, or left floating).
This is illustrated in Figure hereafter:
Figure 4-3.
Control Signal Settings (SLEEP, RS, STAGG, BIST and DRTYPE)
10
Ω
Control
Signal Pin
10 K
Ω
Control
Signal Pin
Not
Connected
Control
Signal Pin
GND
Low Level (‘0’)
GND
High Level (‘1’)
Table 4-2.
Function
BIST
DMUX Mode Settings - Summary
Logic Level
Electrical Level
Static drive Dynamic drive
0
1
10
Ω to ground
10 K
Ω to ground
N/C
V(BIST) = V
IL
V(BIST)
≥ V
IH
Description
BIST: Checker board on output data
Normal conversion
0 10
Ω to ground
SLEEP
STAGG
1
0
10 K
Ω to ground
N/C
10
Ω to ground
10 K
Ω to ground
1
RS
0
1
N/C
10
Ω to ground
10 K
Ω to ground
N/C
DRTYPE
0
1
10
Ω to ground
10 K
Ω to ground
Note: 1. Refer to
for logical levels.
V(SLEEP) = V
V(SLEEP)
≥ V
V(STAGG) = V
V(STAGG)
≥ V
V(RS) = V
V(RS)
≥ V
IL
IH
IL
IH
IL
IH
V(DRTYPE) = V
IL
V(DRTYPE)
≥ V
IH
Power reduction mode (the outputs are fixed at an arbitrary LVDS level)
Normal conversion
Staggered mode
Simultaneous mode
1:2 ratio
1:4 ratio
DR/2 mode
DR mode
1076D–BDC–10/12
43
e2v semiconductors SAS 2012
EV10AS150B
4.4.1
4.4.2
DMUX Ratio
The demultiplexer ratio is programmable thanks to the RS Ratio selection signal.
Table 4-3.
DMUX Ratio Selection Settings
RS
0
1
DMUX Ratio
1:2
1:4
Figure 4-4.
DMUX in 1:2 Ratio
Input Words:
1, 2, 3, 4, 5, 6, 7, 8…
1:2
Output Words:
Port A 1 3 5 7 …
Port B 2 4 6 8 …
Port C Not Used
Port D Not Used
Figure 4-5.
DMUX in 1:4 Ratio
Input Words:
1, 2, 3, 4, 5, 6, 7, 8, 9…
Output Words:
1:4
Port A 1 5 9 …
Port B 2 6 10 …
Port C 3 7 11 …
Port D 4 8 12 …
DMUX Data Ready Output Clock Selection (DRTYPE)
DMUX Data Ready Output Clock is used to latch EV10AS150B output data.
Two modes for the output Data clock rate can be selected (via DRTYPE):
• DR mode: The output data clock frequency is 1/4 th
the ADC Clock frequency (assuming 1:4 DMUX
Ratio), and switches at twice the output data rate. Therefore only the rising edge is considered as active for output data registering (Rising edge of differential output Data clock (DR, DRN) is located at center of the digital data pulse).
• DR/2 mode: The output data clock frequency is 1/8 th
the ADC Clock frequency (assuming 1:4 DMUX
Ratio), the output clock switches at same rate as the digital data. Therefore, both output Data clock rising and falling edges are active for output data registering (Rising and Falling edge of differential output Data clock (DR, DRN) are located at center of the digital data pulse).
This is illustrated in the following figures:
Figure 4-6.
DR Mode
DR
Data Out
Figure 4-7.
DR/2 Mode
DR
Data Out
44
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.4.3
Table 4-4.
DMUX Output Clock Type Selection Settings
DRTYPE
1
0
When DRTYPE is left floating, the default mode is DR.
DMUX Output Clock Type
DR
DR/2
DMUX Output Data Mode (STAGG)
Two output modes are provided:
• Staggered: the output data come out of the DMUX the one after the other;
• Simultaneous: the output data come out of the DMUX at the same time.
In staggered mode, the output clock for each port is provided by the DRA, DRAN, DRB, DRBN, DRC,
DRCN and DRD, DRDN signals which corresponds respectively to the AORN, AOR, BORN, BOR,
CORN, COR, DORN and DOR.
The Simultaneous mode is the default mode (STAGG left floating or at logic “1”).
The Staggered mode is activated by the means of the STAGG input (active low).
Figure 4-8.
Simultaneous Mode in 1:4 Ratio (STAGG = 1)
DR
in DR mode
DR in DR/2 mode
Data Out Port A
N
Data Out Port B
Data Out Port C
N + 1
N + 2
Data Out Port D
N + 3
N + 4
N + 5
N + 6
N + 7
Figure 4-9.
Simultaneous Mode in 1:2 Ratio (STAGG = 1)
DR in DR mode
DR in DR/2 mode
Data Out Port A
N
Data Out Port B
N + 1
N + 4
N + 5
1076D–BDC–10/12
45
e2v semiconductors SAS 2012
EV10AS150B
Figure 4-10. Staggered Mode in 1:2 Ratio (STAGG = 0)
Data Out Port A
N
DRA in DR mode
DRA in DR/2 mode
Data Out Port B
DRB in DR mode
DRB in DR/2 mode
DR in DR mode
DR
in DR/2 mode
N - 1 N + 1
N + 2
N + 3
Figure 4-11. Staggered Mode in 1:4 Ratio (STAGG = 0)
N
Data Out Port A
DRA in DR mode
DRA in DR/2 mode
Data Out Port B
DRB in DR mode
DRB in DR/2 mode
N + 1
N - 2 N + 2
Data Out Port C
DRC
In DR mode
DRC in DR/2 mode
N - 1
Data Out Port D
DRD in DR mode
DRD in DR/2 mode
DR in DR mode
DR in DR/2 mode
N + 4
N + 3
N + 5
N + 6
46
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.4.4
4.4.5
4.4.6
Out-of-range Bit and Data Ready Output Clocks
In simultaneous output mode:
The (AOR/DRAN, AORN/DRA), (BOR/DRBN, BORN/DRB), (COR/DRCN, CORN/DRC) and
(DOR/DRDN, DORN/DRD) signals are used to add the out-of-range bit to the ADC output data.
On each port (A, B, C, D) the out-of-range bit (*OR/*ORN) indicates (logic “1”) when the analog input overrides the ADC 0.5 Vpp Full Scale voltage span, in case of underflow or overflow.
When the analog input exceeds the positive full scale, the 10 bit digital outputs data remain at high logical state, with (*OR,*ORN) at logical one. When the analog input falls below the negative full scale, the
10 bit digital outputs data remain at logical low state, with (*OR,*ORN) at logical one again.
In Staggered output mode:
The (AOR/DRAN, AORN/DRA), (BOR/DRBN, BORN/DRB), (COR/DRCN, CORN/DRC) and
(DOR/DRDN, DORN/DRD) signals will output the Data Ready Output Clock signal (one for each port), centered on the corresponding Data.
In this case, the overflow can be computed with all 10 digital output data detection at logic “1”, (binary mode), and underflow can be computed with all 10 digital output data detection at logic “0” (binary mode).
In 1:2 DMUX Ratio, DR/DRN and DRB/DRBN are identical.
In 1:4 DMUX Ratio, DR/DRN and DRD/DRDN are identical.
DMUX Power Reduction Mode (SLEEP)
The power reduction (SLEEP) mode allows the user to reduce the power consumption of the device
(DMUX part in Sleep mode). In this mode, the DMUX part consumption is reduced by 0.9W.
The Power reduction mode is active when SLEEP is low.
The device is in normal mode when SLEEP is high.
When the device is not used, minimal Power consumption is obtained with NAP (ADC part) and SLEEP
(DMUX part) modes used simultaneously.
DMUX Clock input Delay Cell (CLKDACTRL)
A tunable delay cell (CLKDACTRL) is in serial on the 11 Bit DMUX Data path to fine tune the data vs.
clock alignment at the interface between the ADC and the DMUX.
The delay is controlled via the CLKDACTRL pin. This delay can be tuned around default center value. It ranges from –100 ps to 100 ps for CLKDACTRL varying from V
CCD
/ 3 to 2 × V
CCD
/ 3.
This function results in a delayed internal clock signal.
This pin must always be biased and it is recommended to set CLKDACTRL at V
CCD
/3.
With the recommended value, it is normally not necessary to tune the CLKDACTRL voltage over the full specified clock rate, temperature range and power supply voltage range.
Figure 4-12. DMUX Clock Input Cell Block Diagram
CLK/CLKN
CLKDACTRL
2
Delay
(-100 to 100 ps)
2
Internal clock signal
1076D–BDC–10/12
47
e2v semiconductors SAS 2012
EV10AS150B
4.4.7
DMUX Built-In Test
The Built-In Self Test allows to test rapidly the DMUX block of the device.
It is activating via the BIST bit (active low) a checker board like pattern generator.
When this signal is left floating, the BIST is inactive.
When in BIST mode, a clock must be applied to the device, which can be set to 1:2 or 1:4 mode. The output clock mode DRTYPE can be either DR or DR/2.
In the BIST mode, all the bits are either all at low or high level (even and odd bits are in phase opposition) and transition every new cycle.
In order to have a deterministic output sequence, it is necessary to perform an asynchronous reset.
The output sequence is then:
Table 4-5.
BIST Output Sequence in 1:4 Mode
BIST sequence Binary (D9 …. D0)
DATA N on Port A
DATA N+1 on Port B
DATA N+2 on Port C
DATA N+3 on Port D
DATA N+4 on Port A
DATA N+5 on Port B
DATA N+6 on Port C
DATA N+7 on Port D
DATA N+8 on Port A
...
10 1010 1010
01 0101 0101
10 1010 1010
01 0101 0101
01 0101 0101
10 1010 1010
01 0101 0101
10 1010 1010
10 1010 1010
...
Table 4-6.
BIST Output Sequence in 1:2 Mode
BIST sequence Binary (D9 …. D0)
DATA N on Port A
DATA N+1 on Port B
DATA N+2 on Port A
DATA N+3 on Port B
DATA N+4 on Port A
...
10 1010 1010
01 0101 0101
01 0101 0101
10 1010 1010
10 1010 1010
...
Hexa (D9…D0)
0x2AA
0x155
0x2AA
0x155
0x155
0x2AA
0x155
0x2AA
0x2AA
...
Hexa (D9…D0)
0x2AA
0x155
0x155
0x2AA
0x2AA
...
Out-of-range Bit
0
1
1
0
0
1
0
...
1
0
Out-of-range Bit
0
1
0
...
1
0
48
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.5
ADC 3 Wire Serial Interface (ADC Controls)
4.5.1
3WSI Address and Data Format
The 3WSI is a synchronous write only serial interface made of 3 wires.
This 3 wire bus is activated with the SLDN control pin going low (please refer to “write timing” in next section).
The length of the word is 14 bits: 10 for the data and 4 for the address.
Bit d[9] of Data is MSB and bit a[3] of address is MSB.
The maximum serial logic clock frequency is 100MHz.
4.5.2
3WSI Timing
The 3WSI is a synchronous write only serial interface made of 3 wires:
• SCLK: serial clock input
• SLDN: serial load enable input
• SDATA: serial data input.
Please refer to Table 2-4 on page 7 for logical levels of SLCK, SLDN, SDATA and RESET.
The 3WSI gives a “write-only” access to up to 16 different internal registers of up to 10 bits each.
The input format is fixed with always 4 bits of register address followed by always 10 bits of data.
Address and data are entered MSB first.
The write procedure is fully synchronous with the clock rising edge of SCLK and described in the write chronogram
For proper initialization of 3WSI default settings, an asynchronous reset pulse on pin RESET is required.
The RESET pin combined with the SLDN pin must be used as a reset to program the chip to the “reset setting”.
• RESET high: no effect
• RESET low and SLDN low: programming of registers to default values
SLDN and SDATA are sampled on each rising edge of SCLK clock (clock cycle).
SLDN must be set at “1” when no write procedure is done.
A minimum of one clock rising edge (clock cycle) with SLDN at “1” is required for a correct start of the write procedure.
A write starts on the first clock cycle with SLDN at “0”. SLDN must stay at “0” during the complete write procedure.
In the first 4 clock cycles with SLDN at “0”, 4 bits of register address from MSB (a[3]) to LSB (a[0]) are entered.
In the next 10 clock cycles with SLDN at “0”, 10 bits of data from MSB (d[9]) to LSB (d[0]) are entered.
An additional clock cycle with SLDN at “0” is required for parallel transfer of the serial data d[9:0] in the register addressed with address a[3:0].
1076D–BDC–10/12
49
e2v semiconductors SAS 2012
EV10AS150B
This gives 15 clock cycles with SLDN at “0” for a normal write procedure.
A minimum of one clock cycle with SLDN returned at “1” is requested to end the write procedure, before the interface is ready for a new write procedure.
Any clock cycle with SLDN at “1” before the write procedure is completed, interrupts this procedure and no data transfer to internal registers is done.
Additional clock cycles with SLDN at “0” after the parallel data transfer to the register (done at 15 th
consecutive clock cycle with SLDN at “0”) do not affect the write procedure and are ignored.
It is possible to have only one clock cycle with SLDN at “1” between two successive write procedures.
10 bits of data must always be entered even if the internal addressed register has less than 10 bits.
Unused bits (usually MSB’s) are ignored. Bit signification and bit position for the internal registers are
Figure 4-13. 3WSI Timing Diagram
RESET
SLCK
SLDN
SDATA
Internal
Register
Value
Previous setting Default setting a[3] a[2] a[1] a[0]
11 12
13
14 15 d[2] d[1] d[0]
Table 4-7.
Name
Tsclk
Timings Related to Serial 3 Wire Serial Interface
Parameter
Period of SCLK
Min
10
Twsclk High or low time of SCLK
Tssldn Setup time of SLDN before rising edge of SCLK
Thsldn Hold time of SLDN after rising edge of SCLK
Tssdata Setup time of SDATA before rising edge of SCLK
Thsdata Hold time of SDATA after rising edge of SCLK
Twlreset Minimum low pulse width of RESET
Tdreset
Minimum delay between an edge of RESET and the rising edge of SCLK
2
4
5
4
2
5
10
New Setting
Typ Max
ns ns ns ns
Unit
ns ns ns ns
Note
50
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
4.5.3
3WSI Register Description
Note: Addresses and/or data bits not described herein are not accessible via Graphical User Interface
(GUI).
Table 4-8.
3WSI Settings
Address
0000
0001
0010
0011
0100
Description
Clock Duty Cycle
Sampling Delay
Adjust Coarse
Sampling Delay
Adjust Fine
Gain Adjust
Offset Adjust
register D4:D0 parameter value
Register D4:D0 parameter value
Register D9:D8 parameter value register D7:D0 parameter value register D7:D0 parameter value register D7:D0 parameter value register D9:D0
Default Value
10000
35/65%
10000
15 ps
10
60 ps
00000000
0 ps
10000000
0 dB
10000000
0mV
0000000000
Recommended
Value
10000
35/65%
11111
+30 ps
Max Value
11111
30/70%
11111
+30 ps
11
90 ps
11111111
Min Value
00000
40/60%
00000
0 ps
00
0 ps
00000000
SDA OFF
(See state register description below)
30 ps
11111111
+0.5 dB
11111111
0 ps
00000000
–0.5 dB
00000000
+20mV –20mV
See description hereafter 0110 State Register
Notes: 1. It is recommended to adjust clock duty cycle at 35/65%
2. It is recommended to adjust clock adjust to +30 ps
3. SDA Coarse & SDA fine: Total possible = total Coarse + total Fine = 90 + 30 ps = 120 ps
Step
0.3%
1 ps
30 ps
120 fs
0.004 dB
156 µV
Table 4-9.
State Register Description
Setting for Address :0110
Mode SDA OFF
Mode SDA ON
Mode binary output
Mode gray output
Two’s complement OFF
Two’s Complement ON
Data Ready Reset (DRR) inactive High
Data Ready Reset (DRR) inactive Low
NAP Mode OFF
NAP mode ON
X
X
X
X
D9
X
X
X
X
0
1
X
0
X
X
D8
X
X
X
1
X
X
X
X
X
X
D7
X
X
X
X
X
X
1
X
X
0
D6
X
X
X
X
X
X
0
X
1
0
D5
X
X
0
X
X
X
X
X
X
X
D4
X
X
X
X
X
X
X
X
X
X
D3
0
1
X
X
X
X
X
X
X
X
D2
X
X
X
X
X
X
X
X
X
X
D1
X
X
X
X
X
X
Comments:
1.
SDA Mode: to take into account the value of the Sampling Delay Adjust register, bit D3 of the state register should be asserted to 1. For applications requiring extremely low clock jitter and no interleaving of several ADCs, it is recommended to assert bits D3 of state register to 0.
2.
NAP mode of the ADC part reduces its power dissipation (in standby mode, both terminations of the differential data output buffers of the ADC are driven at same value).
X
X
X
X
D0
X
X
X
X
X
X
1076D–BDC–10/12
51
e2v semiconductors SAS 2012
EV10AS150B
4.5.4
4.5.5
4.5.6
NAP Mode
The NAP mode is controlled via the 3WSI serial interface, reducing the ADC power dissipation by 0.9W.
When the device is not used, minimal Power consumption for the combined ADC and 1:4 DMUX chips is obtained by turning on the NAP mode (on ADC part) and the SLEEP mode (DMUX) simultaneously.
Binary or Gray Output Mode
It is possible for the user to choose between the binary or Gray output data format. Gray coding may be used in order to reduce the effect of BER when occurring, by storing Gray output codes.
Digital Data format selection is made using the 3WSI (default selection is binary), programming the state register at address “0110”.
Binary two’s complement is also available asserting bit D6 at 1 of state register at address “0110”.(only taken in account if bit D5 is asserted at 0, i.e. in binary output mode).
Sampling Delay Adjust Function (SDA) – ADC Interleaving
This function is of most importance for applications based on time interleaving of multiple ADCs, in order to increase the actual sampling rate. In interleaved system the channels relative phasing has to be matched finely to avoid intermodulation spurs. The SDA function is monitored through the 3 Wire Serial
Interface (3WSI), for remote fine alignment of ADC Aperture delays.
The SDA fine tuning has to be done after proper alignment of ADC Gains and Offsets, also monitored by the 3WSI.
A control voltage is applied on the tunable delay line, through embedded control DACs: 2 Bit DAC
(coarse tuning) and 8 Bit DAC (fine tuning). The coarse 2 Bit DAC allows for coarse alignment of the
ADCs aperture delays, with a step (resolution) of 30 ps. The 8 Bit control DACs allows for fine phase alignment of ADCs, with a resolution of 30 ps/256 = 117 fs. This is more than enough for very fine phase alignment of multiple ADCs. Once tuned, the ADCs track each other versus temperature.
It is reminded that the ADC intrinsic jitter is 120 fs rms, (with SDA = OFF).
The sampling delay adjust enables a tuning of the aperture delay of each channel over a range of 120 ps with a first coarse tuning over a range of 90 ps and then a fine tuning on 8 bit over a range of 30 ps for better accuracy of the settings.
It is pointed out that the ADC intrinsic jitter is 120 fs rms if SDA is turned OFF, and becomes 150 fs rms if SDA is turned ON, and 170 fs rms if SDA is fully tuned. This is related to extra delay cells in the sampling clock path, which are bypassed if SDA is OFF. If the SDA is turned on, an amount of + 60 ps extra time delay adds onto the sampling clock path.
For reference, the measured SNR at Fin = 2500 MHz (–1 dBFS) is 50 dB with SDA = OFF, 49.2 dB with
SDA turned on, and 48.5 dB with SDA turned ON and fully tuned.
For optimum dynamic performance (low jitter), it is recommended to disable SDA (Mode SDA OFF).
52
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 4-14. Programming the SDA Tunable Delay Line Through the 3WSI
SDA Coarse tuning on 2 Bit: 30 ps step
SDA Fine tuning on 8 Bit : 117 fs resolution
4.5.7
SDA = OFF
(ADC Jitter
= 120 fs rms
60 ps
00
SDA = ON
(ADC Jitter
= 150 fs rms
30 ps
01
30 ps
10
SDA = ON
30 ps 30 ps
11
SDA = ON fully tuned
(coarse)
SDA = ON fully tuned
(coarse + fine)
ADC Jitter =170 fs
ADC Gain Control
The 3-Wire bus interface also allows for adjusting the gain by enabling a fine tuning by ±0.5 dB and an
8-bit resolution.
4.5.8
Offset Control
The 3WSI allows to control the offset of the ADC with a tuning range of ± 20mV, and an 8 bit resolution.
An accurate tuning of aperture delay, gain and offset allows to interleave two ADCs with minimum number of external analog components, thus providing an equivalent 5.2 Gsps ADC. Due to matching difficulties of the roll off of the input frequency response beyond 4 GHz, it is not recommended to use the
ADC in interleaved mode for coding of broadband signal beyond 4 GHz.
4.6
Die Junction Temperature Monitoring
A die junction temperature measurement setting is available, for maximum junction temperature monitoring (hot point measurement).
The measurement setup is described in the Figure hereafter:
Note: If not used, DIODE pin can be left floating.
Figure 4-15. ADC Diode for Die Junction Temperature Monitoring Setup
DIOD E
D+
1 nF
Temperature sensor
D-
AGND
Diode pin is provided so that the diode can be probed using standard temperature sensors.
For information, the forward voltage drop (V
DIODE
) across diode component, versus junction temperature,
(including chip parasitic resistance), is given below (I
DIODE
= 1 mA).
1076D–BDC–10/12
53
e2v semiconductors SAS 2012
EV10AS150B
Figure 4-16. Junction Temperature Versus Diode Voltage for I = 1 mA
Junction Temperature Versus Diode voltage for I = 1mA
890
880
870
860
850
840
830
820
810
800
790
780
770
760
750
740
730
720
710
700
980
970
960
950
940
930
920
910
900
-30 -20 -10 0
y = -1,195x + 911,855
10 20 30 40 50 60 70 80 90
Junction temperature (°C)
100 110 120 130 140
5.
Applications Information
5.1
Bypassing, Decoupling and Grounding
5.1.1
Decoupling capacitor
Globally, the ADC is decoupled following a four row decoupling capacitors strategy, providing adequate decoupling over frequency taking into account the external decoupling: at Evaluation Board and Package surrounding level, and the internal ADC decoupling (transparent for the end-user): at internal
Package decoupling and on chip decoupling level.
ADC external decoupling: typical values
1.
At Evaluation Board incoming level: (1 st
level): the capacitor should be chosen depending on
DC/DC regulator characteristics.
2.
Close to Package surrounding: 2 nd
level: 144 nF per analog supply (V
CCA5
& V
CCA3
), 192 nF for
V
CCD
and 288 nF for V
PLUSD
.
It is recommended to decouple all power supplies to ground as close as possible to the device balls.
ADC internal decoupling: typical values
3.
Inside Package: (3 rd
level): 100 pF per power supply plane to ground plane.
4.
On chip decoupling level: (4th level): nearly 300 pF per power supplies.
54
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 5-1.
EV10AS150B Decoupling Schema
Level 1: Power management
Capacitor: should be chosen depending on DC/DC regulator characteristics
EV10AS150B
Level 3: Package
Capacitor, ~100 pF
Level 4: On Chip
Capacitor, ~300 pF
5.1.2
Level 2: PCB board
Capacitor, 47 nF // 1nF
Power Supplies Decoupling Schema
Figure 5-2.
Power Supplies Decoupling Schema
EV10AS150B
V
CCA3
V
PLUSD
X3(min)
47 nF
1 nF
AGND
DGND
GND
V
CCA5
V
CCD
X3(min)
47 nF
1 nF
AGND
DGND
GND
GND = common ground plane
1 nF
1 nF
47 nF
X6(min)
GND
47 nF
X4(min)
GND
1076D–BDC–10/12
55
e2v semiconductors SAS 2012
EV10AS150B
5.1.3
5.1.4
V
CCA5
V
CCA5
V
CCA5
V
CCA3
V
CCA3
V
CCA3
V
PLUSD
V
PLUSD
V
PLUSD
V
PLUSD
V
PLUSD
V
PLUSD
V
CCD
V
CCD
V
CCD
V
CCD
Decoupling Capacitors Package Implementation
The table below indicates the pins to pins connected together with a pair of decoupling capacitance
(47 nF // 1 nF).
Table 5-1.
Supply
Decoupling Capacitors Implementation
Total decoupling per supply
144 nF
List of pins to be connected together with 47 nf // 1 nF
A27, A26, A24, B27, B26, B24, C27, C26, C24, D27, D26, D24,
E26, E24, F25
L26, L25, M27
144 nF
288 nF
192 nF
R21, T21, U21
A25, B25, B22, C25, C22, C20, D25, D22, D20, E25, E22, E20, F24,
F22, F20
R20, T20
K26, K25,L27, M26, M25, N26, N27
C19, C16, C15, C14, C13, D19, D13, E19, F19
T15, T14, T13, U15, U14, U13
C11, C9, C7, C6, C5, D11, D9, D7, D6, D5
T11, T9, T7, T6, U11, U9, U7, U6, U5
C4, E3, J4, J3
L4, L3, N4, N3, R4, R3, U4
C12, C10, C8, D16, D12, D10, D8, E17
D4, D3, E4, G4, G3
K4, K3, T4, T3, U3
R16, T17, T16, T12, T10, T8, T5, U12, U10, U8
Decoupling Capacitors PCB Implementation
Figure 5-3.
Decoupling Capacitors PCB Implementation
Package
Evaluation board
(Component side)
Balls
Component and back side traces
1 nF capacitors
Evaluation board
(back side)
Vias
47 nF capacitors
56
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
5.1.5
Power Plane (Analog and Digital)
For performance reasons it is always recommend to use separate supplies for digital and analog circuitry.
The digital supply should only be used for parts placed over the digital ground plane, i.e. all pure digital parts. The analog supply is used for all analog and mixed-signal parts.
Analog supplies
Î V
CCA3
and V
CCA5
Digital supplies
Î V
PLUSD
and V
CCD
V
PLUSD
supply must be isolated from analog supplies
Note: V
PLUSD
supply is dedicated to digital output buffer only
a). If this constraint is not respected, this will induce capacitance between the overlapping areas, which is likely to cause RF emissions to pass from one plane to another.
Figure 5-4.
Overlapping of Analog and Digital Power supplies planes
Analog
Digital
Analog VCCA3
Analog VCCA5
Digital VPLUSD
Digital VCCD
Ground
a) Bad
Analog VCCA3
Digital VPLUSD
Ground
Analog VCCA5
Digital VCCD
c) Optimum
Ground b) Correct
Since DGND and AGND are separated on package and chip, a continuous low impedance ground plane can be used for both Digital and Analog ground.
Moreover, power supply pins should be decoupled directly to the ground plane. The ceramic capacitor should be located as close as possible to the IC power pins.
The sampling clock generation circuitry should be considered like analog circuitry and also be grounded and heavily-decoupled to the common digital and analog ground plane. It also should be isolated from noisy digital circuits.
So, the ground plane not only acts a low impedance return path for decoupling high frequency currents but also minimizes RF emissions. Because of the shielding action of the ground plane, the circuit’s susceptibility to external RF is also reduced.
The isolation between FPGA Ground plane and ADC ground plane is recommended but not mandatory, depending on switching noise level which may be injected by FPGA.
1076D–BDC–10/12
57
e2v semiconductors SAS 2012
EV10AS150B
In case of separate ground between ADC and FGPA, they should be DC connected together by a 0 ohm resistor, making use of the parasitic (high resonant) inductance of the resistor element to reject the HF spikes.
This recommendation for optimum isolation must be followed carefully for optimum rejection of Fclk/4 clock spur in 1:4 DMUX mode.
Information regarding partitioning of package level:
DGND is only allocated to Output Buffers of ADC and digital section + Digital output buffers of 1:4
DMUX. AGND ground plane is allocated to ADC only, and needs to be isolated from 1:4 DMUX output buffers switching into 40 differential 100
Ω terminations. Switching Noise is mainly generated by back reflection effects due to L,C parasitics of 100
Ω terminations. This leads to switching noise into V
PLUSD and DGND which gives energy to Fclk/4 clock related spur if fed back to the ADC Ground.
58
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Figure 5-5.
Analog and Digital Partitioning of Power and Ground Planes for Optimum Isolation
Ground planes
Optional but not mandatory
R=0 Ohm
High resonant
MCM
ADC +
DMUX
Common Digital and
Analog ground plane
Output buffers
FPGA or µC
FPGA Digital
Ground
Power supplies planes
ADC
AGND
DMUX
DGND
MCM package internal ground planes partitioning:
AGND vs DGND
V
CCA5
(5.2V)
V
CCA3
(3.3V)
Analog Power
Supplies
Regulator 1 and 2
MCM
ADC +
DMUX
Output buffers
V
PLUSD
(2.5V)
V
CCD
(3.3V)
Digital Power
Supplies
Regulator 3
FPGA or µC
FPGA Digital
Power
Regulator 4
1076D–BDC–10/12
59
e2v semiconductors SAS 2012
EV10AS150B
5.2
Analog Input Implementation
5.2.1
ADC Analog Input Terminations
The DC analog common mode voltage is provided by ADC and the analog input signal must be used in
AC coupling mode (with a DC blocking capacitor).
The front-end input preamplifier in on chip terminated (100
Ω differential, 50Ω single-ended with accurate thin film TaN resistors with temperature coefficient close to 0
° C), driven by 50Ω controlled impedance lines (100
Ω differential) with the multilayer EBGA317 Package.
This allows flat input Voltage Standing Wave Ratio (VSWR) over frequency, (< 1.25:1 measured for packaged ADC, from DC up to 3 GHz (less than 1.3% of incoming power reflected from the ADC input, <
0.05 dB transmission loss), and < 1.5:1 from 3 GHz to 5 GHz Full power input bandwidth of the ADC
(less than 4% of power reflected from ADC input, < 0.17 dB transmission loss).
The low input VSWR together with the 5 GHz (–3 dB) full power input Bandwidth of the ADC, allows operation up to the 3 rd
Nyquist region (including L_Band and S_Band) with negligible carrier level fluctuations related to packaged ADC input impedance deviation from ideal Z = 50
Ω (including package parasitics).
The on-chip terminations (in-phase V
I N
and inverted phase V
I N N
) are actually based on a resistive voltage splitter of 55
Ω + 550Ω biased under +3.3V to ground, providing proper 50Ω termination
(550 // 55 = 50), together with 3V internal DC common mode biasing. Since the internal analog input DC common mode is +3V, the differential Analog inputs must be driven through high resonant (> 5 GHz) DC blocking capacitors.
Driving the EV10AS150B in single-ended for ADC Full Scale voltage span:
If single-ended driven, the applied in-phase (V
IN
) input voltage amplitude for ADC full-scale voltage span is 0.5V peak-to-peak, (i.e. –2 dBm Full Scale input power into 50
Ω on-chip termination.)
For proper symmetric input impedance matching, the inverted phase input (V
INN
) must be externally terminated by 50
Ω to Ground through a DC blocking chip capacitor. The inverted phase (V
INN on-chip 50
Ω terminated and biased the same way as the in-phase analogue input.
) is internally
Driving the EV10AS150B in differential for ADC Full Scale voltage span:
If entered in differential, each analog input (in-phase V
IN
and inverted phase V
INN
) shall be entered with
±0.125V peak = 0.25V peak-to-peak amplitude, to comply with 0.5V ADC Full Scale voltage span.
Since the 0.5Vpp Full Scale input voltage span is applied onto 100
Ω termination instead of 50Ω, the ADC
Full Scale differential input power into 100
Ω loading is –5 dBm instead of –2 dBm into 50Ω if singleended driven.
Low cost low profile sub-miniature unbalanced to balanced transformers (baluns) are commercially available, designed specifically for driving differential inputs and/or output locations for fast sampling ADCs in the GHz range. These baluns are providing 50
Ω impedance on unbalanced input, and 100Ω impedance on the balanced differential output port, making it easy to use on surface mount application boards to drive AC coupled differential Analog inputs and clock inputs.
For optimum SFDR performance, the maximum phase balance shall not exceed 12 degree and less than 1.5 dB for amplitude balance, over the entire band of operation. For example, for operation over the
2 nd
Nyquist region at 2.6 Gsps, baluns featuring less than 6 degree Phase Balance and 1 dB amplitude
Balance within a band of operation of 1.6 GHz to 3 GHz are available (for roughly 1 dB insertion loss).
Baluns covering a larger band of operation (e.g. 400 MHz to 3000 MHz) with 12 degrees max Phase unbalance and 1.5 dB Amplitude Balance are also convenient.
60
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Since the ADC tolerates up to 12 degrees of phase unbalance without impacting the dynamic performance (SFDR), it is not mandatory to use cascaded (double) transformer to improve phase balance characteristics. Double transformer implementations may degrade the inherent band flatness of the device with additional insertion loss.
Figure 5-6.
ADC Full Scale Single Ended vs. Differential Analog Input
Vin = 500 mVpp
ADC Full Scale
Single ended
+ 250 mVp
-
Vin = 500 mVpp
ADC Full Scale
Differential
VIN
+125 mV
VINN
VINN (DC = +3V)
+ 500 mVpp
+ 250 mV - 250 mV
5.2.2
Full Scale input power:
0.5 Vpp into 50
Ω = -2 dBm
Full Scale input power:
0.5 Vpp into 100
Ω = -5 dBm
Differential Versus Single-ended Input Dynamic Performance (SNR, SFDR, THD)
The EV10AS150B implements a front-end Track and Hold preamplifier which has been designed in order to be entered either in differential or single-ended, up to maximum operating sampling rate without significant loss in dynamic performance in either 1 st
and 2 nd
Nyquist for large signals close to ADC Full
Scale:
Same SNR values are measured in single-ended and differential, 1 st
and 2 nd
Nyquist (delta SNR <
0.2 dB).
Regarding linearity performance (Total Harmonic Distortion (THD) and Spurious Free Dynamic Range
(SFDR)):
• If differential driven, it is the odd Harmonics (mainly H3) which mainly dictates the ADC large signal linearity performance (SFDR and THD), whereas the even harmonics (H2, H4, …) only little contributes to THD.
• In single-ended driven, the even harmonics (H2, H4, …) will become significant with the same weighting as the odd Harmonics (H3,H5,…). The 2 nd
Harmonic will rise to the same level as 3 rd
Harmonic, mainly in the 2 nd
Nyquist.
The EV10AS150B front-end preamplifier has been boosted to maintain the 2 nd
Harmonic (H2) level below –60 dBFS in single-ended over 1 st
and 2 nd
Nyquist, even for large signals close to ADC Full
Scale. Since the 2 nd
harmonic weighting is not higher than the 3 rd
Harmonic level (H3), the SFDR performance will not be impacted.
1076D–BDC–10/12
61
e2v semiconductors SAS 2012
EV10AS150B
5.2.3
However, the THD will be somewhat impacted, since the harmonics are RSS summed since even and odd harmonics have now similar weightings in single-ended:
The THD is slightly impacted by –1 dB in the 1 st
Nyquist, and –2 dB in the 2 nd
Nyquist (Fin < 3000
MHz).
In the 3 rd
Nyquist for large signals close to ADC Full Scale, (e.g. Fin = 4000 MHz, –1 dBFS), the 2 nd
Harmonic impact on THD in single-ended will become much more important (H2 = + 6 dB versus differential mode):
Therefore in the 3 rd
Nyquist, it is highly recommended to drive the inputs in differential for optimum
Spurious Free Dynamic Range (SFDR) and optimum total harmonic distortion performance (THD).
Note: For THD computation, the 10 1 st
low rank harmonics are taken into account.
Differential versus Single-ended Input: Rejection for Fclk/4 Spurs
If the 1:4 DMUX Ratio is selected, the rejection of the Fclk/4 clock related spur becomes an issue since the clock spur is located at the middle of the Nyquist zone.
With 1:2 DMUX Ratio, the clock related spur is located at Fclk/2, and does not affect the center of the band of operation.
The rejection of Fclk/4 becomes 6 dB more efficient if the ADC is differentially driven. (A delta of –6 dB is measured in differential vs. single-ended for the Fclk/4 level).
On EV10AS150B evaluation board, the measured Fclk/4 level is:
• –72 dBFS in single-ended, corresponding to –74 dBm with –2 dBm Full Scale input power (0.5Vpp in
50
Ω).
• –78 dBFS in differential, corresponding to –82 dBm with –5 dBm Full Scale input power (0.5Vpp in
100
Ω).
The ADC has been carefully implemented at chip layout level and packaging level to ensure maximum electric isolation between the noisy Digital output buffers Section and Analog sections of the ADC. (Analog sections include the logic section and the timing circuitry of the ADC).
At application board level:
The absolute value for Fclk/4 in either single-ended or differential mode can be even further improved by improving the electric isolation of the +2.5V power supply: The +2.5V Power plane dedicated to the 40
Digital Output Buffers drives most of the Fclk/4 switching noise energy, and has to be isolated from the other remaining power supplies.
Refer to Figure 5-5 on page 59 for board layout recommendation for optimum isolation.
5.3
Clock Input Implementation
The EV10AS150B differential clock input buffer is based on fast regeneration amplifiers, in order to feature a square wave like sampling clock for the Track and Hold, featuring very fast internal slew-rates to ensure low internal sampling jitter.
The EV10AS150B differential clock input buffer is on-chip 100
Ω terminated (50Ω on each single-ended clock input).
The clock input buffer is biased in the same way as the differential analog input preamplifier:
The differential on-chip clock inputs terminations are based on resistive 55
Ω + 550Ω voltage splitter biased under +3.3V to Ground, providing 5
Ω impedance (550Ω // 55Ω = 50Ω) together with 3V internal
DC common mode biasing for both inputs.
62
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
5.3.1
Driving the EV10AS150B with a Sinewave Clock Input
The SNR rolloff in the high input frequency region (2 nd
and 3 rd
Nyquist) is dictated by sampling jitter:
The ADC sampling jitter is 120 fs rms, to be RSS summed with external sampling clock jitter:
To achieve optimum SNR in the 2 nd
& 3 rd
Nyquist (e.g; SNR > 48 dB at Fin = 3000 MHz, the external clock sinewave requirements are mainly two-fold:
The external sinewave clock signal phase noise and clock signal slew-rate.
The Single Side Band (SSB) phase noise floor of the reference sinewave clock signal shall not exceed
155 dBc/Hz (at 1 MHz from carrier).
The slew-rate of the 2.6 GHz clock input shall not be lower than 5 GV/s at 2.6 GHz frequency.
Accordingly, at 2.6 GHz sampling clock frequency, a sinewave clock input can be directly entered provided the sinewave amplitude is 0.632 Vpp minimum into the 100
Ω differential output to ensure 5 GV/s minimum slew-rate, and 155 dBc/Hz phase noise floor.
Sinewave Clock slew-rate effect on SNR:
Due to very fast internal slew-rates of regeneration stages, a sinewave clock can be entered without impacting the SNR so far the sinewave slew-rate is higher than 5 GV/s minimum (8 GV/s recommended).
These fast slewing front-end clock buffer regeneration stages provides lowest internal rms time domain jitter, as illustrated by the following relationship:
Rms(time jitter) = Rms(voltage noise) / Rms(slew-rate)
Assuming the ADC clock input is driven by a transformer (single-ended unbalanced 50
Ω to balanced
100
Ω differential), the minimum clock input power shall not be lower than –3 dBm into 50Ω input impedance = 0.45 Vpp at transformer input yielding 0.45 Vpp.SQRT(2) = 0.632 Vpp across 100
Ω differential outputs applied to the ADC.
A minimum swing of 0.632 Vpp at 2.6 GHz corresponds to a minimum slew-rate of nearly 5.2 GV/s.
Therefore, the 2.6 GHz external sinewave clock signal voltage amplitude applied across the 100
Ω termination shall not be lower than 0.632 Vpeak to peak for optimum SNR performance in the 1 st
and 2 nd
Nyquist.
For sinewave clock frequencies lower than 2.6 GHz, the sinewave amplitude shall be increased accordingly, to preserve the minimum slew-rate of 5 GV/s, without exceeding 2 Vpp maximum operating amplitude applied to the differential clock input. For example, with a 1.3 GHz sinewave clock, to keep the same slew-rate of 5 GV/s, the minimum voltage amplitude to be applied to the ADC clock input shall not be lower than 2 × 0.632 Vpp = 1.264 Vpp across the 100
Ω differential clock inputs, corresponding to + 3 dBm into 50
Ω transformer input (which is twice the clock input power of a 2.6 GHz sinewave input).
Typical recommended clock input power for a 2.6 GHz sinewave clock for optimum SNR over full temperature range shall be +1 dBm into 50
Ω transformer input impedance, which corresponds to 0.710
Vpp.SQRT(2) = 1 Vpp across 100
Ω across differential outputs applied to the ADC clock inputs. This corresponds to a slew-rate of 8.16 GV/s at 2.6 GHz.
Therefore, typical recommended slew-rates to ensure flat SNR over operating temperature range shall be in the range of 8 GV/s.
Maximum operating clock input power shall not exceed + 7 dBm, which corresponds to 2 Vpp on differential 100
Ω clock input. This corresponds to slew-rates of 15.7 GV/s.
1076D–BDC–10/12
63
e2v semiconductors SAS 2012
EV10AS150B
Maximum ratings for differential clock input is 3 Vpp which corresponds nearly to +11 dBm maximum clock input power.
Sinewave Clock Phase noise vs. time domain jitter effect on SNR rolloff at high input frequencies:
The SNR due to ADC intrinsic clock jitter contribution only (i.e. 120 fs rms), excluding external clock source jitter (jitter free external clock source), is 50 dB at Fin = 2500 MHz, –1 dBFS.
Taking into account the external clock source time domain jitter contribution on SNR performance, the
ADC intrinsic clock jitter of 120 fs rms has to be RSS summed to the external rms clock jitter, yielding to total sampling clock Jitter:
For example with a 100 fs rms external clock source jitter, the total sampling clock jitter will be
SQRT(120
2
+ 100
2
) = 156 fs rms.
With a 120 fs rms clock source jitter, the total jitter will be 170 fs rms:
For the EV10AS150B, the measured SNR at Fin = 2500 MHz for –1 dBFS analog input amplitude is
50 dB with a jitter free external clock source, (taking only into account the ADC intrinsic jitter of 120 fs rms). With 120 fs rms external clock source time domain jitter RSS summed with the 120 fs rms intrinsic
ADC clock jitter, (leading to 170 fs rms total jitter), the measured SNR becomes 48.5 dB.
Relationship between sinewave clock source phase noise and time domain jitter:
The 100 fs rms external time domain jitter corresponds approximately to an external sinewave clock source with (white) phase noise floor spectral density of 155 dBc / Hz, integrated from 1 MHz up to
5.5 GHz:
The 5.5 GHz upper limit of integration corresponds to the ADC clock input bandwidth. The 1 MHz lower limit of integration corresponds to the flat section of white phase noise, excluding 1/f and close in phase noise contribution for 1 st
order calculations.
The integration bandwidth for the phase noise floor is taken from 1 MHz up to 5.5 GHz ADC clock input bandwidth, leading to:
Integrated SSB phase noise floor spectral density: 155 dBc/Hz, 1 MHz to 5.5 GHz:
–155 dB + 10.log(5.5.10
9
–1.10
6
) = –155 + 97.4 dB = –57.6 dB.
This corresponds to the Single Side Band (SSB) phase noise power spectral density 10.log(Rad
2
/Hz) expressed in dBc/Hz, integrated over nearly 5.5 GHz. The total integrated Double Side Band phase noise power expressed in radian (rms) is:
SQRT (2.10
–57.6 / 10
) = 1,86.10
–03
radians (rms).
In time domain, the rms jitter is obtained by dividing the total phase noise power (radians rms) by the carrier frequency.
For example, with a 2.6 GHz sinewave clock carrier, the time domain jitter will be 1,86.10
–03
radians
(rms) / 2.
π.2,6 GHz = ~ 114 fs rms
Therefore for a given constant phase noise floor, the external clock source time domain jitter will be slightly better when increasing clock frequency (due to faster signal slew-rate).
64
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
The 1/f phase noise contribution from 10 KHz to 1 MHz can be neglected so far the mean noise level in this region is below 130 dBc / Hz:
Integrated SSB mean phase noise spectral density: 130 dBc/Hz, 10 KHz to 1 MHz:
–130 dB + 10.log(10
6
–10
4
) = –130 + 60 dB = –70 dB
Integrated double SSB phase noise power in radians (rms):
SQRT(2.10
–70 / 10
) = 0,447.10
–03
radians (rms)
To be compared with 1,86.10
–03
radians (rms) 155 dBc/Hz phase noise integrated over 5.5 GHz. The total RSS summed phase noise from 10KHz to 5.5 GHz is:
SQRT ((1,86.10
–03
radians)2 + (0,447.10
–03
radians)
2
) = 1,91.10
–03
radians
Therefore the close-in phase contribution remains negligible. The close-in phase noise (1/f2 and 1/f3) contributions can also be neglected since this type of phase noise is related only to long term jitter drift effects which are not of concern here regarding the registration length of the Digital Data.
For example, considering the registration length time window for a 32K FFT at 2.6 Gsps:
32768x385 ps = 12.616 µs, corresponding to 79.3 KHz.
As a result, the close-in phase noise below 70 KHz has not to be taken into account for the integration.
As a conclusion, considering a phase noise floor between 1 MHz and 5.5 GHz, any clock source phase noise floor better than 155 dBc/Hz will improve somewhat the SNR:
For example, a 160 dBc/Hz phase noise clock source will feature only 67 fs rms time domain jitter for a
2.6 GHz carrier, and a 165 dBc/Hz clock source will feature only 38 fs rms.
In these cases, the external clock source contribution to total jitter will be negligible, compared to the 120 fs rms intrinsic jitter of the ADC, and SNR performance in the 2 nd
and 3 rd
Nyquist will be improved.
Better SNR performance over frequency will then necessarily go through the improvement of intrinsic
ADC jitter (< 50 fs rms for instance).
On the other hand, using a lower performance phase noise floor sinewave source will cause extra rolloff of the SNR in 2 nd
Nyquist region.
Example: 150 dBc/Hz phase noise will feature a time domain jitter of 188 fs rms time domain jitter for a
2.6 GHz carrier, and a 145 dBc/Hz phase noise will feature 375 fs rms time domain jitter at 2.6 GHz.
Therefore the impact on SNR (and therefore ENOB performance) for very high analog input frequencies is no more negligible against ADC intrinsic jitter, and the SNR performance will be impacted in these cases.
As a conclusion, the recommended phase noise floor shall be at least 155 dBc / Hz for optimum SNR performance over multiple Nyquist zones. Any improvement in phase noise floor will help to save a few dBs especially in the 2 nd
and 3 rd
Nyquist zones.
Note: If a fixed clock source can be used, (i.e.: non swept clock), the sinewave clock source can be band-pass filtered to further improve the phase noise floor.
1076D–BDC–10/12
65
e2v semiconductors SAS 2012
EV10AS150B
5.3.2
Driving the EV10AS150B with a Square Wave Differential Cock Input
If the clock input signal is a square wave, the incoming signal slew-rate becomes independent to signal amplitude and frequency:
So far the slew-rate of the square clock signals is constant and controlled (> 5 GV/s), the SNR performance of the ADC will be independent to clock signal frequency and amplitude.
Consequently with square wave clock signals, much lower signal amplitudes and frequencies can be entered to the ADC as with sinewaves. With square waves, the SNR performance will be mainly dependant on time domain jitter, which shall remain below 100 fs rms, to ensure optimum SNR performance especially in the 2 nd
and 3 rd
Nyquist as detailed in the previous sections.
With a square wave clock input, a minimum amplitude of ±125 mV peak = 250 mVpp can be entered without impact on linearity (SFDR and THD) and SNR. LVDS logic compatible and LVPECL logic compatible clock inputs can therefore be used to drive the EV10AS150B ADC clock input, so far the time domain jitter of the square wave is < 100 fs rms, and slew-rate is > 5GV/s. So it is not actually the rise/fall time of the square wave clock signals which are of interest, but the actual Slew-rate of the clock edges.
For LVPECL square wave clock input featuring a typical rise time (20% to 80%) of 75 ps for 800 mV voltage swing, the corresponding signal slew-rate is roughly 500 mV/75 ps = 6.6 GV/s, which is convenient.
In addition to that, the time domain jitter shall be < 100 fs rms. (e.g.: 70 fs rms).
LVDS and/or LVPECL compatible low jitter regeneration signals Buffers are commercially available, featuring less than 100 fs rms time domain jitter.
Of course, the sinewave to square wave regeneration buffers shall be driven by a low phase noise sinewave reference clock source (> 155 dBc/Hz), as stated in the previous sections:
The additional jitter due to the square wave regeneration circuitry has then to be RSS summed to the driving sinewave signal time domain jitter (e.g.: 100 fs rms time domain jitter which are achieved by a
155 dBc/Hz sinewave clock source will have to be RSS summed with a 70 fs rms square wave jitter, yielding to SQRT(100
2
+ 70
2
) = 122 fs rms total external jitter (sinewave & square wave)).
In these conditions, the SNR performance of the ADC will be nearly 49 dB at Fin = 3000 MHz, (–1 dBFS at 2.6 Gsps). It is reminded that an SNR of 50 dB at (3000 MHz, –1 dBFS) is achieved if the external clock source jitter contribution can be neglected, only taking into account the 120 fs rms intrinsic sampling clock jitter of the ADC, assuming ideal (jitter-free) external clock sources with ultra fast slew-rates
(> 10 GV/s).
Note 1: ADC Minimum sampling clock frequency:
The square wave clock input frequency shall not be lower than 500 MHz minimum, to avoid SNR and
THD rolloff due to front-end Track and Hold droop rate. Recommended minimum value for clock frequency over temperature is 1 GHz to avoid any roll-off in dynamic performance.
Note 2: ADC SNR performance at low analog input frequencies:
The sampling clock jitter effect on SNR is negligible at low input frequencies (oversampling conditions).
The SNR performance for low input frequencies is mainly dictated by the ADC front-end Track and Hold input referred thermal noise integrated over the 5 GHz Analog input Bandwidth (times
π/2 assuming 1 st order rolloff).
The measured SNR in oversampling conditions is 54 dB.
The ADC deviation from ideal quantification noise (Differential Non Linearity: DNL) on SNR performance is negligible compared to input referred thermal noise:
The measured Differential Non Linearity (DNL) is < 0.5 LSB peak (= 0.2
LSB
rms) at 2.6 Gsps 1 st
and 2 nd
Nyquist, which is the same order of magnitude as quantification noise Q/SQRT(12).
66
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Note 3: Effect of 3WSI SDA (Sampling Delay Adjust) tunable delay line on ADC intrinsic jitter
The previous assumptions are made with ADC intrinsic clock jitter = 120 fs rms, which is assuming the sampling clock adjust is de-activated (SDA OFF), which is the case by default setting at ADC Reset.
If multiple ADCs need to be interleaved, the respective Aperture delays of the ADCs need to be carefully aligned, and fine tuned with the “SDA” function monitored by the 3 Wire Serial Interface (3WSI), by activating the SDA function (SDA ON).
Turning “ON” the SDA will activate an internal tunable delay line in serial with the sampling clock path,
(which is bypassed if SDA is turned OFF). By turning ON the SDA, the ADC intrinsic jitter becomes 150 fs rms (with minimum delay), since some additional extra delay cells will contribute to jitter, which will slightly impact the SNR performance by 1 dB at Fin = 2500 MHz. (50 dB with SDA turned OFF, and 49 dB if SDA is turned ON). If SDA is turned ON and fully tuned (maximum delay), the ADC intrinsic jitter becomes 170 fs rms.
Refer to chapter 4.6.6. for more information about Sampling Delay Adjust function.
5.4
LVDS Output Implementation
Output Data, Output Clock (Data Ready) and out-of-range bit are LVDS signals that needs to be 100
Ω differentially terminated.
Output data:
• In-phase (Ai) and inverted phase (AiN) digital outputs on DEMUX Port A (with i = 0…9)
• In-phase (Bi) and inverted phase (BiN) digital outputs on DEMUX Port B (with i = 0…9)
• In-phase (Ci) and inverted phase (CiN) digital outputs on DEMUX Port C (with i = 0…9)
• In-phase (Di) and inverted phase (DiN) digital outputs on DEMUX Port D (with i = 0…9)
A0, B0, C0, D0 are the LSB, A9, B9, C9, D9 are the MSB
Output Clock (Data Ready)
• In-phase DR and inverted phase DRN.
• In-phase DRA and inverted phase DRAN on DEMUX Port A used in staggered mode.
• In-phase DRB and inverted phase DRBN on DEMUX Port B used in staggered mode.
• In-phase DRC and inverted phase DRCN on DEMUX Port C used in staggered mode.
• In-phase DRD and inverted phase DRDN on DEMUX Port D used in staggered mode.
Out of Range
• In-phase AOR and inverted phase AORN on DEMUX Port A used in simultaneous mode.
• In-phase BOR and inverted phase BORN on DEMUX Port B used in simultaneous mode.
• In-phase COR and inverted phase CORN on DEMUX Port C used in simultaneous mode.
• In-phase DOR and inverted phase DORN on DEMUX Port D used in simultaneous mode.
Each of these outputs should be terminated by 100
Ω differential resistor placed as close as possible to the differential receiver (inside receiver is even better).
In 1:2 DMUX Ratio the unused Output Data and out-of-range bit (Port C and D) could be leave floating without 100
Ω differential resistor.
1076D–BDC–10/12
67
e2v semiconductors SAS 2012
EV10AS150B
Figure 5-7.
Differential Digital Outputs Terminations (100
Ω LVDS)
EV10AS150B
Output Data
Z0 = 50
Ω
Data Out
Differential Output buffers
Data OutN
Z0 = 50
Ω
100
Ω Termination
To Load
5.5
DRR and ASYNCRST Implementation
FPGA outputs can be directly connected to ADC DRR and ASYNCRST inputs.
Figure 5-8.
DRR and ASYNCRST Signals Implementation
FPGA
EV10AS150B
DRR
D0
D1
Note: Take care of signal integrity to avoid parasitic transitions.
ASYNCRST
68
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
6.
Package Description
6.1
Package Outline
Figure 6-1.
EBGA 317 Package Outline
8
A1 Corner
Ni Plated
A4
DETAIL A
SIDE VIEW
DETAIL A
D
TOP VIEW
A
E
B e
A2
A1
DETAIL B
E1
27
26 24
25
22
23
20
21
18
19 17
16 14
15
12
13
10
11 9
8
7
6
5
4
3
2
1
A
C
E
G
B
D
F
J
H
K
L
N
R
M
P
U
W
T
V
REF.
DIMENSIONAL REFERENCES
MIN.
NOM.
A
A1
A2
A4
D
D1
E
E1 b
1.25
0.50
0.75
0.10
34.80
24.80
0.60
1.45
0.60
0.85
35.00
33.02 (BSC.)
25.00
22.86 (BSC.)
MD
ME
N bbb
0.75
27
19
317 ddd e
Q
0.25
1.27 TYP.
MAX.
1.65
0.70
0.95
35.20
25.20
0.90
0.25
0.20
D1
BOTTOM VIEW
Q
A
DETAIL B e bbb C
Q b (Nx PLACES)
4
0.30
M
0.15
M
C
C
A B
NOTES:
1. ALL DIMENSIONS ARE IN MILLIMETERS.
2. "e" REPRESENTS THE BASIC SOLDER BALL GRID PITCH.
3. "MD" AND "ME" REPRESENTS THE BASIC SOLDER BALL MATRIX SIZE IN
"D" AND "E" DIRECTION RESPECTIVELY, AND SYMBOL "N" IS THE
MAXIMUM ALLOWABLE NUMBER OF BALLS AFTER DEPOPULATING.
4. DIMENSION "b" IS MEASURED AT THE MAXIMUM SOLDER BALL DIAMETER
5.
PARALLEL TO PRIMARY DATUM C .
DIMENSION "ddd" IS MEASURED PARALLEL TO PRIMARY DATUM C .
6. PRIMARY DATUM C AND SEATING PLANE ARE DEFINED BY THE
SPHERICAL CROWNS OF THE SOLDER BALLS.
7. PACKAGE SURFACE SHALL BE Ni PLATED.
8. BLACK SPOT FOR PIN 1 IDENTIFICATION.
9. ENCAPSULANT SIZE MAY VARY WITH DIE SIZE.
10. "A4" IS MEASURED AT THE EDGE OF ENCAPSULANT TO THE INNER EDGE
OF BALL PAD.
11. DIMENSIONING AND TOLERANCING PER ASME Y14.5 1994
12. THE OUTLINE DIMENSION IS REFERENCE TO JEDEC MS034.
13. FOR QUALIFICATION PURPOSE ONLY.
C
SEATING PLANE 6 ddd C
5
1076D–BDC–10/12
69
e2v semiconductors SAS 2012
EV10AS150B
6.2
Land Pattern Recommendation
Figure 6-2.
EBGA 317 Land Pattern Recommendation
TOP VIEW
A
B
D e
BOTTOM VIEW
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12 10 8
11 9 7
6
5
4
3
2
1
N
P
R
T
U
W
V
A
B
C
D
E
G
F
H
J
K
L
M
E
C
Note: Not solder PADS
LAND PATTERN
RECOMMENDATIONS
E
A B C D E e b
25.00
35.00
0.85
22.86
33.02
1.27
0.60
b
D e
All dimensions are in millimeters
70
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
6.3
Thermal Characteristics
As there is no JEDEC standard definition for the thermal resistance applied to a multi-die device, only the thermal resistance for each die (ADC block powered ON only or DMUX block powered ON only) is provided.
All results were computed with ANSYS thermal simulation tool and with the following assumptions:
• Half geometry simulation
• ADC heating zone = 1.9 × 1.9 mm
2
• MUX heating 4.0 × 4.0 mm
2
• No air, pure conduction, no radiation
• ADC block ~ 4.0W
• DMUX block ~ 2.2W
6.3.1
Table 6-1.
Thermal Resistor
ADC block ON only
DMUX block ON only
Rth junction to bottom of balls
7
° C/W
3.9
° C/W
Note: 1. Assumed board size = 53 × 43 mm².
Rth junction to
Top of Case
4.1
° C/W
1.5
° C/W
Rth junction to
8
° C/W
4.9
° C/W
Rth junction to ambient
17.1
° C/W
13.9
° C/W
Thermal Management Recommendations
In still air and 25
° C ambient temperature conditions, the maximum temperature of 106° C + 25° C
= 131
° C is reached for the ADC block. It is consequently necessary to manage heat from the
EV10AS150 very carefully to avoid permanent damages of the device due to over temperature operation.
In no air cooling conditions, an external heatsink must be placed on top of package. An electrical isolation may be necessary as the TOP of the package is at ground potential.
It is advised to use an external heatsink with intrinsic thermal resistance better than 4
° C/Watt when using air at room temperature 20–25
° C. At 60° C, the external heatsink should have an intrinsic thermal resistance better than 3
° C/Watt.
Figure 6-3.
EV10AS150B-EB Evaluation Board Heat Sink Outlines
35 x 35 x 18 heat sink
Thermal conductive foil both sides
Note: ADC performance is improved at higher temperature.
1076D–BDC–10/12
71
e2v semiconductors SAS 2012
EV10AS150B
6.4
Moisture Characteristics
This device is sensitive to the moisture (MSL3 according to JEDEC standard). Its shelf life in sealed bag is 12 months at < 40
° C and < 90% relative humidity (RH).
Once the bag is opened, devices that will be subjected to infrared reflow, vapor-phase reflow, or equivalent processing (peak package body temp. 260
° C for ROHS versions and 220° C for non ROHS versions) must be:
• Mounted within 168 hours at factory conditions of
≤ 30° C/60% RH, or
• Stored at
≤ 20% RH
Before mounting, devices will require baking if the humidity indicator is > 20% when read at 23
° C ± 5° C.
If baking is required, devices may be baked for:
• 192 hours at 40
° C + 5° C/–0° C and < 5% RH for low-temperature device containers, or
• 24 hours at 125
° C ± 5° C for high-temperature device containers
7.
Ordering Information
Table 7-1.
Ordering Information
Part Number
EVX10AS150BTP
Package
EBGA317
EVX10AS150BTPY EBGA317 RoHS
EV10AS150BCTP EBGA317
EV10AS150BVTP
EV10AS150BCTPY
EV10AS150BVTPY
EV10AS150BTP-EB
EBGA317
EBGA317
RoHS
EBGA317
RoHS
EBGA317
Temperature
Ambient
Ambient
Commercial grade
0
° C < T amb
; T
J
< 90
° C
Industrial grade
–40
° C < T amb
; T
J
< 110
° C
Commercial grade
0
° C < T amb
; T
J
< 90
° C
Industrial grade
–40
° C < T amb
; T
J
< 110
° C
Ambient
Screening
Prototype
Prototype
Standard
Standard
Standard
Standard
Prototype
Comments
Production version
Production version
Production version
Production version
Evaluation Board with soldered
EVX10AS150BTP prototype in
EBGA317 package
72
1076D–BDC–10/12 e2v semiconductors SAS 2012
EV10AS150B
Table of Contents
5 Applications Information ...................................................................... 54
1076D–BDC–10/12
i
e2v semiconductors SAS 2012
EV10AS150B
ii
1076D–BDC–10/12 e2v semiconductors SAS 2012
How to reach us
Home page: www.e2v.com
Sales offices:
Europe Regional sales office e2v ltd
106 Waterhouse Lane
Chelmsford Essex CM1 2QU
England
Tel: +44 (0)1245 493493
Fax: +44 (0)1245 492492 mailto: [email protected]
e2v sas
16 Burospace
F-91572 Bièvres Cedex
France
Tel: +33 (0) 16019 5500
Fax: +33 (0) 16019 5529 mailto: [email protected]
e2v gmbh
Industriestraße 29
82194 Gröbenzell
Germany
Tel: +49 (0) 8142 41057-0
Fax: +49 (0) 8142 284547 mailto: [email protected]
Americas e2v inc
520 White Plains Road
Suite 450 Tarrytown, NY 10591
USA
Tel: +1 (914) 592 6050 or 1-800-342-5338,
Fax: +1 (914) 592-5148 mailto: [email protected]
Asia Pacific e2v ltd
11/F.,
Onfem Tower,
29 Wyndham Street,
Central, Hong Kong
Tel: +852 3679 364 8/9
Fax: +852 3583 1084 mailto: [email protected]
Product Contact:
e2v
Avenue de Rochepleine
BP 123 - 38521 Saint-Egrève Cedex
France
Tel: +33 (0)4 76 58 30 00
Hotline: mailto: [email protected]
Whilst e2v has taken care to ensure the accuracy of the information contained herein it accepts no responsibility for the consequences of any use thereof and also reserves the right to change the specification of goods without notice. e2v accepts no liability beyond that set out in its standard conditions of sale in respect of infringement of third party patents arising from the use of tubes or other devices in accordance with information contained herein.
e2v semiconductors SAS 2012
1076D–BDC–10/12
EV10AS150B
iv
1076D–BDC–10/12 e2v semiconductors SAS 2012
advertisement
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project