Fast-20 active negation current

Kevin Gingerich S=Kevin_Gingerich%S=Gingerich%G=Kevin%TI at
Thu Jan 26 08:37:00 PST 1995

  To  SCSI Reflector   X400
>From  Kevin Gingerich  GING
Subj  Fast-20 active negation current
I'm not sure this went out the first time. Sorry if it is a duplicate. /Kevin/
-********** ORIGINAL MSG RECEIVED ON 01/25/95 AT 07:16 FOLLOWS **********-
/pn=scsi/mbx1=scsi at
  To: SCSI Reflector   X400
From: Kevin Gingerich  GING
Subj: Fast-20 active negation current
The load that the single-ended SCSI bus presents to a driver is time varying.
The worst-case for assertion is a fully charged and loaded bus and for
is a fully discharged and loaded bus. The characteristic impedance of a fully
loaded bus may be as low as 57% of the cable impedance or about 50 ohms. A
driver in the middle of the bus "sees" two transmission lines in parallel or
To achieve the thresholds of VIH = 2 V and VIL = 1 V on the first transition,
there must be sufficient current available to drive the load(s) to these
levels. For negation,
        VO(t=0+) = (IOH(t=0+) + IOL(t=0-)) x 25 + VO(t=0-)
        2 = (IOH + 0.048) x 25 + 0.25
        IOH = 0.022 A
It is difficult, at best, to simulate t=0- conditions in a test circuit. The
"kick" needed from the active-negation may be verified by driving a 90 ohm
ground referenced resistor to 2 V. This requires different test circuits for
low-to-high and high-to-low driver output conformance testing.
Since these are ac output current requirements, you may make a case for some
sort of frequency limitations. However, I suspect it will be difficult (but
impossible) to come to an agreement on the requirement and how to test it.
Texas Instruments, Inc.
4307725 at

More information about the T10 mailing list