[R-390] BFO Problem/Question for a newly restored R-390A

[Tim Shoppa]

"Chuck Curran" <ccurran at wi.rr.com> wrote:
> The main problem now is that the BFO would kick in, and then weaken and drop
> out within 10-20 seconds.  Sometimes it did not work at all, and I had to
> flip the switch repeatedly to get it to kick in.

My guess would be a leaky capacitor that gets worse when voltage
is applied (of course you wouldn't notice it until you turned on
the BFO!). IIRC there are at least a couple of black beauties in that
area. One is "just" some sort of bypass on the current-regulated filament
string through the BFO and PTO and is hardly critical.

> I first checked the resistances on V505 comparing all to the manual values,
> most were "close" to nominal values, but off enough to cause a strong
> interest, especially on pin 5.  I then removed the IF deck and started to
> dig in a bit deeper.  Upon checking individual resistances I quickly found
> that R530, the 22K plate resistor for this Hartley oscillator, had changed
> from a correct value of 22K to 33K.

Pretty common occurence! Almost all 2.2K plate resistors in my 390A's
were up by several hundred percent.

>  That did not seem to be good, so I
> installed a new  22K resistor.  OK, many of you know how crowded things are
> in that part of the 390A.

Well, in my yellow striper the BFO oscillator tube socket was crumbling
into tiny pieces of dust. I managed to replace that, although it was an
extended project.

> without being sure how much the shaft had rotated.  I cleverly told myself
> that it was surely like a variable cap, with a 360 degree rotation pattern
> that would just repeat itself.  Maybe not?

Not. In fact you will notice that the shaft is left-hand threaded, such
that when you turn it clockwise it backs out of the BFO. This moves a
slug in and out of an inductor core. It's also the reason that the flexible
coupling is also compressible.

> Here is my question, sorry for the long diatribe.  Should I start turning
> the BFO shaft until I achieve a tone at the expected points?  Is this shaft
> on a threaded adjustment, with 3, 4, or 5 turns possible so I can achieve
> this?  

Yes. Just don't turn it past its limits, if you're doing all this with
a counter then you will know when you're turning in the wrong direction
and not go very far that way.

When you're done, you might want to check the flexible/compressible
coupling to make sure that it's not stretched to its limit either.

> As an alternative, I thought I could try to sample this BFO oscillator
> output with my frequency counter, and see where it is.  I think I can do
> this using a tube test socket and just sample the signal off the available
> exposed V505 plate contact point.

I do fine with just sticking a scope probe near the tube (may
want to remove the tube shield first). That's enough of a signal to
register on my frequency counter or my scope. Also works great
for the crystal oscillators (the bandswitched one, the 200kHz one,
and the 17MHz one.) You can also see the spiky divided-down 100kHz
calibrator at the next dual triode in the calibrator too.

I know the "classic" R-390A tuneup procedure only needs signal generators
and voltmeters, but if you've got a scope or a counter then a lot of
this stuff becomes trivially simple.

Tim.