[vcf-midatlantic] Repair Workshop Wrap-up 3/30 & 3/31

[Herb Johnson]

... and it's a two-fer of appreciation, with Bill Dromgoole's report on 
how the Univac repairs "get done".

Some of the failures you describe - gradual loss of function across 
cards - suggests to me general problems like power-supply voltages. YOu 
found a suspect supply, I see.

There's some really cheap little DC digital voltmeter boards from China 
for a few dollars. Putting a bunch of those on your DC lines may allow 
you to monitor them as you do repairs, look for sagging DC. Your "real" 
voltmeter can confirm the voltage; these widgets aren't that accurate 
but don't have to be. YOu might also 'scope out the DC, look for noise 
and glitches. As you know you need a lot of bandwidth to see DC spikes 
and dips which can trigger logic incorrectly.

Also, chemical changes occur in board contacts as current flows. 
Corrosion over decades may create a bit of chemistry. Then when current 
is applied, chemistry (corrosion) may increase resistance between boards 
and their edge-sockets. Removal/insertion will fix that, for awhile. 
I've had good results with DeOxIt chemicals - even with those IC's with 
"black leads".

A swipe of the DeOxIt brush across the board edge may work. I often 
follow that with some scrubbing with a white papertowel, to remove the 
oxide while maintaining that chemical coating. You can see dark-stuff 
come off. If there are pin contacts, a soft used toothbrush is a good 
scrubber. YOu don't want to "scrape", that removes gold and nickel! But 
you know these things.

Point being: never underestimate the problems with "contacts".

Herb Johnson

-- 
Herbert R. Johnson, New Jersey in the USA
http://www.retrotechnology.com OR .net