Date: June 19, 1990 To: Cable Working Group Participants From: Lawrence J. Lamers Subject: June 19, 1990 Cable Working Group Meeting Larry Lamers called the meeting to order at 1:30 p.m., Tuesday, June 19, 1990. He thanked John Lohmeyer of NCR for hosting the meeting. The attendance list was circulated. Participants were requested to add or verify their facsimile numbers so they could be contacted if needed. As is customary, the people attending introduced themselves. The final agenda was as follows: 1. Re-test of Madison 4099 Cable 2. Correlation of capacitance to crosstalk 3. Single-ended crosstalk measurements 4. Testing Cable Translators 5. Action Items 6. Next Meeting The following people attended the meeting: Name Organization ------------------------------ ------------------------------ Mr. Robert C. Herron 3M Company Mr. John Knudson 3M Company Mr. James Fiala 3M Company Mr. Al Wilhelm Adaptec, Inc. Mr. James R. Schmidt Adaptec, Inc. Mr. Charles Brill AMP, Inc. Mr. C. Edward Reynolds AMP, Inc. Mr. Bob Whiteman AMP, Inc. Mr. William R. Sopchak Amphenol Interconnect Mr. Jan V. Dedek Ancot Corp. Mr. Robert Otis Apple Computer Mr. Peter M. Blackford Astro Cable Company Mr. Thomas Debiec Belden Wire & Cable Mr. Wills Xu C&M Corp. Mr. John Guennewig Cinch Connector Mr. Fred Hengelhaupt Connective Technologies, Inc. Mr. Ed Haske Distributed Logic Corp. Mr. Terry Maezawa Furukawa Electric Amer, Inc. Mr. Kurt Chan Hewlett Packard Co. Mr. Tom Ryan Hitachi Cable Mr. Brian Griggs Hitachi Cable Manchester Mr. Ted Dubbs ITT Cannon Mr. D. Shaff JAE Mr. Robert Bellino Madison Cable Corp. Ms. Donna Pope Maxoptix/Maxtor Corp. Mr. Lawrence J. Lamers Maxtor Corp. Ms. Julie Duvall Molex Inc. Mr. Richard Wagner Montrose Products Company Mr. Andrew Jue National Semiconductor Mr. Gary Murdock National Semiconductor Mr. John Lohmeyer NCR Corp. Mr. Scott Peplinski NCR Corp. Mr. Ed Cieniawa Panduit Corp. Mr. John Emerson Prime Computer Mr. Gerald Houlder Seagate Technology Mr. Gene Milligan Seagate Technology Mr. Vit Novak Sun Microsystems, Inc. Mr. Kenneth Plourde Tempflex Cable Inc. Mr. D. W. Spence Texas Instruments Mr. Harvey Waltersdorf Thomas & Betts 40 People Present The following new documents were distributed at the meeting: Document Doc Date Author Description of Document ------------- -------- --------------- --------------------------------------- X3T9.2/90-93 6/18/90 K. Chan Single-ended crosstalk Measurement X3T9.2/90-91 6/18/90 J. Fiala Crosstalk-capacitance testing The results of the meeting 1. Re-test of Madison 4099 cable [Xu] The testing was not done. Wills Xu questioned whether the testing would make any material contribution to the committee efforts. 2. Correlate capacitance to crosstalk [Fiala] (90-091R0) Jim Fiala reported on the testing he had done of twisted pair cables to correlate capacitance and crosstalk. The results were not as expected. There appears to be no correlation between wire-to-wire capacitance and crosstalk. He also checked capacitance to ground, and again there was no correlation with crosstalk. See document 90-091R0 for details. The testing was done on special cross-talk fixtures using a cable that was the same as 4084 cable (IPI cable) 105 ohms. Jim suggested aluminum clad wires to control impedance; however there is a problem with terminating to an IDC contact. The center wires would then need increased diameter to .046 inches. Bill Spence asked if varying the wrap pitch could alleviate some of the cross talk. The response was that this changed the wire lengths and hence increased the skew. Gerry Houlder suggested counter-clockwise and clockwise wraps on different layers. While this might be possible it is a manufacturing problem. The bottom line is that the cables are going to have to be checked with the time consuming and laborious process of measuring the crosstalk on each and every pair as Jim finally did on the cable tested. Jim promised some further results at the next meeting. 3. Single-ended crosstalk measurements [Chan] Kurt Chan reported on the results he had of single-ended crosstalk measurements. The testing was done by driving alternating ones and zeros at 1 Mhz. Crosstalk is complex function. What was measured was the low voltage on negative going peak, i.e., the signal that will cause a drop below 2.0 volts. See document 90-093R0 for details. The averages differ from best case to worst case by 88 mv, and within each cable the pairs vary by 200 mv. What this suggests is that you don't throw money at cable technology, but instead pick the signal to pair location in the cable. This will minimize crosstalk. Apparently what the signal to pair location is for the cables that have been tested in the past is not known. The cables need to be dissected to look at what signals are connected to what pairs. The unshielded ribbon cable had the least crosstalk and was used to normalize the results of the shielded cables tested. The round cable is constructed in three layers, of 2, 8, and 15 pairs. The testing suggests that REQ/ACK signals be put in the center where impedance is highest, that reserved and TERMPWR lines be in the middle 8 pairs, and the data signals be in outside pairs. However having data in the outside pairs where impedance is lowest may be a problem because the edges (assertion and negation) are not as clean. Previous testing had indicated that keeping impedance high on data signals was desirable. Kurt stated that he will measure the best and worst cables from vendors to see if anything is gained. Kurt stated that if the edge of the first assertion is missed the signal must make a round trip to get detected. This could be up to 60 ns and thus the timing specification is violated. Therefore the edges must be clean enough to get detected on the first edge. 4. Testing cable translators This issue was not addressed. 5. Action Items 1) Jim Fiala to get further crosstalk results. 2) John Lohmeyer to get copy of NCR cable specification for reference. 3) Kurt Chan to measure best- and worst-case cables. 6. Next Meeting The next meeting is scheduled for July 9, 1990, at the Midway Motor Lodge in Rochester, MN. The starting time is 2:00 p.m.