Totally aside from the power meter issue, what you are doing with the
ATU is like loading a tube amplifier more lightly with the PI network.
Lighter loading will improve efficiency, but may result in more
distortion. One reason solid state amplifiers are generally less
efficient than tube amplifiers is that transistors (including FET's)
become extremely nonlinear when they are driven close to saturation.
That is, when the voltage across the device is near zero at the bottom
of the cycle. As a result, solid state amplifiers have to be loaded more
"heavily", to keep the transistors far from saturation. This, of course,
results in more dissipation.
I haven't tried it, but it's possible the KPA1500 could be loaded more
lightly by manually adjusting the ATU values to improve efficiency on CW
or RTTY, where linearity isn't a big issue. That might reduce PA
temperature a little, though I'm not sure it would be worth the trouble.
Given how sensitive the KPA1500 is to small changes in SWR, though,
setting this up across a band would be a tedious process.
The rather large variations in PA current I see at 1500 W output at a
moderate SWR, like 1.3:1, is probably due to the same effect. (A 1.3:1
SWR could be a load impedance of anything from 39 to 65 Ohms.)
73,
Scott K9MA
On 8/31/2019 09:15, Roger D Johnson wrote:
I think I've discovered a fly in my ointment! The efficiency
that the amp displays is based on it's determination of the
output power. If the amp is looking into a higher Z load,
the indicated power is greater than normal resulting in a
higher efficiency reading.
Oh well, back to the drawing board!
73, Roger
On 8/31/2019 9:32 AM, Roger D Johnson wrote:
It occurred to me that the KPA1500's built in tuner might be of
some help.
Here's my reasoning. If we decrease the drive power to lessen
the output power but keep the same voltage, the amplifier wants
to see a higher load impedance.
Here are the results of some quick tests:
Base line is the amp set for 1000 watts output and eff is 44%
I the set the tuner so the amp "sees" a load of 60 Ohms (actually)
58 Ohms with the values of L and C available. Eff went up to 53%!
I then set the tuner to present a 70 Ohm load (actually 68 Ohms).
Eff went up to 59%!
This was into a dummy load. I realize that, if you're using the tuner
to match an antenna, this could be very complicated.
If Elecraft could figure some way to give us a choice of voltages
for differing power levels, this problem would go away.
73, Roger
--
Scott K9MA
k...@sdellington.us
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft@mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to arch...@mail-archive.com
