[Collins] found the problem with kwm-2

[Dr. Gerald N. Johnson]

On 12/28/2010 11:31 PM, Fern wrote:
>         I finally found the problem with my kwm-2 which had two issues with
> audio feedback on transmit which occured when releasing the ptt as well as a
> slow drop off of the ptt. I found C47 (0.047mfd) , which goes to pin #3 of
> V14A and 275 B+,  was leaking. I will replace it with a 0.068 mfd rated @
> 600 vdc as the used to be 275 vdc is now much higher with higher line
> voltages plus it has to contend with a reversal of 100 vdc which was taxing
> the original cap real bad.

Revision 11 in the 1978 9th edition manual. Orange Drop suggested.

Digging into that obnoxious late parts list, I see the old C47 is one of 
the black beauties, molded oiled paper capacitor. They were leaky when 
new and leak more now. You can save hours of trouble shooting by simply 
replacing all you can find with orange drops. Don't bother to measure 
the leakage, they leak.

Having much more than 275 volts makes the PA tubes draw more off 
resonance current and so makes them touchier to tune.

Check the bias supply, the original selenium rectifier is way beyond its 
expected life time and can let the available bias voltage drop and that 
affects TR switching. In the extreme the M-2 can be trying to transmit 
and receive at the same time in some stages. A selenium with lots of 
reverse leakage is hard on the bias supply filter capacitor also which 
is not to say that an old leaky filter cap isn't hard on the selenium. 
Later power supplies used a silicon diode there.

Also found that both 68K resistors R20 and
> R47(which are in parallel)  which go to pin#7 the cathode of V4B had changed
> dramatically in value. Those two resistors were also taxed to the max as
> they are 2 watts each but will replace with a 35K rated at 5 watts.
>      73 from Fern   VE7GZ
>
>
Knepper has run into a rig with the 68K resistors removed and the 
cathode to ground circuit replaced with a 10 volt zener. My self, I'd go 
for a 15 volt zener and a 270K or 330K to +275 to be sure the zener is 
always turned on. It sure reduces power dissipated in those two 68K 
resistors but may cause the vox to hang longer since the cathode voltage 
won't be rising with more relay tube drive. I'd also put the 12K 
resistor in the relay tube plate because of that lack of cathode voltage 
rise. I've not tried this mod, only analyzed it.

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