Ultra320 SCSI with Receiver Equalization, 25 meters into a Backplane with 6 loads. Russ Brown Quantum Corporation

Transcription

1 T1/-153r Ultra32 SCSI with Receiver Equalization, 25 meters into a Backplane with 6 loads Russ Brown Quantum Corporation SCSI Physical Working Group Meeting 7 March 2 Dallas, TX

2 U32 25 Meter Cable Test Overview 6 March 2 T1/-153r Russ Brown - slide 2! Quantum's goal for Ultra 32 SCSI is to have a solution that is so robust it could be extensible to Ultra 64.! In order to demonstrate that our Receiver Equalization scheme is extremely robust, we want to test it at conditions beyond the specified limits of Ultra 16.! The first of these was to test U32 using a 25 meter round cable into a fully loaded 6-slot backplane.! The signals were measured to find the eye opening with ISI, reflections and crosstalk including:! Amplitude errors;! Timing shift errors;! Miscellaneous noise.! The following describes the test and results.

3 6 March 2 T1/-153r U32 25m Cable Test Setup Russ Brown - slide 3! Margins were evaluated with the same techniques as used for our other Ultra32 data:! Transmitter driving voltage: +/- 4mV.! Transmitted Pattern: 2µs of "111..." training pattern followed by 8µs random data.! The equalizer input signals are captured differentially with a Tektronix TDS694C oscilloscope by probing at the backplane.! The equalizer output signal is generated by Spectre, simulating in transistor level models and using captured data as input stimulant.! For the bit cell: the training pattern ("11...") defines the cell boundary and cell center.

4 6 March 2 T1/-153r Russ Brown - slide 4 5 Ω aggressor XTALK source 1 U32 25m Cable Test Schematic Cable + backplane configuration 1 or 2 Perfect terminators DB(P1) 5 Ω random data source* DB() DB(1) 5 Ω aggressor XTALK source 2 receiver board identification: -XT8bd1 to XT8bd9 are on ribbon cable, -XT8bp1 to XT8bp6 or XT8bp1 to XT8bp15 are on the back-plane * TEK 241 HP8113A Perfect terminators Db1+ Db+ IC DbP1+ Db1- Db- DbP1- Differential probe on victim Termination 1Ω, located at end of back-plane

5 6 March 2 T1/-153r U32 25m Cable Test Configuration Russ Brown - slide 5! 25 meter Amphenol cable assembly using Madison 28AWG round shielded cable plus 6-slot backplane.! Waveforms 1Gs/s Balanced 1 Ω Termination 25m bp1 bp3 bp5 bd 25m 28AWG round shielded cable Backplane Single balanced transmitter, position zero bp2 bp4 bp6 supplied by Amphenol, bp1 bp6, receivers on backplane

6 6 March 2 T1/-153r Error Sources Russ Brown - slide 6! Error sources are used to define the range over which a receiver characteristic may typically vary from the ideal sample point, i.e., the actual sample point may lie anywhere within a box defined by 2 times -to-peak height and 2 times -to-peak width of the errors. 3 mv 2 ns U32 Eye Mask! Amplitude error sources define height, and timing error sources define width, e.g., set-up time margin is measured as the distance from the eye diagram waveform to the box.

7 U32, Eq Input, 25m Cable, bp1 Uneualied Signal m round cable + loaded 6-slot bp1 1 Cell Center Output Amplitude (mv) Output amplitude U32 Eye Mask KEY: pattern Random pattern ed ine 111 pattern Sampling Time 1pS Time (ns) 1-8 Conclusion: Failing Margin (Increasing amplitude would still fail) 6 March 2 T1/-153r Russ Brown - slide 7

8 U32, Rx Equalized, 25m Cable, bp1 6 March 2 T1/-153r Russ Brown - slide Equalizer Output 25m round cable + loaded 6-slot bp1 1 Cell Center Output Amplitude (mv) Output amplitude KEY: Red ine 111 pattern Random pattern Sampling Time 1pS x 1-8 Time (ns) U32 Eye Mask Conclusion: Excellent Margin

9 Set-up and Hold vs Eye Opening, bp1 6 March 2 T1/-153r Russ Brown - slide 9 Eye Opening Amplitude (mv) Eye openning Amplitude setup hold time vs eye opening amplitude 25m round cable + loaded 6-slot bp1 Solid line = set-up time Dashed line = hold time solid line setup time dash linehold time U32 Eye Mask Equalizer Rx Equalizer Output Output Unequalized Signal Unequallzed signal setup and hold time x 1-9 Set-up and Hold Time (ns)

10 U32, Eq Input, 25m Cable, bp Unequalized Signal 25m round cable + loaded 6-slot bp3 3 Cell Center 3.3 Output Amplitude (mv) Output amplitude KEY: pattern Red line 111 pattern Random pattern U32 Eye Mask Sampling Time 1pS x 1-8 Time (ns) Conclusion: Failing Margin (Increasing amplitude would still fail) 6 March 2 T1/-153r Russ Brown - slide 1

11 U32, Rx Equalized, 25m Cable, bp3 6 March 2 T1/-153r Russ Brown - slide 11 Equalizer output m 3 round cable + loaded 6-slot bp3 Cell center Output Amplitude (mv) Output amplitude U32 Eye Mask KEY: Red line pattern Random pattern Sampling Time 1pS x 1-8 Time (ns) Conclusion: Excellent Margin

12 Set-up and Hold vs Eye Opening, bp3 6 March 2 T1/-153r Russ Brown - slide m round cable + loaded 6-slot bp3 35 Eye Opening Amplitude (mv) Solid line = set-up time Dashed line = hold time Rx Equalizer Output Unequalized Signal U32 Eye Mask Set-up and Hold Time (ns)

13 U32, Eq Input, 25m Cable, bp6 ata file: tin2.dat m round cable + loaded 6-slot bp6 Input eye diagram: Output Amplitude (mv) Output amplitude U32 Eye Mask KEY: pattern Random pattern Sampling Time 1.25e-1 x 1-8 Time (ns) Conclusion: Failing Margin (Increasing amplitude would still fail) 6 March 2 T1/-153r Russ Brown - slide 13

14 U32, Rx Equalized, 25m Cable, bp6 6 March 2 T1/-153r Russ Brown - slide m round cable + loaded 6-slot bp6 4 Output Amplitude (mv) KEY: pattern Random pattern U32 Eye Mask Time (ns) Conclusion: Excellent Margin

15 Set-up and Hold vs Eye Opening, bp6 6 March 2 T1/-153r Russ Brown - slide setup hold time vs eye opening amplitude 25m round cable + loaded 6-slot bp6 Eye Opening Amplitude (mv) Eye openning Amplitude solid Solid line line setup = set-up time time dash linehold Dashed time line = hold time Equalizer Rx Equalizer outputoutput Unequalized signal Unequalized Signal U32 Eye Mask setup and hold time x 1-9 Set-up and Hold Time (ns)

16 6 March 2 T1/-153r U32 25m Cable Test Conclusions Russ Brown - slide 16! A Receiver Equalization scheme is extremely robust, capable at operating beyond the specified limits for Ultra 16.! The specification for the maximum bus path length between terminators (25 meters point-to-point and 12 meters multidrop interconnect) does not need to be changed for U32 in SPI-4.