[R-390] R390 progress 6AK5/6AJ5

[Dan Merz]

Hi,  I decided to get into the non-a 1st/2nd oscill crystal box and see if
jiggling/cleaning the xtal pins would clear up the problem I've been having
getting the 7-8 Mhz band working.

The task was easier than I thought because Collins must have anticipated my
skill.  I took the oscillator chassis out,  removed the box that covers the
oven/xtal array on the end away from the two tubes and got down to my
previous point of trying to get the heater box off.  It was relatively
simple,  the four standoff posts unscrew and then the oven can be slipped
off without even disconnecting the heater wires.  I removed each crystal,
put a tiny amount of deox on each pin and reinserted them.  I put the unit
back in place with the oven slipped back on but without replacing the oven
cover to see if anything was improved.  It wasn't !!

Some background: I had noticed before all this disassembly that the screen
voltage on the 2nd oscill (the one that was giving a bad 10/20 Mhz waveform
for the 7-8 band was about 90 volts instead of the 60 volts specified.  The
plate voltage was close to the specified 160 volt value.  The 1st
oscillator, which has nearly identical circuit,  had more or less the
correct voltages on screen and plate.  I am using 6AK5's instead of the
specified 6AJ5's.  Switching the 6AK5's in the two oscillators did not
change the voltages so I ruled out an individual tube effect.  Putting a
6AJ5 into the 2nd oscill did bring the screen down to 70 volts,  closer to
the spec. value.  This is a bit of a puzzle.  I still haven't figured out if
I should be concerned about the 90 volts on the screen using the 6AK5 in the
2nd oscillator.  If both oscillators showed this,  I'd shrug it off as 6AK5
vs 6AJ5.  But the 1st oscill seems to operate to spec with the 6AK5.

Now after the xtal disassembly, with no improvement,  I decided to tweak
T402 the output transformer in the plate circuit since this is one of the
items that is adjusted in a complete alignment of the oscillator circuit. I
was hesitant to mess with this initially because most of the bands seemed to
work ok.  But turning this produced a big effect on the waveform of the
10/20 Mhz output and I was able to get it looking like I thought it should
with a nice 20 Mhz harmonic without the distorted form I saw before.  I now
suspect, even though most of the bands were "working ok" ,  this transformer
was not adjusted correctly when I got the set.  I note in the manual that
T402 is adjusted only once on band 31 and the procedure is somewhat cryptic.
It is adjusted so that the trimmer for band 31 can produce only one peak on
a VTVM over the trimmers whole range.  I interpret this to mean that the
inductance of T402 is reduced to the point so that only one set of overtone
harmonics are produced using the trimmer,  and presumably this is the
correct overtone for band 31.  My question is: when using a VTVM to monitor
this adjustment,  does production of an overtone produce a peak in reading?
It seems to make sense to me that it would.  The waveform in seeking out
overtones,  as the trimmer is adjusted,  goes from complete waves where the
average pp would be high through regions where the peaks are not reached
then to the next overtone where the peaks are again completely formed.  Am I
interpreting this correctly?

I'm  embarking on doing the complete alignment of this oscillator,  Dan.