> something is amiss with cabling.  Back in the old days, when we used coax
> thin-net cabling (thank God it's gone), these strange problems came up
where
> I used to work.  Some computers could see some others, some computers were
> completely useless, one could see everything and worked just fine.  Guess
> which one had the problem.
>
> This guy moved his office around and instead of extending the coax loop
(as
> he was specifically instructed), he just hooked an extension piece (about
> 10' - 15') to the Tee connector.  All I can figure is that it acted like a
> "matching stub" and caused nulls (from reflections cancelling real data)
in
> certain spots on the lan.  That (and the silver satin story) are the
> strangest things I've seen in dealing with networks.  But tomorrow is
> another day.  ;-D


This is an ABSOLUTE no-no forbidden verboten must not do etc etc thing in
coax LANs.
Can and will destroy the whole network functionality. Such a "stub" is a
variable fraction of a wavelength long depending on its length, and places
an impedance bumb at that point which can be anything from a dead short to a
very high impedance point and will happily spread echoes across the whole
network. Good fun.

Another interesting tricj is to insert a length of different impedance coax
in the main backbone. Similar effect but not as severe usually as the bump
is better defind (usually 50/75/50 ohms) although the length will affect
what the echoes do. Rather hard to find without a TDR or similar tool.


        RM

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads