Re[2]: remote twinax connection problem

[Buck Calabro <mcalabro@xxxxxxxxxxxx>]

>
What kind of PC's are you running.  Sounds like a PC conflict..
>
>But all at the same time, and healing at the same time?  Can that be the
>PCs?  that doesn't seem to me to be a common point, but rather three
>independant points acting alike..

Had a very similar problem years ago.  IBM CE told us that our twinax 
line was too long, even though it was physically under the maximum 
length.  After much debate, we realised that he was measuring the
end-to-end resistance of the line, and THAT was out of spec.  We had
a rats nest of barrel connectors, extra cable (for those offices that were
temporarily vacant) and so on.  Once we removed those surplus barrel 
connectors, the problem got MUCH better.

We suspected a bad twinax card, but had no way of finding out which
one was bad until we bought a brand new IBM twinax card and swapped
it into each PC in turn until the failures went away.  We used the "marginal"
card in a local PC without any problems.

One of our "engineer" types explained it this way: One "marginal" twinax
card can have enough of a mis-match to fail on any given day, and when it
does, it eats enough of the signal that the other twinax cards can't see it
either.  Dumb terminals work because they have strong transmitter/
receiver circuitry: they're not limited by how much power they can draw from
the PC's bus.  Such a "marginal" card will only show up when it's looking
at an attenuated signal: say 1000 feet or so...

Hope this helps

Buck Calabro
Commsoft

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to "MIDRANGE-L@midrange.com".
| To unsubscribe from this list send email to MAJORDOMO@midrange.com
|    and specify 'unsubscribe MIDRANGE-L' in the body of your message.
| Questions should be directed to the list owner/operator: david@midrange.com
+---