Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site batman.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!genrad!panda!talcott!harvard!seismo!ut-sally!utastro!cositex!cyb-eng!batman!gene
From: gene@batman.UUCP (Gene Mutschler)
Newsgroups: net.micro
Subject: Re: Looking for advice on building 10-12 copies of a circuit board
Message-ID: <120@batman.UUCP>
Date: Wed, 10-Jul-85 18:20:59 EDT
Article-I.D.: batman.120
Posted: Wed Jul 10 18:20:59 1985
Date-Received: Wed, 17-Jul-85 03:59:10 EDT
References: <146@python.UUCP>
Distribution: net
Organization: Burroughs Austin Research Center, TX
Lines: 22

> I need to build ten or twelve copies of a circuit board,
> and would prefer not to do it by hand (a wire-wrapped
> prototype involved several hundred point-to-point
> connections).  How do people feel about some of the
> options available?
> 
>  - Automated wire wrap done from a complete schematic
>    and a component map by a company like Augat?
> 
>  - Printed circuit board, with quite an expense up front
>    in getting the design and the artwork right.
We have used this at Burroughs ARC.  We used Multiwire at the time,
although Burroughs can now do the same thing internally.  The first
time, we did a lot of things by hand.  There was a lot of up front work
as well as the substantial up front expense.  Debugging the resulting
board was somewhat tedious.  The second time around, we had some tools
on Mentor workstations.  The up front work was less, and the debugging
was much simpler, as we had a better handle on the design.  It would have
been extremely simple, but somewhere along the line (we think at Multiwire)
a node got dropped.  Unfortunately it had about 70 connections.  Ultimately
the board had to be remade, but we were able to basically plug the remade
version right in.