Path: utzoo!utgpu!water!watmath!clyde!bellcore!rutgers!cmcl2!nrl-cmf!ames!pasteur!ucbvax!agate!saturn!eshop
From: eshop@saturn.ucsc.edu (Jim Warner)
Newsgroups: comp.protocols.appletalk
Subject: Re: PhoneNet cabling
Message-ID: <3611@saturn.ucsc.edu>
Date: 4 Jun 88 17:10:10 GMT
References:  <614@mtxinu.UUCP>
Reply-To: eshop@saturn.ucsc.edu (Jim Warner)
Organization: University of California, Santa Cruz
Lines: 30

In article <614@mtxinu.UUCP> ed@mtxinu.UUCP (Ed Gould) writes:
>
>As I think more, though, I remember that the cable that PhoneNet uses
>is flat, untwisted telephone cable.  So it might work better to use
>two pairs, selecting one conductor from each for the PhoneNet, and
>grounding the other.
>

No.  Apple's LocalTalk is RS-422 balanced differential signalling.
The two wires should be twisted together for the maximum noise
immunity.  You can read about it in EIA-422-A, available from the
Electronic Industries Association (Dec 1978).  Furthermore,
"grounding the other [wire]" has little meaning in an interbuilding
connection.  It is not safe to assume that ground will be the
same in different buildings.  

My reading of RS-422 does not lead to the conclusion that Farallon's
distance limits are conservative.  But whatever works works.  The
rules appear to be empirical rather than based on engineering 
analysis.

I have peeled open a Farallon transformer and looked at the measures
they have taken to guard against transients.  They use the same
transient suppression components that are used to protect 120 VAC
circuits.  I honestly can't imagine how these are at all appropriate
to protecting the attached equipment.

Jim Warner
Univ of Cal
Santa Cruz