Date: October 29, 1990 To: X3T9.2 Membership From: Lawrence J. Lamers, Chairman & X3T9.2 Secretary Subject: October 29, 1990 - Cable Working Group Meeting Larry Lamers called the meeting to order at 1:00 p.m., Monday, October 29, 1990. He thanked Bill Spence of Texas Instruments for hosting and arranging the meeting. As is customary, the people attending introduced themselves. A copy of the of the X3T9.2 membership list was circulated for attendance and corrections. The final agenda was as follows: 1. Color Coding Report - Lamers 2. Single-ended Fast SCSI - Spence 3. Cable Testing Methods - Fiala & Grant The following people attended the meeting: Meeting Attendees Name Status Organization ------------------------------ ------ ------------------------------ Mr. Robert C. Herron A 3M Company Mr. Denis D. Springer O 3M Company Mr. James Fiala S 3M Company Mr. Doug Makishima S Adaptec, Inc. Mr. Robert Otis A Apple Computer Mr. Peter M. Blackford P Astro Cable Company Mr. David Hess P Berk-Tek, Inc. Mr. Robert Liu P Fujitsu America, Inc. Mr. Terry Maezawa P Furukawa Electric Amer, Inc. Mr. Kurt Chan P Hewlett Packard Co. Mr. Mike Peper A Hewlett Packard Co. Mr. Howard Wang O Hitachi Mr. Chuck Grant A Madison Cable Corp. Mr. Lawrence J. Lamers P Maxtor Corp. Mr. Ken Erickson O Micropolis Corp. Mr. Gary Murdock A National Semiconductor Mr. John Lohmeyer P NCR Corp. Mr. Luke A. Perkins S NCR Corp. Mr. James McGrath O Quantum Corp. Mr. Gene Milligan A Seagate Technology Mr. D. W. Spence P Texas Instruments Mr. Edward R. Schurig O Texas Instruments RESULTS OF MEETING 1. Color Coding Larry Lamers reported on the events of the previous plenary meeting. The vote at the plenary meeting was 29-5 for no color coding. An implementors note from Bill Spence was accepted for inclusion in SCSI-3. It reads as follows: "For single-ended operation, cable assemblies incorporating shielded round, twisted-pair cable comprising a core surrounded by two or more layers of twisted pairs, the following principles shall be observed: (1) The REQ and ACK pairs shall be in the core. If there are more than two pairs in the core, the remainder shall contain ground pairs. (2) The data and parity pairs shall be in the outermost layers." 2. Retention Mechanism Kurt Chan owes us information on dimensions and the implementors note. This was not provided at the meeting. 3. Single-ended 3.1 REQ/ACK Signal Quality and Fast Single-ended [90-159] (Chan) Simulation and prototyping have shown that totem-pole single-ended drivers and receivers with increased hysteresis provide significant benefits to single-ended signal quality. Fast SCSI operation appears feasible. 3.2 Cable testing Bill Spence has tested Madison and Astro cables. He brought in the test results (see document 90-156). He plans to test Montrose cables within the week and have the test results in the next mailing. The Madison and Astro cables were indistinguishable even though the Madison cable was 79 ohms impedance and the Astro cable was 90 ohms impedance. Bill is planning to do a standardized impedance test on cables. What is desired is to specify a number of cable parameters for SCSI-3, which distinguish a good working cable. Document 90-157 shows the noise coupling into REQ/ACK, that characterizes a bad cable. The impedance and rise time degradation are also key factors. Dal Allan asked why a cable assembly test is not available that automatically determines a good cable. Such a device is needed for incoming inspection. Several factors are involved: signal isolation can be easily tested; rise time degradation can not be measured on a short cable and it needs a scope to check; characteristic impedance may be testable. Jim Fiala reported that he has received a literature package describing equipment that does these tests using HP equipment. John Lohmeyer suggested that a per foot and a per system basis be used for specifying rise time degradation. Jim Fiala suggested that square wave voltage attenuation be included, since a 32 gauge wire could meet the impedance and rise time degradation and not meet a square wave voltage attenuation. Chuck Grant suggested that a DC resistance requirement could be used in place of square wave voltage attenuation . Jim Fiala stated network testing centers are being set up on each coast for testing FDDI cables. He suggested that these centers be enlisted to test SCSI cables. Jim McGrath wants to have a discussion on glitch filtering and interoperability. The characteristics of a glitch need to be defined. National Semiconductor has trapezoidal transceivers (up to 17 Mhz) that deal with glitches, but more expensive. Gary Murdock has tested at fast transfers and found spikes going through the transition region because of the open collector drivers. Totem pole drivers may solve this problem. 4. Cable Testing Methods Chuck Grant presented his test method using a TDR. He explained how to measure impedance correctly with a TDR such that rise time does not affect impedance measurement. Jim Fiala defended the appropriateness of the n-termination test. He stated that if you vary a pulse generator this caused a change in impedance. Also the repetition rate at which the signal is injected affects the impedance measurement. 5. 68-Conductor Cable John Lohmeyer asked if other people were experiencing low impedance values with 68-conductor cable. It was apparent that nobody present had looked at 68-conductor cable from a parametric view. If there is a problem lurking out there it needs to be identified.