[Elecraft] K2 transmit rapid drift on 10 meters

[Bill DeHaven]

While tuning up my K2 filters with spectrogram and tuning the bandpass
filters, I decided to see how the transmitted CW sounded on my K2 Rev B
with keying mod, temperature compensated oscillator mod, and KAT1 tuner,
so I drug out my 75S-3B receiver to listen as I tuned up into my dummy
load.  Everything sounded great on 80, 40, 20, and 15 meters, but on 10
meters, yuk.

Not only did the output sound low compared to what I was picking up on
15 meters for instance, but there was a very bad frequency drift on key
down that never seemed to stop.  I was measuring power output using the
detector on the DL1 dummy load and it seemed to be putting out the
expected power.

I also borrowed an HP8640 signal generator from work to test the
receiver sensitivity, since my XG1 seems to be not generating the
correct levels.  With the HP8640 I was able to test the receiver
sensitivity on all bands and found that the receiver is functioning very
well, even on 10 meters.

Any ideas on why the 10 meter transmit signal is so bad?  I suspect that
17 meters may be bad also but don't have a receiver covering that band.

There is very little circuitry that is not common to both transmit and
receive, so, since the receiver is working properly,  it seems like
perhaps there may be a parasitic oscillation in the transmitter output
stages.  However, the ever handy test finger placed on all the circuitry
reachable from the top of the RF board made no change to the 10 meter
transmitted tone or drift. The bang the sides and top test had no effect
either. 

Regards,

Bill DeHaven (ex N6ED, and I wish I hadn't let that one expire!)