Minutes of X3T10 SPI-2 LVDS Working Group Meeting

John Lohmeyer - SCSI jlohmeye at Symbios.COM
Sun Dec 24 19:20:01 PST 1995


* From the SCSI Reflector, posted by:
* <John Lohmeyer - SCSI <jlohmeye>>
Please note:  My normal email account is down (and may not be up for
              several days due to the holidays).  I am posting this message
              from another email account.  Please post any responses to
              john.lohmeyer at symbios.com.  Thanks -- John

Minutes of SPI-2 LVDS Working Group Meeting                     X3T10/96-102r0

Accredited Standards Committee*
X3, Information Technology
                                                   Doc. No.: X3T10/96-102r0
                                                       Date: December 24, 1995
                                                    Project: 1142-D
                                                  Ref. Doc.:
                                                   Reply to: John Lohmeyer

To:         Membership of X3T10

From:       John Lohmeyer, Chair X3T10
            Bill Ham, SPI-2 Technical Editor

Subject:    Minutes of X3T10 SPI-2 LVDS Working Group Meeting
            Milpitas, CA -- December 15, 1995


                                       Agenda

1. Opening Remarks

2. Approval of Agenda

3. Attendance and Membership

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

5. Setup Time Budget (96-103) [Bridgewater]

6. Releasing Bus from Active Negation [Uber]

7. DIFFSENSE Specification [Uber]

8. Voltage Mode Drivers [Bridgewater]

9. Meeting Summary [Ham]

10. Meeting Schedule

11. Adjournment




                              Results of Meeting

1.    Opening Remarks

John Lohmeyer, the X3T10 Chair, called the meeting to order at 9:30 a.m.,
Friday December 15, 1995.  He thanked Jim McGrath of Quantum for hosting the
meeting.

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

2.    Approval of Agenda

The agenda was developed at the meeting.

3.    Attendance and Membership

Attendance at working group meetings does not count toward minimum attendance
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. Wally Bridgewater  V  Adaptec, Inc.             wally at eng.adaptec.com
Mr. Richard Moore      V  Adaptec, Inc.             richard_moore at corp.adaptec
                                                    .com
Mr. Simon Ngo Sy       V  Adaptec, Inc.             simonn at eng.adaptec.com
Mr. Chris Burns        V  Adaptec, Inc.             chrisb at eng.adaptec.com
Mr. Dennis R. Haynes   O  Burr-Brown Corp.          haynes_dennis at bbrown.com
Mr. Richard Kulavik    V  Burr-Brown Corp.          kulavik_Richard at bbrown.com
Mr. Jaff Lin           V  BusLogic                  jaffl at buslogic.com
Mr. Joe Chen           P  Cirrus Logic Inc.         chen at cirrus.com
Dr. William Ham        A# Digital Equipment Corp.   ham at subsys.enet.dec.com
Ms. Nancy Cheng        A# Hitachi Computer Products n_cheng at hitachi.com
Mr. Dan Colegrove      A# IBM Corp.                 colegrove at vnet.ibm.com
Mr. Richard Greenberg  V  IBM Corp.                 richg at vnet.ibm.com
Mr. Dean Wallace       P  Linfinity Micro           75671.3443 at compuserve.com
Mr. Ting Li Chan       A  QLogic Corp.              t_chan at qlc.com
Mr. Richard Uber       V  Quantum Corp.             duber at tdh.qntm.com
Mr. Farrokh Mottahedin V  Quantum Corp.             fmottahe at qntm.com
Mr. Brian N. Davis     A# Seagate Technology        brian_davis at notes.seagate
                                                    .com
Mr. Dave Guss          A  Silicon Systems, Inc.     dave.guss at tus.ssi1.com
Mr. Bob Masterson      V  Sonoran Tech.             rwmast at aol.com
Mr. Vit Novak          A  Sun Microsystems, Inc.    vit.novak at sun.com
Mr. John Lohmeyer      P  Symbios Logic Inc.        john.lohmeyer at symbios.com
Mr. Mark Jander        V  Symbios Logic Inc.        mark.jander 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. Paul D. Aloisi     P  Unitrode Integrated       Aloisi at uicc.com
                          Circuits

26 People Present

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

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

Terminator specification:

Bill Ham discussed the changes to the LVDS terminator specification.

The first area of discussion was the differential impedance requirement.  The
new test circuit developed at the last working group was discussed.  The
method of specifying this performance requirement was accepted but the actual
values of the resistance were questioned.  Based on earlier inputs from
Unitrode that the sign of the reflections should always be such that the
terminator is lower impedance than a loaded bus a set of lower numbers for the
terminator were included in the draft standard.  These lower numbers decrease
the signal amplitude and were not accepted.  Further it was noted (see below)
that the leakage current from devices could have a significant effect on the
terminator bias.  This was noted and placed on a list of concerns.  The issue
was not resolved in this meeting.  It was noted that a more sensitive receiver
would be very beneficial here.

A value of slightly over 100 ohms was deemed to be optimal for the nominal
terminator resistance.

Output leakage current:

After some discussion, the output pin off current was specified at +/- 20 uA.
This level is required to accommodate 2.5 V IC processes that have higher
leakage than % or 3.3V processes.  This raised the question of what happens on
a fully-loaded bus when all the devices are leaking at the same extreme of
this specification (15 * +/- 20 uA = +/- 300 uA total leakage current).

Terminator common-mode impedance test circuit:

See Table 3 and Figure 3.  Kevin Gingerich suggested that the specification
should include AC characteristics  over the range of DC to 300 MHz.
Eventually, we settled on a range of DC to 1 MHz since only the common mode
noise needs to be accommodated -- not the SCSI signals.  This test was
accepted as valid with the addition of the frequency requirements.

Terminator balance test:

This test measures the balance of the terminator between the two sides.  Any
curve within the box (Vmin to Vmax and delta Vmin to delta Vmax) is legal.  A
10 mV imbalance was allowed.  Specified frequency range is DC to 1 MHz.  This
test was accepted as valid with the addition of the frequency requirements.

Driver current requirements:

Late in the afternoon, we re-visited the signal budget again.  Bill developed
a model of the driver and receivers, including crosstalk, leakage current,
unbalanced termination, and the receiver sensitivity of 70 mV.  We do not have
much margin for cable loading or attenuation with a 5 mA current mode driver
and the present common mode range.  It appears that we need to increase the
driver current again, perhaps to around 7 or 8 mA.  Or we need to increase the
receiver sensitivity and reduce the terminator bias voltage.  Limiting the
common mode range is another possibility.  There are other ways to keep the
chip power dissipation down that can be applied. Stay tuned.

5.    Setup Time Budget (96-103) [Bridgewater]

Wally Bridgewater presented a Proposed Change for Setup Time Budget (96-103).
It contains a justification for changing the setup time for Fast-40 from 5 ns
to 9 ns at the driver and from 1 ns to 5 ns at the receiver, adding 4 ns to
each number.  Wally said the additional 4 ns is needed because the received
data voltage levels are asymmetric due to the bias currents in the
terminators.

Kevin Gingerich suggested that the receiver design could compensate for the
offset voltages.

Bill Ham suggested that the setup time be increased by 2 ns to 7 ns transmit
and 3 ns receive.  He was concerned that increasing the budgets any higher
would create problems with the Fast-80 budgets.  Wally countered that the Fast-
20 Differential budget had 4 ns of time for the external drivers that could be
shifted to the transmit setup time. After a little more haggling, we ended up
with agreement on adding 3.5 ns to get 8.5 ns transmit setup time and 4.5 ns
receive setup time.

Fast-80 transmit setup and hold times were each changed to 5.75 ns and the
Fast-80 receive setup and hold times were each changed to 1.75 ns.  Bill
suggested that, with these changes, still higher speeds beyond Fast-80 could
not be achieved without protocol changes or going to a more tightly specified
interconnect.

6.    Releasing Bus from Active Negation [Uber]

Richard Uber bought up an issue regarding spurious glitches on LVDS lines
caused when the driver goes from the active negation to the released states.
The lines may appear to be true for up to a round-trip trip (approximately a
BUS SETTLE DELAY).  The major time of concern was the transition from MESSAGE
IN phase to BUS FREE phase.  Bill Ham asked the protocol chip vendors to look
at their chips to see whether such glitches would have any adverse affects on
their operation.

7.    DIFFSENSE Specification [Uber]

Richard Uber presented a foil pointing out several deficiencies in the
DIFFSENSE specification.  The present document shows this information as a
circuit example rather than as a performance specification.  Having lobbed the
grenade, Richard sat down and Bill Ham picked up foils and markers; he spent
over an hour developing a new specification for DIFFSENSE (see SPI-2 Rev 4).
These changes do not materially affect real designs underway but will improve
the way the DIFFSENS is specified in the standard.

8.    Voltage Mode Drivers [Bridgewater]

Wally Bridgewater presented the idea of using voltage-mode drivers with a
separate 1 volt supply.  The advantage is that the transistors would have a
very low voltage drop (~0.1 volts) so the on-chip power dissipation would be
quite low.  The disadvantage is that external voltage regulators would be
required.

There was some concern about how large of a common mode voltage range can be
supported; Wally thought the transistors could operate somewhat outside the
rails.  Another concern was the effect of the low-impedance driver on the
transmission line.  Wally plans to do more investigation for the Dallas
meeting.

It was generally agreed that the specifications for the drivers in the present
draft depend too strongly on a current source method of implementation.  By
forcing a current source and a set of compliance voltages one boxes in a power
dissipation in the chip.  As there is nothing intrinsic about the use of
current mode drivers with LVDS Kevin Gingrich and Bill Ham both called for
removing the required use of current mode drivers.  At the January meeting we
will develop performance specifications that do not preclude the use of
voltage mode drivers.  This will open up new ways to manage the chip power
dissipation and leave the implementation details to the suppliers.


9.    Meeting Summary [Ham]

Generally, this meeting explored several second order facets of the low
voltage differential transmission schemes.  While there were some new points
raised these were all of a tweaking nature.  No show stoppers have been found
yet.

At the request of the newly-formed SCSI Trade Association, a goal of acheiving
technical stability by the March `96 meeting was set.

10.   Meeting Schedule

The next meeting of SPI-2 Study Group will be Monday January 8, 1996, in
Dallas, TX at the Doubletree Hotel at Park West (214-869-4300), hosted by
Quantum Corp.  Another SPI-2 Study Group meeting is scheduled for February 5,
1996 in Denver, CO hosted by Symbios Logic -- It appears that the February 5
meeting will be re-scheduled due to a conflict with an FC-0 meeting and the
need for a two-day meeting to meeting the March `95 stability deadline.
Tentatively, the meeting will be moved to January 29-30, 1996 still in Denver,
CO.

11.   Adjournment

The meeting was adjourned at 5:25 p.m. on Friday December 15, 1995.

--
John Lohmeyer             E-Mail:  john.lohmeyer at symbios.com
Symbios Logic Inc.         Voice:  719-573-3362
1635 Aeroplaza Dr.           Fax:  719-573-3037
Colo Spgs, CO 80916     SCSI BBS:  719-574-0424 300--14400 baud





More information about the T10 mailing list