So I built a preliminary mockup of a 2x 813 class C amp on a sheet of aluminum and used a B&W 850A for the coil in the Pi net. I also used #12 copper for the rf interconnects.
I used the TANKIT program for a plate load of 4750Ω’s and a Q of 15. I also have the east and west coast charts for Pi component values as a reference. The TANKIT program has 14μH for the coil, the B&W in the 80M position measures, for me, at 14.3μH.
I am driving the whole mess backwards with an Icom 746, watt meter in between, and also monitoring the 4750Ω load resistor across the tubes with a scope.
It tunes up on 3.75 Mhz close to the TANKIT values. I check for the -3db points by tuning the Icom above and below 3.75Mhz while looking on the scope for 0.707 x the voltage at resonance. I find a Q of 12.6… 3.75Mhz / delta f.
The problem and question. I am trying to get a general idea of what to expect when I actually build this on a new chassis. Size of the pi coil, padding cap for loading etc. The problem is I have to drastically reduce the inductor, 50%, to make any sizeable increase in Q.
The west coast handbook suggests a Q of 15 to 20 for class C plate modulated amp. I’m aiming for a Q of 18 and cannot get it any higher than 14 on 80M.
I have not tried 40M – 15M. Of course the problem child with 2-813’s, 10M, works fine with the stock coil! Q is around 24 but if I short a few turns of the coil Q will increase greatly, around 37-40 as it should.
Any ideas or suggestions as to what is going on here??
Is my test procedure correct?
Does anyone even pay attention to Q on the lower bands.
Just trying to build a clean amp and learn along the way.
Oh ya the auto tuner in the Icom is off.
Thanks Again
Ted / KC9LKE.
