[Milsurplus] R-390 hints

[David Olean]

Hi surplus gurus,

I am overhauling a Collins R-390, a fairly low serial numbered unit, 
(#870) from the 1951 production order.  (NOT AN R-390A) When I got it, 
it was dead and a tag said it had a bad 6082 voltage regulator.  I noted 
all sorts of crispy surge resistors in the power supply as well as in 
the regulator circuit in the audio amplifier.  I replaced all the fried 
parts and tried operating the audio/ regulator unit all by itself on the 
bench. I was greeted by no regulation at all. I traced the problem to a 
0.22 ufd Sprague Vitamin Q cap that had long un-insulated leads. Someone 
had pushed it aside and the bare wire was touching another circuit. This 
caused the regulator to malfunction. When I fixed that, I saw that the 
regulator was stuck on 220 volts rather than 180 volts.   A bit more 
sleuthing produced some out of tolerance parts, no doubt caused by the 
excessive voltages that burned up all the 47 ohm surge resistors! 
Incidentally, the 0.22 ufd cap tested perfect with zero leakage after 68 
years.  I am running the receiver on 115 vac with solid state 
replacements for the tube rectifiers. At 185 volts regulated, the 6082s 
are not all that hot. I can touch the tops, but not the sides near the 
plates.

So I got the rig running with 185 volts regulated, and things started to 
work, but many of the lower 8 bands were inop.  I noted that fiddling 
with the Megacycle band knob would sometimes cause a band to start 
working.  After fixing a stuck tuning rack and broken slug, I narrowed 
the problem down to a bad crystal oscillator sub assy. The 1st crystal 
oscillator was not working properly.  It would show up as poor 
sensitivity on a band and the crystal output was low and at twice the 
frequency that it was supposed to be on! A look at the schematic showed 
that Collins switched in trimmers and fixed caps for each band. If the 
switch was messed up or dirty, the trimmers were not connected and the 
output tuned circuit was at a much higher resonant freq as a result. 
This explains the crystal output at double the correct frequency.  The 
second crystal oscillator for 8 MHz and above seems OK.

The previous owner had attempted to remove the crystal oscillator sub 
assembly. It was missing a screw or two on the back panel, and the 
internal green phillips head screws on the oscillator were rather 
loose.  So my question is. How difficult is it to open up the crystal 
oscillator and get at the band switch? It looks like a bear to work on, 
but possibly someone has done it and has a few ideas on how to proceed.  
I suspect that the switch inside is not clean or maybe is not exactly 
indexed.  Things are packed so tightly, that you cannot really see the 
switch or get at it. Dis assembly looks difficult and I am sure there 
are many pitfalls to watch out for.

Any helpers?

Dave K1WHS