[Elecraft] Low output on higher bands

[Bill Coleman]

When I first built my K2 back in the spring, I seem to remember being 
able to get 10 watts on all bands, with no problem.

Now, I'm seeing a lower output on the higher bands. Let me clarify what 
I've done so far:

* CAL CUR is set for 3.5 Amps.
* For these tests Po28 is set to 15 watts
* TX current in TUNE with power set to minimum is 0.50-0.52 Amps.
* PA Transistor bias voltage with power at minimum is 0.638 volts
* Using 25 Amp supply that is nominally 14.1 V at no load (12.8 V at 20A)
* T2 is wound neatly and spaced off the PC board slightly. T1 is wound
  neatly, but not spaced off the board. 

Initially, I measured the following TX output to a 50 ohm dummy load 
(power out set to 15.2 watts):

(>= 15 watts on all lower bands, except 13.8 watts on 1850)
21050 - 12.5
21400 - 11.9
24930 - 11.7
28050 - 9.3
28600 - 8.8
29650 - 5.2

I decided to re-align the 17m/15m and 10m/12m transformers. I re-peaked 
them according to the directions in the manual. This resulted in:

21050 - 13.0
21400 - 12.2
24930 - 12.3
28050 - 9.7
28600 - 9.0
29650 - (not measured)

Better, but still not great. After this, I decided to re-wind T4. I 
originally wound it according to the directions, with the secondary wound 
over the primary. Since then, I've been told that winding the two 
interleaved helps performance on the higher bands. Re-winding was hard, 
mainly because I had no more white wire, and had to use the limited 
length of wire on the core. After this:

21050 - 14.0
21400 - 13.5
24930 - 12.8
28050 - 9.9
28600 - 9.3
29650 - 5.5

At this point, I'm totally satisfied with the performance on every band, 
except 10m. I've ordered the KPA100, so as long as the base K2 produces 
10 watts, the KPA100 ought to be able to produce it's rated power.

I'm puzzled why I can't get full rated output on 10m. When I first built 
the K2, I believe I was getting 10 w, at least on 28200, which is the 
calibration point. Now I can't get more than 9.9 watts.

Gary has suggested earlier that one PA transistor may be weak. Both seem 
to get warm, although I'm unsure if they are "equally" warm or not. 

I put a scope probe on the collector of each transistor. I see similar 
waveforms with similar amplitude on each. I wonder how I can tell if one 
has failed or is weak otherwise.

Suggestions?


Bill Coleman, AA4LR, PP-ASEL        Mail: aa4lr@arrl.net
Quote: "Not within a thousand years will man ever fly!"
            -- Wilbur Wright, 1901