[SI-LIST] : IBIS models, how to fill in the missing pin info and other.

jaydiep@us.ibm.com
Tue, 31 Aug 1999 15:08:31 -0400

Andy Ingraham wrote,

>>If you mean something like a 74ABT family, then I would expect about the
>>same variation as one might see from one vendor to another. But be careful
>>about different drive levels within the same family, i.e., 64mA vs. 24mA.
I agree, and add that from my experience with ABT and other families, one must
be very careful. The ABT family for example, was introduced by both Philips
and TI, but even though they both made some common parts, the designs and fabs'
processes were NOT the same, even for those parts. Even from just Philips, a
number of different input circuits and buffer cells were used for even parts
with the same output current rating. One really needs to dig to find out
which part uses which subcircuits, as their characteristics do vary. It was
never clear why this all happened, but I would guess that the various parts
may have been designed by different groups. My testing showed the difference
in just edge rates between even a lowly ABT245 from the two suppliers was
significant.

>>Some modern logic parts are built using a gate array process, in which case
>>many or all parts in that family may use the same exact I/O cell designs.
>>(A recent example I saw of this was a 74xxx16yyy 16-bit buffer.) In that
>>case, a family model ought to be about as accurate as one for a specific
>>part number in the family.
Yes, you're right, some do use gate arrays. Some do not, however, and only
the supplier can tell you which is which (unless, of course, you want to test
them all and see how the characteristics vary). Some "gate arrays" in the
case of buffer parts are really what I would call device arrays, so even that
is not sufficient to produce equivalent parts from an SI perspective.

Jay Diepenbrock

Senior Engineer
Interconnect Technology & Qualification
IBM Global Procurement, B8UA/061, RTP, NC
Phone: 919-543-8804 Fax: 919-543-3642
Email: jaydiep@us.ibm.com

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****