Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!ucbvax!hplabs!hpda!hpcuhb!hpscdc!rkarlqu
From: rkarlqu@hpscdc.HP.COM (Rick Karlquist)
Newsgroups: sci.electronics
Subject: Reliability:  54 vs. 74 TTL
Message-ID: <5770009@hpscdc.HP.COM>
Date: 9 Jun 88 20:21:00 GMT
Organization: Hewlett-Packard, Santa Clara Div.
Lines: 29

Does anyone out there know what is the precise difference between
Commercial temperature range and Military temperature range IC's?

The question came up:  If the ambient is say 69 degrees C, is a
54XX logic IC more reliable because it has been "derated from 
125 degrees C" that a Commercial one that is only rated to go
from  0 to 70 degrees C?  Let's assume both are in a frit seal
ceramic package so we are not confused with comparisons between
plastic and hermetic package reliability.  My understanding is
that there is _no_ increase in reliability in the above case
and in fact there is no permanent damage done to the commercial
IC if it is operated at temperatures up to 125 C, it just won't
meet its specs, may not even "play".  But once returned to 
normal temperatures, it will be fine.  The only problem is that
reliability decreases a factor of 2 for each 10 degrees increase,
but then that would also be true of a 54XX part.

My understanding is that 54XX parts get built on the same line
as 74XX parts, and the design is the same.  For example, they
don't design 54XX parts to use less power to compensate for the
greater temperatures involved.  Now we're talking about plain
vanilla 54XX parts here, not 54XXJANTX parts which _do_ get
special T.L.C. treatment.

By the way, are JANTX parts really more reliable that commercial
parts (on the average) or is it merely that more is known about
the reliability?  I have heard that some JANTX parts are actually
less reliable.  

Can anyone out there shed some light on this?