Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!houxm!mtuxo!mtunh!mtung!mtunf!ariel!vax135!petsd!pesnta!greipa!decwrl!decvax!ucbvax!info-vax
From: info-vax@ucbvax.ARPA
Newsgroups: fa.info-vax
Subject: Re: VT241 (really cabling woes)
Message-ID: <8966@ucbvax.ARPA>
Date: Thu, 11-Jul-85 06:08:16 EDT
Article-I.D.: ucbvax.8966
Posted: Thu Jul 11 06:08:16 1985
Date-Received: Sat, 13-Jul-85 16:23:40 EDT
Sender: daemon@ucbvax.ARPA
Organization: University of California at Berkeley
Lines: 45

From: Bob Sutterfield 

As Richard Garland @CU20B has pointed out, pins 1 (protective gnd) and 7
(signal gnd) are wired together on most DEC VT devices.  I had this
confirmed for me in a most distressing situation by a local Terminals Field
Service rep:

My standard station cabling connects pins 2, 3, 7, and 20, with 2 and 3
crossed (standard `null modem').  When we wired an office complex with many
users on Rainbows (acting like VT102's) to Xyplex cluster controllers, most
came out allright.  When one secretary turned hers on one morning, with new
station cable connected to COMM port, she noticed funny lights, sounds, and
smells coming from her machine.  Upon further investigation (most of the
day), it turned out that the office had been wired for power about 23 years
ago, before it was common practice to wire the ground plug of the 120V
outlet to a real ground.  All the `ground plugs' in that complex are tied to
the neutral leg of the AC!!!!  This makes them float wherever they happen to
go, with no reason to agree on a common reference.

In short: ground plug goes to chassis of Rainbow, into RS232 asynch
controller area, where pin 1 (`protective' ground, remember?) is wired to
pin 7 (signal ground), thence over station cable into cluster controller (on
pin 7), where pins 1 and 7 aren't wired together, and pin 1 is really wired
to power ground.  Well, it turns out that there happened to be somewhere
near *82 volts* between the two `grounds' that day.

That voltage pushed current through the ICs on boards in both the cluster
controllers and the Rainbow.  Chips really exploded, and signal traces on
the printed circuit boards acted like fuses (several layers deep).  And, I
had no idea what the problem with the cluster was (thought it was software
at first), so I used some (clear throat) ineffecient diagnosis techniques,
roasting yet more equipment.

The carnage:	Two Rainbow motherboards
		Two Xyplex cluster controllers
		Probably around $7000 in all.
All was replaced under maintenance, although neither Xyplex nor DEC really
should have - it was a personal-favor-for-a-big-customer situation on both
counts.

The moral:  Check the validity of your grounds before installing any DEC
terminal products in an area with older wiring.

Mothers, don't let your babies do what I done...