Minutes of SPI-2 LVD Working Group Meeting -- 3/11/96

Lohmeyer, John JLOHMEYE at cosmpdaero.co.symbios.com
Tue Mar 19 13:52:00 PST 1996

* From the SCSI Reflector, posted by:

Minutes of SPI-2 LVD Working Group Meeting                  X3T10/96-134 r0

Accredited Standards Committee*
X3, Information Technology
                                                   Doc. No.: X3T10/96-134r0
                                                       Date: March 19, 1996
                                                    Project: 1142-D
                                                  Ref. Doc.:
                                                   Reply to: John Lohmeyer

To:         Membership of X3T10

From:       Ralph Weber, X3T10 Secretary
            Bill Ham, SPI-2 Technical Editor
            Larry Lamers, Vice Chair X3T10

Subject:    Minutes of SPI-2 LVD Working Group Meeting
            San Diego, CA -- March 11, 1996


1. Opening Remarks

2. Approval of Agenda

3. Attendance and Membership

4. Test Data (96-141) [Ham]

5. Transmission Line Model (96-123 & -124) [Gingerich]

6. Timing Budget [Ham]

7. SPI-2 Document Review (X3T10/1142D) [Ham]

8. Releasing Bus from Active Negation [Uber]
   8.1 Digital Filtering of  Negation Release Glitches [Lohmeyer]

9. Standing Waves [Bridgewater]

10. Termination Balance [Ham]

11. Glitch Filtering [Ham]

12. Drive Currents [Moore]

13. Meeting Schedule

14. Adjournment

                              Results of Meeting

1.    Opening Remarks

Larry Lamers, the X3T10 Vice-Chair, called the meeting to order at 9:00 
Monday March 11, 1996.  He thanked Skip Jones of QLogic for hosting the

As is customary, the people attending introduced themselves and a copy of 
attendance list was circulated.

2.    Approval of Agenda

The agenda was approved with the following additions:

      10. Termination Balance [Ham]
      11. Glitch Filtering [Ham]
      12. Drive Currents [Moore]

3.    Attendance and Membership

Attendance at working group meetings does not count toward minimum 
requirements for X3T10 membership.  Working group meetings are open to any
person or organization directly and materially affected by X3T10's scope of
work.  The following people attended the meeting:

         Name          S        Organization         Electronic Mail Address
 ---------------------- -- ------------------------- 
Mr. Norm Harris        P  Adaptec, Inc.             nharris at eng.adaptec.com
Mr. Lawrence J. Lamers A  Adaptec, Inc.             ljlamers at aol.com
Mr. Tak Asami          A# Adaptec, Inc.             asami at dt.wdc.com
Mr. Wally Bridgewater  V  Adaptec, Inc.             wally at eng.adaptec.com
Mr. Richard Moore      V  Adaptec, Inc. 
            richard_moore at corp.adaptec
Mr. Edward Fong        P  Amdahl Corp.              esf10 at amail.amdahl.com
Mr. Ron Roberts        A  Apple Computer            rkroberts at aol.com
Dr. Benjamin Monderer  V  Box Hill Systems Corp.    ben at boxhill.com
Mr. Clifford E. Strang P  BusLogic                  skip at buslogic.com
Mr. Louis Grantham     P  Dallas Semiconductor      grantham at dalsemi.com
Mr. Siegfried Schmalz  V  Dallas Semiconductor      schmalz at dalsemi.com
Mr. Ken Jeffries       V  Dell Computer Corporation ken at volente.us.dell.com
Dr. William Ham        A# Digital Equipment Corp.   ham at subsys.enet.dec.com
Mr. Roger Cummings     O  Distributed Processing    cummings_roger at dpt.com
Mr. Ralph O. Weber     P  ENDL Associate            roweber at acm.org
Mr. Mike Chennery      A# Fujitsu 
                  mchenery at fcpa.fujitsu.com
Mr. Andy Chen          A  Fujitsu Computer Prods    achen at fcpa.fujitsu.com
Mr. Dan Colegrove      A# IBM Corp.                 colegrove at vnet.ibm.com
Mr. Dan Strevey        V  Intellistor, Inc.         dstrevey at intellistor.com
Mr. Dean Wallace       P  Linfinity Micro 
          75671.3443 at compuserve.com
Mr. Wayne E. Werner    O  Lucent Technologies       wew at alupo.att.com
Mr. Ting Li Chan       A  QLogic Corp.              t_chan at qlc.com
Mr. John A. Fobel      O  Rancho Technology, Inc.   scsi at trancho.com
Mr. Gene Milligan      P  Seagate Technology        Gene_Milligan at notes.
Mr. Gerald Houlder     A  Seagate Technology        Gerry_Houlder at notes.
Mr. Brian N. Davis     A# Seagate Technology 
       brian_davis at notes.seagate
Mr. Dave Guss          P  Silicon Systems, Inc.     dave.guss at tus.ssi1.com
Mr. Daniel E. Moczarny V  Silicon Systems, Inc. 
    dan.moczarny at tus.ssi1.com
Mr. Robert N. Snively  P  Sun Microsystems Computer bob.snively at eng.sun.com
Mr. John Lohmeyer      P  Symbios Logic Inc. 
       john.lohmeyer at symbios.com
Mr. Pete Tobias        A  Tandem Computers          tobias_pete at tandem.com
Mr. Kevin Gingerich    V  Texas Instruments         4307725 at mcimail.com
Mr. Tokuyuki Totani    P  Toshiba America           totani at tix.netcom.com
Mr. Kenneth J. Hallam  P  UNISYS Corporation        ken.hallam at mv.unisys.com
Mr. Paul D. Aloisi     P  Unitrode Integrated       Aloisi at uicc.com

35 People Present

Status Key:  P    -  Principal
             A,A# -  Alternate
             O    -  Observer
             L    -  Liaison
             V    -  Visitor

4.    Test Data (96-141) [Ham]

Bill Ham presented test data that was so hot off of the oscilloscope that he
had not fully analyzed it.  The data was based on a controlled load board, a
tool that allows the placement of a known load on a SCSI bus.  In the
discussion, Bill noted that the tests used a symmetrical driver, because he
has been unable to build an asymmetrical driver from discrete components.

One significant conclusion Bill reached based on the testing was that a 
filter is needed.  This is caused by reflections from clustered loads much 
in the single ended case defined in SPI-2.  The glitch filter would operate 
ignoring the signal state for between 5 ns and 7 ns after a transition is
recognized.  It was noted that this does not mean that a 5 to 7 ns low pass
analog filter be used.  Digital filtering was recommended.

Bill's test configuration usually involved several loads placed near one
terminator with a long (often 26 m) run of unloaded cable to the driver and
other terminator.  In several cases, Bill showed that 20 pF loads cause
unacceptable signal degradation in this configuration.  He also showed that 
lower per-node capacitance budget, such as 10 pF, produced a much better
signal to the point where adequate margin was seen.  This result is highly
dependent on the bus length and other physical configuration parameters. 
worst case occurred when a long string of one bits was followed by a single
zero bit (or vice versa).  The first changed bit signal had the lowest 
It was noted that this effect is precisely what causes the need for d.c.
balance in a manner similar to 8b-10b.

5.    Transmission Line Model (96-123 & -124) [Gingerich]

Bill Ham had incorporated the model in Annex B of the SPI-2 revision 5.  As
part of the SPI-2 review, Kevin Gingerich walked the group through the Annex
B.  One of the first steps was to remove all references to EIA/TIA 
A complete set of tests and limits are now in place for the LVD SCSI

6.    Timing Budget [Ham]

Rev 05 contained a proposal for a specific method for defining the timing
parameters: assertion period, negation period, set up time, and hold time.
This proposal followed the Fast-20 model.  This turned out to be one of the
most useful parts of the meeting because it forced real consideration of the
receiver thresholds.

The issue identified was the behavior of real receivers with respect to the
idea that they have a 30 mV sensitivity.  This 30 mV level was the d. c.
level or the level to be used if the differential signal for assertion is
approximately the same as the differential signal for the negation. 
when the differential signal for the assertion is large (as with a long 
of "1's" and the negation signal is small (as with a short "0") the actual
receiver threshold must be larger for the negation detection (if the same
timing values are to be met).  This is caused by "precharging" within the
receiver circuitry with the long assertion that must be discharged and
recharged by the short negation.  This observation is critical in 
what an acceptable signal really is and will significantly affect the
configurations that are possible.

The present thinking is that one should use a negation threshold that is 
times the assertion voltage level prior to the negation.  This allows a very
deterministic way of evaluating signals for their acceptability.

The above discussion also applies to the complementary transitions from
negation to assertion.

This impact of this on the acceptability of different configurations is not
presently known.

7.    SPI-2 Document Review (X3T10/1142D) [Ham]

Bill distributed revision 5 of the SPI-2 proposed working draft, dated March
11, 1996.  See other sections of these minutes for the details of this 

8.    Releasing Bus from Active Negation [Uber]

Richard Uber was not present.  Discussion of this topic was deferred to the
next meeting.  It was decided at the last meeting that it would be required
for devices to ignore glitches for a bus settle delay after release from
active negation.  It is not clear that this item needs further discussion.

      8.1   Digital Filtering of  Negation Release Glitches [Lohmeyer]

Discussion of this topic was deferred to the next meeting.

9.    Standing Waves [Bridgewater]

This subject was not addressed at this meeting but still needs review.
Continue to keep this item active.

10.   Termination Balance [Ham]

The group discussed the termination balance test as part of the proposed
working draft document review.  This subject was addressed as a result of
reflector traffic that pointed out that the specification in rev 04 was not
possible to meet.  After a lot of good discussion it was decided to adopt 
proposed wording in rev 05 but to remove the hard limits on delta Vmin and
delta V max.  This test is still viewed as a balance test for the LVD SCSI
terminators and will remain in the standard.

11.   Glitch Filtering [Ham]

During the presentation of his test data, Bill noted the need for a digital
glitch filter on all signals.  The glitch filter Bill described would ignore
all signal activity for between 5 ns and 7 ns after a transition has been

12.   Drive Currents [Moore]

Due to lack of time this item was not addressed at this meeting.

13.   Summary of Results

The group agreed that development of the SPI-2 standard is not yet complete.
A major area of concern was the impact of the revised view of the receiver
threshold on the allowed configurations.  The group agreed to a target of
stabilizing the LVD SCSI portions of SPI-2 at the May meeting.

14.   Meeting Schedule

Pending approval by X3T10, the next meeting of SPI-2 Working Group will be
Monday-Tuesday April 22-23, 1996, in San Jose, CA at the Sheraton Inn
Milpitas, CA (408-943-0600) hosted by Adaptec.  Another SPI-2 Working Group
meeting will be held May 6, 1996 in Fort Lauderdale, FL.

15.   Adjournment

The meeting was adjourned at 5:09 p.m. on Monday March 11, 1996.

More information about the T10 mailing list