[R-390] Z-503

[David Olean]

Hi Larry

I have been digging around my R-392 since mid February!

I was not sure how to progress with checking the AGC system, so I came 
up with a simple test.  The radio was distorting on strong stations 
after some amount of warmup time.  From memory, I think I was seeing 
-2.15 volts with about a -40 dBm input signal if things were sort of 
OK.  So what I did was couple in some negative voltage thru a 470K 
resistor into the AGC (black) test point.  I set the supply voltage to 
about 2 volts, and then went to each stage controlled by the AGC to see 
what ended up on the tube grid, pin 1.  This would test all of the 
bypass caps in the AGC line. I could also see any current draw in the 
AGC system by monitoring the voltage drop across my 470K resistor. There 
is a 1 meg resistor to ground in the AGC circuit.  I saw just about -2 
volts at all of the tube grids. This told me that all those Vitamin-Q 
caps in the AGC line were working well.  The only other area had to be 
the IF amplifier.

I focused on the IF amplifier chain, looking at stage gain. When I got 
to the 2nd stage, it looked like gain was not that great. -70 dBm 
antenna input produced 140 mv on the grid of V-502. -50 dBm was about 3 
volts peak to peak, but I noticed that when I pulled out the tube, the 
voltage soared. With little AGC voltage available, this surprised me and 
I figured that something was severely loading down the input circuit.  
Looking at the schematic I could see a 1000 pf capacitor across the 
primary of the IF transformer.  I could also see that there was some 
jitter with the rf voltage jumping around a bit in V-502 on the grid and 
the plate, but it was fine in V-501.  I suspected that one of the silver 
micas was shorting out and loading down the circuit.  I took apart the 
2nd IF can and disconnected the capacitor and it tested fine on two  
different (cheapy) capacitor meters. It had the correct value and showed 
no leakage. I then applied 250 volts DC and measured about 0.20 volts DC 
leaking through it. I also tested the suspect cap with my newly acquired 
GenRad 1657 LCR meter. Without putting DC bias on the part, it checked 
OK with no leakage. I need to fab up a fixture to get at the four test 
terminals rather than just two, on the 1657, so I can apply DC voltage 
across the cap, or I  am going to miss all these caps. Other caps in the 
transformer were tested as well. I replaced the postage stamp cap with a 
new 1000 pf silver mica, and now everything is working again. I re 
aligned the IF stages since I messed with all those caps in the IF can, 
but it was still pretty close after all the work.

Now the IF stage tunes the same whether the signal is weak or strong. 
With a flakey cap in the signal path, gain drops as the signal gets 
stronger.This impacts how much AGC volts get developed.  I get the same 
single peak in the pass band, whether weak or strong now. The distortion 
is gone. Only at about -10 dBm does the AGC fail. That is good enuf for me.

Out of three R-392s restored, I am keeping just this last one. I have 
the PTO  done up pretty well and accuracy is +/- 200 Hz or so across the 
whole band. Sensitivity is fantastic on all bands. Its a keeper!

I have a feeling that these SM caps are going to be a pain in the 
future. They are all giving up now!

73

Dave K1WHS

On 3/21/2020 9:02 AM, Larry H wrote:
> Hi Dave, Were you able to make any progress on your 2 392s that have 
> drifting agc voltage?
>
> Regards, Larry
>
> On Sun, Feb 16, 2020 at 8:42 AM David Olean <k1whs at metrocast.net 
> <mailto:k1whs at metrocast.net>> wrote:
>
>     I am having a similar problem with two R-392 receivers here. They
>     work
>     great when first turned on, but after some warmup, the AGC voltage
>     starts dropping, and the signals start to distort as the AGC voltage
>     does not limit the gain in the various stages. The lower AGC
>     voltage is
>     the problem. How the AGC voltage drops is a mystery for now.
>
>     Rather than individually test the capacitors, I got a can of freeze
>     spray, and will apply some freezing spray while monitoring the AGC
>     voltage.  I should be able to zero in on the problem quite fast with
>     this method. (I hope!)
>
>     I would not suspect alignment at all.  You could have a bad mica
>     cap in
>     the RF stages, but these will only affect a single band. (2-4,
>     4-8, 8-16
>     etc)  Trying the radio on different bands should show a pattern if
>     an RF
>     cap is defective. I had a bad cap in the 1-2 MHz band that started
>     shorting out with about 8 volts P to P across it. It was located
>     in one
>     of the 1-2 MHz cans.  I saved the cap to demonstrate how
>     capacitors may
>     fail. It tests OK on many cap testers. The only way I found it was to
>     put some DC voltage across it and look for noise on an oscilloscope.
>     Above 8 volts DC, I started to see DC voltage appear on the
>     opposite end
>     of the capacitor. Below 8 volts, there was no leakage!
>
>     I just bought a GenRad 1657 in an effort to test capacitors. I have a
>     feeling that some mica caps will be hard to isolate with such
>     testers.
>     I'll fill in more when I get some more experience.
>
>     73
>
>     Dave K1WHS
>
>