Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site kitty.UUCP
Path: utzoo!linus!decvax!genrad!panda!talcott!harvard!seismo!rochester!rocksanne!sunybcs!kitty!larry
From: larry@kitty.UUCP (Larry Lippman)
Newsgroups: net.dcom
Subject: Re: Re: lightning, PACXs and computers
Message-ID: <231@kitty.UUCP>
Date: Sun, 4-Aug-85 18:22:54 EDT
Article-I.D.: kitty.231
Posted: Sun Aug  4 18:22:54 1985
Date-Received: Wed, 7-Aug-85 01:47:19 EDT
References: <1397@islenet.UUCP> <29116@lanl.ARPA>
Distribution: net
Organization: Recognition Research Corp., Clarence, NY
Lines: 69

> > > Briefly, all of the on-campus computer facilities with data communications
> > > lines extending beyond their immediate buildings (typically RS232
> > > 3-or-4-wire leased phone circuits or similar) suffered burned out
> > > ports and burned out terminals ...
> 
> We are in the midst of our second storm season at our new campus.  It's
> a small campus, but we do have several terminals in buildings separate
> from the computing center (where the cpu is located).  Except for scale,
> we are experiencing (repeatedly!) exactly what happened in Hawaii.
> 
> Our remote terminals are connected with rs232 cables, underground, in
> conduit.  Our power supply to the VAX is protected.  We have an Emulex
> CS11 (DH11 emulator) with 4 distribution panels.  During most electrical
> storms we have lost one of the panels and various other peripherals
> including terminals and our line printer (which is in the computing
> center).  Our first attempt at defense was to install individual surge
> protectors at the power connection of each remote terminal.  Next we
> added surge protectors to our dialup phone lines.  I assumed that the
> data lines to the remote terminals *couldn't* be the culprits, since
> they are underground and in conduit.
> 
> The latest advice (from one of DEC's field people) is that this is a
> typical problem with sites where equipment is in several buildings;
> that, during an electrical storm, each building will have a different
> potential to ground!  Thus, the ground wire in the rs232 cables suddenly
> carries the difference.  I have disconnected (at both ends) all of the
> terminals outside the computing center until we can get protection on
> the data lines.  We have had several lightning storms since, and (knock
> on wood) no more damage.

	[Sorry about leaving so much of the original article, but I felt it
important for anyone just joining in...]

	I am going to give you a simple and inexpensive suggestion which is
predicated on the following assumptions:

(1)	You are running async at 9,600 baud or less.

(2)	You do not require any RS-232 handshaking leads (RTS, CTS, DSR, DTR,
	etc.) and that any flow control is character oriented.

	The suggestion: Use optically isolated (at both ends) 30 mA current
loop, which requires two pairs - one for RxD and one for TxD.  You can get
fairly inexpensive current loop-to-RS-232 converters with *indivually*
isolated power supplies for well under $ 200.00 per pair.  If you or any of
your people are electronically handy, you can build what you need for less
than $ 60.00 in parts per data line.
	If your DTE and/or DTE already has a current loop option, then you
will not require RS-232 conversion.  You should make certain that any current
loop option is *passive* and optically isolated.  In that case, all you need
do is provide a current limited, isolated source of excitation current for
the RxD and TxD pairs.
	Using current loop as above, there is *NO* DC path from the DTE
apparatus to the DCE apparatus; the current loop which connects the two ends
is optically isolated and floats between them.  As a conservative measure, you
should use shielded pairs, but UNDER NO CIRCUMSTANCES should you connect the
shields to either DTE or DCE apparatus; just connect the shields to good earth
ground at both ends (i.e., water pipe).

	Larry Lippman
	Recognition Research Corp.
	Clarence, New York
	UUCP	{decvax,dual,rocksanne,rocksvax,watmath}!sunybcs!kitty!larry
		{rice,shell}!baylor!kitty!larry
		syr!buf!kitty!larry
	VOICE	716/741-9185
	TELEX	{via WUI} 69-71461 answerback: ELGECOMCLR

	"Have you hugged your cat today?"