[ct-user] Ethernet Problem

[David Robbins]

Dennis Egan wrote:
> 
> When I pulled the BNC plug off the NIC, I found that the terminating plug
> resistor was warm- so warm that I could hardly hold it.  I don't remember
> that from before, so, one at a time, I connected one computer to the cable
> and measured the voltage on the cable.  For 3 of the NIC cards, I have -8.7
> volts on the BNC pin- and on the fourth NIC, I have .33 volts on the BNC
> pin.

those terminating resistors should always be cold, the normal power they
dissipate is tiny.

> 
> Could 3 of the 4 cards been blown?  And still pass the diagnostics?  To add
> further information, both stack matches had numerous blown diodes and one
> had several vaporized traces on the PC board, and a 12V 35Amp power supply
> connected to them was putting out a hefty 26 volts! (probably regulator
> failure).  I didn't have a chance to replace the NICs today, but is that
> voltage on the output normal or do I have 3 blown cards?

yes, i would say you had 3 blown ones, and i would suspect the fourth one.  i
would expect no dc voltage reading on the cable.  when i measure an active cable
here with a handheld dvm i get flat 0v from shield to pin.  you should also test
the terminators to make sure they are still ok.

about passing the diagnostics... yes, depending on what type of diagnostics it
is very possible for the on-card circuitry to still be working so that the card
passes a simple internal test, but still have blown output circuits.  many cards
also include an external loop test where you set one machine up to echo back
while testing the other one, this provides a more realistic end to end test of
the whole card and cable.

sounds like time to turn everything on and let it run for a while, you probably
have more damaged stuff that won't show right away... be very sure to test the
computers well.

-- 
David Robbins K1TTT
e-mail: mailto://k1ttt@arrl.net
web: http://www.k1ttt.net
AR-Cluster node: 145.69MHz or telnet://dxc.k1ttt.net