[Collins] KWM-2A 2 repairs & difficulties (longish)

[Dr. Gerald N. Johnson]

Its only possible to predict the voltage range the bias supply should 
produce because the bias adjust pot is in the power supply and is 
adjusted to set the PA idling bias.

Tubes with grid emission push the S-meter negative more as the receiver 
warms up.

The S-meter pot may not be normal value either from age or an improper 
replacement. And since you found a resistor in the RF gain circuit way 
off in value, its worth checking all the resistors in the radio, the 
carbon comps used are known to drift. 18K to 50K is more than usual 
though, but I recall fixing an old TV set in 1959 where the horizontal 
oscillator was running at half frequency because of a resistor with 
drifted value. It made for weird symptoms, like a sync bar down the 
middle of the screen with just half a picture on each side. Comparing 
the video with the horizontal sweep with a borrowed scope soon showed up 
the problem. It was an RC oscillator so the resistor had to have at 
least doubled in value.

Especially check those in the S-meter stage.

If its not distorting on strong received signals with RF gain full up, 
the AVC is probably working correctly. The low voltage high value AVC 
time constant capacitor in ceramic is more prone to leaking than higher 
voltage ceramics.

One other check would be to hook a VTVM on the AVC line and tune in 
signals of various strengths (or the calibrator and vary the preselector 
tuning) and see if that AVC voltage rises well. There should be zero 
voltage with no signal I think. Another test to make. In most radios AVC 
voltage is zero with no signal.

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.

On 6/2/2011 2:01 PM, Al Parker wrote:
> Hi Jerry,
> Many thanks for your note. I've had several private replies, and have
> responded to each, and will put a note out to the list shortly. Might as
> well cc this one to the list.
> I ended up doing something somewhat similar to your suggestion, and
> others', that didn't involve much dis-assembly. Moving the electrolytic
> off it's post, allowed the tip of my small soldering pencil to access
> pin 8, so I hooked the new cap's + lead over the lead of the resistor
> already there, crimped it with hemostats, and soldered it. Then used the
> teflon from the old cap (I'd twisted both ends of it's leads close to
> their mtg points, using straight hemostats & broken them off) left the
> new cap almost vertical by the socket and att. the teflon covered - lead
> to a ground lug on the strip close to that post. New cap is almost as
> small as the old one, looks neat to me.
> I've done enough tight work on boatanchors to usually figure out what
> needs to be done to get a solid mechanical and acceptable cosmetic job,
> and try to do that (and not cause consequential damages in the process).
> The KWM-2 has lots of that sort of challenge.
> I suspected the bias supply, need to get back into the supply and take a
> look at what's been done in there. I have a"516F- metered test box" that
> was designed by someone & published some yrs ago in one of the Collins
> mags, and it showed -70 +- VDC on the bias line. I"ve thought about
> putting in some load resistors and relay switching in that box to get a
> better indication, plus it doesn't give any indication of ripple.
> Thanks for your thoughts on the AVC line. This appears to be a fairly
> late KWM-2A, it has factory installed plug-in relays, and no black
> beauties. All the bypasses that I chased down are disk ceramics. I'll
> find the time constant caps & ck them. I did swap out tubes in the AVC
> controlled stages, that didn't make any difference before I replaced
> that hi-value resistor, but did help a bit afterwards. I suspect there's
> still something pulling the AVC line down, as the S-meter adj. pot is at
> it's end right now. I"ll be getting back into the thing soon, I'm sure.
> I know Black Beauties well, and agree fully, if they're not bad now,
> they will be soon. I enjoy, and have worked on many, SP-600's.
> Hammarlund must have realized the BB problem as at about s/n 17,500 they
> changed to disk ceramics, and noted in one late manual that that's what
> should be used in any service work.
> Thanks again for your note, and all the help you give on the email list.
> 73,
> Al, W8UT
>
>
> On 6/2/2011 1:21 PM, Dr. Gerald N. Johnson wrote:
>> Check the bias rectifier in the power supply. Originally it was a little
>> boxed selenium and its way past end of life. Leakage there is also hard
>> on the bias filter capacitors. It takes full bias voltage to shut off
>> the TX and the RX for those respective modes. Total loss of bias makes
>> the M-2 operate very oddly. There is an S-meter zero pot and the circuit
>> is not without drift. The standard S-meter circuits use an AVC
>> controlled stage as a simple DC VTVM showing signal strength from the
>> developed AVC voltage. IF tubes with grid emission drive the meter
>> negative. Leaky time constant and bypass capacitors on the AVC line slow
>> or diminish AVC response and lead to distortion on strong signals. The
>> Black Beauties are leaky, not worth the bother of removing them in
>> condition to test and reinstall. Just put in orange drops after cutting
>> out the black beauties.
>>
>> You'll do less damage by cutting out the bad capacitor and wrapping (or
>> simply lapping) the leads of the new part over the pile on the lug than
>> trying to extract the old lead to make a hole. Some cuticle clippers are
>> sturdy enough to cut copper wire and thin to reach deep into the radio.
>> And there are small end nippers and bend side cutters that aid in
>> getting to some locations. If you have to run a new part lead a long
>> distance, it may be useful to sleeve the leads with teflon to insulate
>> from other circuits and to stand being against the soldering iron.
>> Collins bought much teflon sleeving.
>>
>> While it will hurt the stage gain, unhooking that cathode bypass
>> capacitor should clean up the audio with added inverse feedback and
>> proper bias.
>>
>> 73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association
>>
>> On 6/1/2011 1:45 PM, Al Parker wrote:
>>> Hi Folks,
>>> I lived with an AVC problem in my KWM-2A for several weeks, the RF gain
>>> ctrl had minimal affect on the AVC, and the rcvr's controlled stages
>>> were running wide open, it seemed.
> <snip>
>