[R-390] Further AGC measurements

Wed Oct 8 17:35:09 EDT 2014

You're welcome, Alan.  I hope it helps.  I forgot to say that the "excessively loud" symptom is not
unique to a bad AGC chain.  If the signal chain is working, then anything that spuriously moves the
AGC voltage in a positive direction will cause the signal output to be too loud.

When debugging an instrument that's been in bad environmental conditions, or simply one that's
very old, one must guard against expecting the schematic to 100% describe the instrument.
Degraded insulation, leakage paths... there may be things "connected together" that's you'd
never guess from the diagram.  Forgetting this may cut you off from a successful diagnosis.
"It's not what you don't know that'll get you, it's what you 'know' that ain't so."

Dave
________________________________________
From: Alan Victor [amvictor at ncsu.edu]
Sent: Wednesday, October 08, 2014 12:41 PM
To: David Wise
Cc: r-390 at mailman.qth.net
Subject: Re: [R-390] Further AGC measurements

Thank you David. I will take a look at these items in addition to others.

I appreciate your time taken and I will let you know what I find.

The assumption, the radio is not corrupted with mods etc... Is valid. It is quite original but not a pedigree.

On Wed, Oct 8, 2014 at 3:17 PM, David Wise <David_Wise at phoenix.com<mailto:David_Wise at phoenix.com>> wrote:
First, let's assume it hasn't been butchered.  If it has, then all bets are off; anything
could be miswired to anything.

>From a systems point of view, there are three possible causes for subnormal AGC voltage.

1. Signal chain gain is subnormal.  Less signal means less AGC.  This is pretty likely.
 Standard troubleshooting techniques apply.

2. AGC chain is bad.  No matter how much signal you have, AGC will be low.  Symptom is
  excessively loud (maybe distorted) output, since the signal chain will
  be running with excessive gain.

3. Something is loading down or pulling up the AGC line.  Standard suspects are listed below.
You probably have more than one.

 a. Leaky C551.  Look for a change when switching between FAST, MED, and SLOW.
  If this cap has not been replaced, I guarantee it's bad.
 b. Leaky mechanical filter(s).  The foam rubber goes bad.  With power off, look for
  a change in resistance when changing bandwidth.
 c. Leaky antenna trimmer.  This variable cap is tied to AGC.  It's insulated from the
  chassis, but the insulation is exposed to contamination and can easily get leaky.
 d. One or more tubes have grid emission.
 e. One or more coils, switches, etc are leaking to the AGC line.

We've taken care of item a.
Items b, c, d, and e can be diagnosed in a number of ways, but here's what I do.

First, we will partition the AGC system into "the controlled" and "the controlling".
We will analyze the former first.

Disconnect the AGC NOR jumper between TB102 terminals 3 and 4.
Except for R201 and R234 in the RF deck, and a number of tube grids
which will conduct if driven positive, the controlled side of the AGC
net should be a perfect open circuit.

Measure the voltage at TB102-4 with the radio running.  It should be
around 0V or slightly negative.  (The negative comes from the Contact
Potential effect.)  If it's more than about 0.5V positive, you have B+ intrusion,
grid emission, or both.  Pull the following tubes, one by one, until you
see the voltage fall: V201, V202, V203, V204, V501, V502, V503.
If the voltage stayed positive even with all these tubes out, you definitely
have B+ leakage.  (You might also have grid emission; you will have to
check again after you have eliminated the leakage.)

I had much more but I got tired.
Post your results and I will direct you further.

Regards,
Dave Wise

________________________________________
From: R-390 [r-390-bounces at mailman.qth.net<mailto:r-390-bounces at mailman.qth.net>] On Behalf Of Alan Victor [amvictor at ncsu.edu<mailto:amvictor at ncsu.edu>]
Sent: Tuesday, October 07, 2014 11:34 PM
To: Craig Heaton
Cc: r-390 at mailman.qth.net<mailto:r-390 at mailman.qth.net>
Subject: Re: [R-390] Further AGC measurements

Hi Craig,

excellent question and you are probably right! My approach was driven by...
Is this radio electrically sound or is there something so bad that a
mechanical tear down is an absolute necessity? And lets start ripping out
stuff. I looked at CAM alignment, no binding, no slug hangup, decent
sensitivity above 8 MHz. Checked for bad mechanical filters, AH yes,
replaced the .01 uF Vitamin Q (which I subsequently measured and it was
spot on and not leaky), pulled the power supply filter caps, a-ok, and the
8uF acid leaky cap in the audio section, was pristine! That all said,
decided to look at troubelshooting what was to me obviously not right
electrically. That process found several of the 2nd LO xtals way off in
spec. Fixed. The PTO over range and end points are nearly spot on. And so I
continued along the electrical route with this final bit with the AGC. At
this point I think I am finally convinced I must pull units to begin a
probe for out of spec items. The tube extender and playing modules out of
the main frame is attractive to me for troubleshooting but has mixed
reviews. I do have evidence now that the MC bandswitch needs attention,  as
the sensitivity below 8 MHz degraded last night only to return after
changing to say 21 MHz and then returning back to 7 MHz. So, yeah, now I am
seeing some mechanical ugly issues. Finally, the AGC fault is not going to
get resolved without pulling the subchassis. Hence, now its time to do what
you suggested. P.S. the power supply is not converted, not solid state,
supply caps look and play like new, and operation for now is on variac.
. .
On Tue, Oct 7, 2014 at 7:49 PM, Craig Heaton <hamfish at efn.org<mailto:hamfish at efn.org>> wrote:

> Alan,
>
> A couple of questions, sort or wondering if the cart is in front of the
> horse? Going back thru this thread on the R390 e-mail reflector & AM Forum,
> have you replaced the BBOD's & electrolytic caps yet in the entire
> receiver?
> Have all the switches been cleaned? There are so many of those little
> got-ya's which should be addressed first. Have all the variable IF & RF
> transformers been removed/cleaned so there is good pin contact?
>
> I'm rather new playing with these R-390/A's, it takes me about two weeks of
> cleaning, replacing old caps, etc., before any mechanical or electrical
> alignment. After which comes all the bugs to track down. (poor grounds, bad
> pots, leaky caps, more out of speck resistors)
>
> Should the module pull & mainframe breakdown come first?
>
> Craig,
>
> ----------------------------------------------------------------------------
>
> ----------------------------------------------------------------------------
> --------------
>
>
>
> -----Original Message-----
> From: R-390 [mailto:r-390-bounces at mailman.qth.net<mailto:r-390-bounces at mailman.qth.net>] On Behalf Of Alan
> Victor
> Sent: Tuesday, October 07, 2014 6:54 PM
> To: r-390 at mailman.qth.net<mailto:r-390 at mailman.qth.net>
> Subject: [R-390] Further AGC measurements
>
> I completed some more AGC measurements and have some clues.
> However, bottom line, ready for a module pull and mainframe breakdown. For
> the record though, I found the following:
>
> Monitor the AGC TP3-4 points with VTVM. That AGC voltage tracks AGC feed to
> the RF amp, and the three mixer grid control AGC voltages within 0.2 V.
> Still the AGC value always low even on large RF input of 150 uV/30%
> modulation.
>
> First a careful Z measurement again on TP 3 shows ~ 380K. Clearly less than
> the desired 500 K. Possible issue here.
>
> Next, look at all the series grid R values from RF amp and all mixers. The
> values are all on target, except for the RF amp, its 50 K to large!
> Measured
> over 500 K. Possible issue here.
>
> Finally, the most interesting measurement, with the AGC loop open. Function
> switch is set for the AGC ON, MEDIUM time constant.
> The AGC control line is fed via an external power supply with series 10K R
> to pin4. The RF input level is 150 uV/30% AM mod, now I can easily control
> the AGC line so that -7 V (AGAIN AGC OPEN LOOP) is generated at TP3 and the
> carrier meter comes alive.
> Larger input RF level can readily drive the AGC line pin3 to over -12 V.
>
> So the AGC circuit (OPEN LOOP) is functioning. Clearly, with the AGC loop
> closed it is not. Either the AGC source Z is not nearly low enough to
> source
> the required current (includes leakage) or the leakage which is present is
> larger than desired for proper AGC action.
>
> I think it would be helpful to measure the actual power supply current when
> this experiment is run. That would confirm that the current required by the
> AGC loop closed is just to high to support a functioning AGC.
>
> Comments?
>
>
> Alan
> ______________________________________________________________
> R-390 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/r-390
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:R-390 at mailman.qth.net<mailto:R-390 at mailman.qth.net>
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>
>
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net<mailto:R-390 at mailman.qth.net>

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html