Day Night Problem

[Tom Droege <tdroege@veriomail.com>]

For some time I have been looking at a problem.  Sometimes the Mark IV 
writes junk.  Sometimes it is just high data point junk.  An occasional 
full scale point.  So I have been trying to find the problem.  It is an 
intermittent pain.  Today it worked perfectly all day.  No matter what I 
did it worked fine.  This running two systems.  Now it is dark and it is 
failing.  On several occasions it has failed late in the run.  Say 2 
AM.  Normally if things work in the tower where it gets to 45 C in the day 
time it should be OK at night.  Note that it also has worked through the 
winter.  So a temperature problem does not look too likely.  I am currently 
looking for something that starts to oscillate when it gets a little 
cool.  But I just ruled out the most likely candidate.  I am now thinking 
about humidity.  Sigh!  What could humidity be getting into?  No real help 
needed, just sympathy.  ;^)   I will find it.  Then ther are silicon diodes 
that leak a little in the daytime, but are OK at night.  But I can't 
imagine where I have one that could be giving such a problem.  Sigh!

This for some of you that might be wondering what I am doing since I am not 
getting MICHAEL and GLENN out the door.

Tom Droege