[R-390] R390 Restoration

[Barry Hauser]

Hi Chuck & Gang:

You wrote ...

> I restored my Collins R390 over the long weekend.  I had it traded to me
on a
> "it worked the last time I tried it" basis.  Basically, that means it was
broken
> and the "seller" pretty much knew it.

Not unfamiliar, but hey, I thought that wasn't supposed to happen to you ...

You've probably already been in touch with Dave Medley.  BTW, the old links
to his site are wrong .. it's at

http://www.davemed.info/

I have several non-A's/'391's and they all work fairly well, though not all
of them are tweaked up.  My impression so far is there aren't the numbers of
problem components you find on the A, however the MTBF's may just be longer.
The worst trouble-spot is the power supply, as you found, and usually due to
the heat.  I came across an R-391 with a solid stated P/S and audio amp
which runs cool.  It was written up by Dr. Gerald Johnson in Issue #52 of
HSN.

Some previous owner did the mod -- only one tube left on the audio deck.  Al
Solway has replicated it fairly closely and is busy tweaking it for closer
regulation.  While my '391 was not done reversably, his mod is a plug'n
play.

<snipped>
> As soon as I got the power supply removed, I knew I was in the right
> neighborhood.  There was plenty of nip and tuck work that had been done
> under the regulator module including some modifications.  No cooked parts
or
> wiring.  The 47 ohm (?) resistors looked like new.  Noted the hum balance
pot
> so knew the supply had to be balanced out.  Replaced some out of spec
> resistors but found the real problem to be a 100pf mica cap from one of
the
> 6082's to ground that turned into about a 900K resistor.  I will have to
come
> up with some solution to the heat generated by that module though.  A CPU
> fan over one of the holes in the side might go far to help.

I used to think "mica was forever", but I guess not, and especially not when
broiled.

<snipped>

>I have heard of a green gear that if
> you pull or move, thats pretty much it.  I didn't see it and decided to
leave that
> phase alone.

Not exactly.  The green gear is as Dave Medley calls it "a do-nothing gear".
That's its purpose -- to make sure nothing happens.  There are three
possible places where it might be.  1.  Up high attached to the front frame
of the RF deck with a screw, meshing with nothing.  2.  In it's
non-operating operating position on a short shaft down low on the front of
the deck slightly to the right.  The gear is cupped so the teeth are offset
from its center.  Some decks don't have the tapped hole for storage on top,
so the gear is stored flipped over on it's regular shaft such that the teeth
don't engage the two gears it's supposed to freeze when in use. 3.  It's
missing altogether.  You'll see the drilled/tapped shaft down low with
nothing on it.  And maybe a drilled and tapped hole up high and a couple
inches to the left of center on the front ot the deck.

When the non-A RF deck is lifted off the mainframe, the two "halves" of the
geartrain are "unhinged" so to speak and can get out of synch.  Before
removing the deck, the procedure is to place the green gear in position on
its shaft, green side facing out so that it engages and locks the two big
gears together.  As I recall, there's an eliptical hole in the center which
mates to the shaft, but nontheless important to make sure it's seated and
screwed down tight.

However, all that said, there are other ways to secure the gears .. and if
they do get out of whack, it's not the end of the world.  I forgot to put it
in place once when I had to "double back" and remove the RF deck a second
time.  On top of that, I did some absent-mindedly twirling.  It wasn't that
big a deal to fix it -- the manual has the procedure for mechanical
alignment (or "alinement") from the ground up.

There's another oddity with the RF deck to be aware of.  Dangling
participles, uh, no, but there is a dangle factor.  One of the main drive
shafts goes through what is really a slot, not a round hole in the front
plate of the RF deck.  There is some vertical wiggle room.  If someone sets
the deck down hard and the bushing is not completely tight, the thing can
shift.  Too high and the counter will turn, but the rest of the business
won't.  Too low and the counter won't turn, and the RF deck might not seat
100%.  Something to watch out for.

> Left the gears in the RF deck and scrubbed them down with a
> long bristle brush, a toothbrush and carburator cleaner.  Got probably 70%
of
> the old grease out, rinsed the carb cleaner out and followed that with a
> scrubbing with 50/50 Simple Green and water solution then a thorough rinse
> with hot water.  Blew out the gear train with compressed air and let it
sit in
> the 100 degree heat we had one day.  Once dry, I re-lubricated everything,
of
> course.

I dunno if that ionizing type cleaner scare is the real deal.  Simple Green,
409, etc are ionizing cleaners which can saturate phenolic and unglazed
ceramic leaving salts behind which can result in arcing -- so we have been
warned.  Dave Medley writes about flushing with distilled water.

> Pulled each slug rack out and dipped the bearing in the carb cleaner and
gave
> them a twist.  Cleaned everything but the actual slug with a toothbrush
then
> submerged the bearing ends a final time and moved them.  Gunk came out
> and the bearings on the racks move nicely now.   Lubed the bearings and
> faces of the racks with 90W oil and placed them back in the radio.

Somebody recently wrote that there are R-390A's (Motorola's I think) that
have non-rotatable cam follower bearings.  Is that true?  Or were some just
staked too hard.

> No wholesale replacement of capacitors in the IF deck but noted there are
> certainly plenty of them.  I imagine they could give you a fit if one
started
> leaking.

I suppose, but this is one case where the older technology may have been
better.  It's funny, but a lot of my older vintage gear -- WWII and prior --
seems to work with a replacement cap here and there.  OK, some might qualify
for the oil depletion allowance ;-)
>
> Anyone out there in List Land have information on which ones commonly fail
> in this radio?  Other common problems that I ought to chase down before
> they become serious issues?

Dave Medley mentions some components, but mostly in the power supply.  I'm
not aware of any capacitor "hit list".

> Found the 1st RF tube replaced with a 6AK5 v/s a 6AJ5 as are all the
> oscillator tubes.  Also, there is a 6BH6 as the last IF amp.   I wonder if
these
> are helpful mods AKA: W3HM and the 75A-4 mods or did someone popped
> some tubes and just stuck those in 'cause they fit?

Sounds familiar.  I've found some swapparoos too.  Some I think are OK subs,
however there is that thing about the PTO tube in the A's -- whereby too
strong is not desireable.  The 6AK5 and 6AJ5 are not listed as subs for each
other, so maybe some mods were done under the hood.


> Repainted and re-lettered the front panel.  Stripped and painted the
knobs,
> cleaned all the sub-chassis, set the PTO length, balanced the upper and
> lower band end over-run and set up the mechanical alignment for the cams.
> Tied it all together by setting up the Cam Position/PTO length/electrical
> relationships, centered the Antenna Trimmer and then did a full IF and RF
> alignment.

That should be same-ol' same-ol' for you.

>
> To put it bluntly, the radios performance is impressive.  I did not do a
full
> sensitivity test but knew it was very hot.  For some reason, I had it
tuned to
> 15.400 and heard Radio Kuwait with no antenna connected other than the
> shielded BNC cable going to the HP-8640B generator.  I did a spot check at
> the 3.8 mhz RF deck alignment point and was floored.

Well, we're going to need a full review on that cost reduction program fifty
years after the fact.  In the New World Order, everything is subject to
investigation. ;-)

> Ben Wallace and Mike Harris (just finished his) are currently tied for
"most
> sensitive R390A's I have yet to restore."  This one beats both by a hair
but,
> keep in mind, its a different radio.  Much higher build quality in the non
"A"
> but I am not sure there are performance differences between the non "A"
and
> the "A" that would justify what must have been the cost difference.

Maybe not.  While it was a cost reduction thing, there were some outright
improvements.  I would suspect mechanical filters were more expensive than
LC circuits back then, but I guess refined selectivity was more important
than smoothness in the audio.  There is one weak spot in the gear train of
the otherwise heaftier non-A version.  One big gear in the middle is
attached to the front plate with a cover with 2 or 3 screws.  I've noticed
in all of my samples -- that gear wobbles or wiggles -- you see it when you
change directions.  I've been tempted to pull them apart and put a real
bearing in there.

> I'd like to find out the various problem caps and maybe pre-emptively
replace
> them.  Also, the filter caps seem to be either papers a-la the 32V-3 HV
cap or
> are oil filled's.  They could also probably stand to be replaced, again on
a pre-
> emptive basis.

I dunno -- as I mentioned, I've got a lot of old gear with oil filled caps
that still test good -- even ones that have started to seep because the
rubber seals around the terminals have dried out.  Might just need to be
topped up and sealed with silicone.  Reminds me, gotta go check the
newspaper under my AR88LF.  That's one heckuva oily radio.  Maybe it's some
kind of rust prevention system.  Quite aromatic, what with the PCB's -- a
bit pungent, like an old-time garage.

> This does not make me an R390 expert and I have no plans at present to
> start restoring them.  Having no real parts stock and little direct
experience
> makes that idea one who's time has not yet come.  However, its a start.

If nothing else, it's great as a perspective on the origins of the R-390A.
Basically the same, only very different.  One derived from the other, yet
just a few years apart and so much is not the same.  I'd think that's very
unusual in the annals of product development and manufacturing -- of
anything.  Front panel-wise/application-wise, they're nearly the same
device.   And then, there's the R-392.  When tweaked, the performance is
very close -- allowing for limits (b/w selection, etc.)  I recently took the
time to align one of those fully and was amazed at the performance,
including the sound quality when it isn't put through an LS-166.

Barry