MT-097: Dealing with High Speed Logic

MT-097: Dealing with High Speed Logic
MT-097
TUTORIAL
Dealing with High-Speed Logic
WHEN ARE TRANSMISSION LINE TECHNIQUES NEEDED?
Much has been written about terminating PCB traces in their characteristic impedance, to avoid
signal reflections. Tutorial MT-094 presents the basic design equations for microstrip and
stripline transmission lines. However, it may not be clear when transmission line techniques are
appropriate.
A good guideline to determine when the transmission line approach is necessary is as follows:
Terminate the transmission line in its characteristic impedance when the one-way propagation
delay of the PCB track is equal to or greater than one-half the applied signal rise/fall time
(whichever edge is faster).
For example, a 2 inch microstrip line over an Er = 4.0 dielectric would have a delay of ~270 ps.
Using the above rule strictly, termination would be appropriate whenever the signal rise time is <
~500 ps.
A more conservative rule is to use a 2 inch (PCB track length)/nanosecond (rise/fall time) rule. If
the signal trace exceeds this trace-length/speed criterion, then termination should be used.
For example, PCB tracks for high-speed logic with rise/fall time of 5 ns should be terminated in
their characteristic impedance if the track length is equal to or greater than 10 inches (where
measured length includes meanders).
As an example of what can be expected today in modern systems, Figure 1 shows typical rise/fall
times for several logic families including the SHARC DSPs operating on +3.3 V supplies. As
would be expected, the rise/fall times are a function of load capacitance.
In the analog domain, it is important to note that this same 2 inch/nanosecond guideline should
also be used with op amps and other circuits, to determine the need for transmission line
techniques. For instance, if an amplifier must output a maximum frequency of fmax, then the
equivalent risetime tr is related to this fmax. This limiting risetime, tr, can be calculated as:
tr = 0.35/fmax
Eq. 1
The maximum PCB track length is then calculated by multiplying tr by 2 inch/nanosecond. For
example, a maximum frequency of 100 MHz corresponds to a risetime of 3.5 ns, so a 7-inch or
more track carrying this signal should be treated as a transmission line.
Rev.0, 01/09, WK
Page 1 of 9
MT-097
‹ GaAs: 0.1ns
‹ ECL: 0.75ns
‹ ADI SHARC DSPs: 0.5 ns to 1 ns (Operating on +3.3V Supply)
ADSP-21060L
SHARC:
Figure 1: Typical DSP Output Rise Times and Fall Times
REDUCING THE EFFECT OF FAST LOGIC ON ANALOG CIRCUITS
The best ways to keep sensitive analog circuits from being affected by fast logic are to physically
separate the two by the PCB layout, and to use no faster logic family than is dictated by system
requirements. In some cases, this may require the use of several logic families in a system. An
alternative is to use series resistance or ferrite beads to slow down the logic transitions where
highest speed isn't required. Figure 2 shows two methods.
LOGIC
GATE
R
< 2 inches
LOGIC
GATE
CIN
Risetime = 2.2 R·CIN
> 2 inches
LOGIC
GATE
R
LOGIC
GATE
CIN
C
Risetime = 2.2 R·(C + CIN)
Figure 2: Damping Resistors Slow Down Fast Logic Edges to
Minimize EMI/RFI Problems
Page 2 of 9
MT-097
In the first, the series resistance and the input capacitance of the gate form a lowpass filter.
Typical CMOS input capacitance is 5 pF to 10 pF. Locate the series resistor close to the driving
gate. The resistor minimizes transient currents and may eliminate the necessity of using
transmission line techniques. The value of the resistor should be chosen such that the rise and fall
times at the receiving gate are fast enough to meet system requirement, but no faster. Also, make
sure that the resistor is not so large that the logic levels at the receiver are out of specification
because of the voltage drop caused by the source and sink current which flow through the
resistor. The second method is suitable for longer distances (>2 inches), where additional
capacitance is added to slow down the edge speed. Notice that either one of these techniques
increases delay and increases the rise/fall time of the original signal. This must be considered
with respect to the overall timing budget, and the additional delay may not be acceptable.
Figure 3 shows a situation where several DSPs must connect to a single point, as would be the
case when using read or write strobes bidirectionally connected from several DSPs. Small
damping resistors shown in Figure 3A can minimize ringing provided the length of separation is
less than about 2 inches. This method will also increase rise/fall times and propagation delay. If
two groups of processors must be connected, a single resistor between the pairs of processors as
shown in Figure 3B can serve to damp out ringing.
A
STAR CONNECTION
DAMPING RESISTORS
SHARC
DSP
USE FOR RD, WR
STROBES
SHARC
DSP
<2"
10Ω
EACH
SHARC
DSP
SHARC
DSP
NOTE: THESE TECHNIQUES
INCREASE RISE/FALL TIMES
AND PROPAGATION DELAY
SHARC
DSP
B
SINGLE DAMPLING
RESISTOR BETWEEN
PROCESSOR GROUPS
<2"
SHARC
DSP
20Ω
SHARC
DSP
Figure 3: Series Damping Resistors for
High Speed DSP Interconnections
Page 3 of 9
SHARC
DSP
MT-097
END TERMINATION AND SOURCE TERMINATION
The only way to preserve 1 ns or less rise/fall times over distances greater than about 2 inches
without ringing is to use transmission line techniques. Figure 4 shows two popular methods of
termination: end termination, and source termination. The end termination method (Figure 4A)
terminates the cable at its terminating point in the characteristic impedance of the microstrip
transmission line. Although higher impedances can be used, 50 Ω is popular because it
minimizes the effects of the termination impedance mismatch due to the input capacitance of the
terminating gate (usually 5 pF to 10 pF).
TYPICAL DRIVERS:
„ 74FCT3807/A (IDT)
„ 74ACTQ240 (Fairchild)
A
ZO = 50Ω
+1.4V
+3.3V
+5.0V
120Ω
30mW
180Ω
72mW
+1.4V
91Ω
22mW
GROUND PLANE
B
68Ω
29mW
END TERMINATION
ZO ≈ 10Ω
SOURCE TERMINATION
39Ω
RULE OF THUMB:
ZO = 50Ω
USE TRANSMISSION LINE IF DISTANCE IS
MORE THAN 2"/ns OF LOGIC RISE/FALL TIME
50Ω PC BOARD TRANSMISSION LINE DELAY ≈ 1ns / 7"
Figure 4: Termination Techniques for Controlled
Impedance Microstrip Transmission Lines
In Figure 4A, the cable is terminated in a Thevenin impedance of 50 Ω terminated to +1.4 V (the
midpoint of the input logic threshold of 0.8 V and 2.0 V). This requires two resistors (91 Ω and
120 Ω), which add about 50 mW to the total quiescent power dissipation to the circuit. Figure 4A
also shows the resistor values for terminating with a +5 V supply (68 Ω and 180 Ω). Note that
3.3 V logic is much more desirable in line driver applications because of its symmetrical voltage
swing, faster speed, and lower power. Drivers are available with less than 0.5 ns time skew,
source and sink current capability greater than 25 mA, and rise/fall times of about 1 ns.
Switching noise generated by 3.3 V logic is generally less than 5 V logic because of the reduced
signal swings and lower transient currents.
The source termination method, shown in Figure 4B, absorbs the reflected waveform with an
impedance equal to that of the transmission line. This requires about 39 Ω in series with the
internal output impedance of the driver, which is generally about 10 Ω. This technique requires
that the end of the transmission line be terminated in an open circuit, therefore no additional
fanout is allowed. The source termination method adds no additional quiescent power dissipation
to the circuit.
Page 4 of 9
MT-097
HIGH SPEED CLOCK DISTRIBUTION
Figure 5 shows a method for distributing a high speed clock to several devices. The problem
with this approach is that there is a small amount of time skew between the clocks because of the
propagation delay of the microstrip line (approximately 1 ns /7"). This time skew may be critical
in some applications. It is important to keep the stub length to each device less than 0.5" in order
to prevent mismatches along the transmission line.
+3.3V
TRANSMISSION LINE ZO = 50Ω
CLOCK
+1.4V
120Ω
30mW
91Ω
22mW
SHARC
DSP
SHARC
DSP
SHARC
DSP
50Ω PC BOARD TRANSMISSION LINE DELAY ≈ 1ns / 7"
NOTE: KEEP STUB LENGTH < 0.5"
NOT RECOMMENDED FOR SYNCHRONIZED SHARC OPERATION!
Figure 5: Clock Distribution Using End-of-Line Termination
The clock distribution method shown in Figure 6 minimizes the clock skew to the receiving
devices by using source terminations and making certain the length of each microstrip line is
equal. There is no extra quiescent power dissipation as would be the case using end termination
resistors.
ZO ≈ 10Ω
39Ω
*
> 4"
ZO = 50Ω
SHARC
DSP
ZO = 50Ω
SHARC
DSP
ZO = 50Ω
SHARC
DSP
ZO ≈ 10Ω
39Ω
CLOCK
*
ZO ≈ 10Ω
* Same
Package
39Ω
*
Figure 6: Preferred Method of Clock Distribution
Using Source Terminated Transmission Lines
Page 5 of 9
MT-097
TERMINATIONS IN BI-DIRECTIONAL LINKS
Figure 7 shows how source terminations can be used in bi-directional link port transmissions
between SHARC DSPs. The output impedance of the SHARC driver is approximately 17 Ω, and
therefore a 33-Ω series resistor is required on each end of the transmission line for proper source
termination.
ADSP-2106x
ADSP-2106x
OFF
LENGTH > 6"
33Ω
ZO = 50Ω
33Ω
ON
ZO ≈ 17Ω
LINK PORT
TRANSMITTER
LINK PORT
RECEIVER
Figure 7: Source Termination for Bi-Directional
Transmission Between SHARC DSPs
The method shown in Figure 8 can be used for bi-directional transmission of signals from several
sources over a relatively long transmission line. In this case, the line is terminated at both ends,
resulting in a dc load impedance of 25 Ω. SHARC drivers are capable of driving this load to
valid logic levels.
SHARC
DSP
SHARC
DSP
SHARC
DSP
+3.3V
+3.3V
120Ω
30mW
ZO = 50Ω
LENGTH > 10"
+1.4V
91Ω
22mW
120Ω
30mW
91Ω
22mW
SHARC
DSP
SHARC
DSP
SHARC
DSP
NOTE: KEEP STUB LENGTH < 0.5"
NOT RECOMMENDED FOR CLOCKS IN SYNCHRONIZED SHARC OPERATION!
Figure 8: Single Transmission Line Terminated at Both Ends
Page 6 of 9
MT-097
Emitter-coupled-logic (ECL) has long been known for low noise and its ability to drive
terminated transmission lines with rise/fall times less than 2 ns. The family presents a constant
load to the power supply, and the low-level differential outputs provide a high degree of
common-mode rejection. However, ECL dissipates lots of power.
Recently, low-voltage-differential-signaling (LVDS) logic has attained widespread popularity
because of similar characteristics, but with lower amplitudes and lower power dissipation than
ECL. The defining LVDS specification can be found in Reference 1, and References 2 and 3
should also prove useful. The LVDS logic swing is typically 350 mV peak-to-peak centered
about a common-mode voltage of +1.2 V. A typical driver and receiver configuration is shown in
Figure 9. The driver consists of a nominal 3.5 mA current source with polarity switching
provided by PMOS and NMOS transistors as in the case of the AD9430 12-bit, 170 /210 MSPS
ADC. The output voltage of the driver is nominally 350 mV peak-to-peak at each output, and can
vary between 247 mV and 454 mV. The output current can vary between 2.47 mA and 4.54 mA.
The LVDS receiver is terminated in a 100 Ω line-to-line. According to the LVDS specification,
the receiver must respond to signals as small as 100 mV, over a common-mode voltage range of
50 mV to +2.35 V. The wide common-mode receiver voltage range is to accommodate ground
voltage differences up to ±1 V between the driver and receiver.
AD9430 OUTPUT DRIVER
(+3.3V)
(3.5mA)
V+
V–
V–
V+
+1.2V
3.5kΩ 3.5kΩ
(3.5mA)
Figure 9: LVDS Driver and Receiver
The LVDS edge speed is defined as the 20% to 90% rise/fall time (as opposed to 10% to 90% for
CMOS logic) and specified to be less than < 0.3 tui, where tui is the inverse of the data signaling
rate. For a 210 MSPS sampling rate, tui = 4.76 ns, and the 20% to 80% rise/fall time must be less
than 0.3 × 4.76 = 1.43 ns. For the AD9430, the rise/fall time is nominally 0.5 ns.
LVDS outputs for high-performance ADCs should be treated differently than standard LVDS
outputs used in digital logic. While standard LVDS can drive 1 to 10 meters in high-speed digital
Page 7 of 9
MT-097
applications (dependent on data rate), it is not recommended to let a high-performance ADC
drive that distance. It is recommended to keep the output trace lengths short (< 2 in.), minimizing
the opportunity for any noise coupling onto the outputs from the adjacent circuitry, which may
get back to the analog inputs. The differential output traces should be routed close together,
maximizing common-mode rejection, with the 100 Ω termination resistor close to the receiver.
Users should pay attention to PCB trace lengths to minimize any delay skew. A typical
differential microstrip PCB trace cross section is shown in Figure 10 along with some
recommended layout guidelines.
‹
‹
‹
‹
‹
‹
Keep TW, TS, and D constant over the trace length
Keep TS ~ < 2TW
Avoid use of vias if possible
Keep D > 2TS
Avoid 90° bends if possible
Design TW and TG for ~ 50Ω
Figure 10: Microstrip PCB Layout for Two Pairs of LVDS Signals
LVDS also offers some benefits in reduced EMI. The EMI fields generated by the opposing
LVDS currents tend to cancel each other (for matched edge rates). In high speed ADCs, LVDS
offers simpler timing constraints compared to demultiplexed CMOS outputs at similar data rates.
A demultiplexed data bus requires a synchronization signal that is not required in LVDS. In
demuxed CMOS buses, a clock equal to one-half the ADC sample rate is needed, adding cost
and complexity, that is not required in LVDS.
CLOCK GENERATION AND DISTRIBUTION PRODUCTS
Analog Devices offers ultra-low jitter clock distribution and clock generation products for
wireless infrastructure, instrumentation, broadband, ATE and other applications demanding sub
picosecond performance. ADI clock products are ideal for clocking high performance analog-todigital converters (ADCs) and digital-to-analog converters (DACs). ADI clock ICs integrate PLL
cores, dividers, phase offset, skew adjust, and clock drivers in small chip scale packages.
Page 8 of 9
MT-097
REFERENCES:
1.
TIA/EIA-644-A Standard, Electrical Characteristics of Low Voltage Differential Signaling (LVDS)
Interface Circuits, January 30, 2001.
2.
IEEE Std. 1596.3-1996, IEEE Standard for Low-Voltage Differential Signals (LVDS) for Scalable
Coherent Interface, IEEE, 1996.
3.
Cindy Bloomingdale and Gary Hendrickson, "LVDS Data Outputs for High-Speed Analog-to-Digital
Converter," Application Note AN-586, Analog Devices, 2002.
4.
Hank Zumbahlen, Basic Linear Design, Analog Devices, 2006, ISBN: 0-915550-28-1. Also available as
Linear Circuit Design Handbook, Elsevier-Newnes, 2008, ISBN-10: 0750687037, ISBN-13: 9780750687034. Chapter 12
5.
Walt Kester, Analog-Digital Conversion, Analog Devices, 2004, ISBN 0-916550-27-3, Chapter 9. Also
available as The Data Conversion Handbook, Elsevier/Newnes, 2005, ISBN 0-7506-7841-0, Chapter 9.
6.
Walter G. Jung, Op Amp Applications, Analog Devices, 2002, ISBN 0-916550-26-5, Chapter 7. Also
available as Op Amp Applications Handbook, Elsevier/Newnes, 2005, ISBN 0-7506-7844-5. Chapter 7.
Copyright 2009, Analog Devices, Inc. All rights reserved. Analog Devices assumes no responsibility for customer
product design or the use or application of customers’ products or for any infringements of patents or rights of others
which may result from Analog Devices assistance. All trademarks and logos are property of their respective holders.
Information furnished by Analog Devices applications and development tools engineers is believed to be accurate
and reliable, however no responsibility is assumed by Analog Devices regarding technical accuracy and topicality of
the content provided in Analog Devices Tutorials.
Page 9 of 9
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement