Thanks to all who responded to my recent question about tuned input
for a cathode driven 4CX1500B. Special thanks to N1RJ for taking the
time to investigate why I was getting erroneous and confusing
results from several pi network calculators (see his post on the
subject). Roger saved me before I pulled all my hair out!
The cathode impedance is as I had surmised: 100 ohms with 38 pf in
parallel. I wanted to leave a 50 ohm RF load resistor on the tube
side of the input network to absorb excess drive, so the target
impedance in my case became 33 ohms. For 50 MHz I ended up with 150
pf on the 50 ohm side of the pi network, 0.1 uH series L, and
approximately 130 pf (estimated, I used an an air variable) on the
33 ohm side. Add 38 pf contributed by the tube plus any circuit
strays to the latter value. Q is between 4 and 4.5 which is fine for
this amplifier.
Initial testing of the amp had been done with nothing more than the
50 ohm resistor on the input and a capacitor coupling to the
cathode. Adding the pi network made a big difference. Efficiency and
power output went up significantly. Screen current came into line
with what I usually expect for this tube.
Several people suggested a passive grid driven configuration. I have
a 160 through 10m amp on which I will experiment with that. The 6
meter amp wasn't a good candidate for being grid driven. There were
some unusual constraints due to this amp's former life as a grid
driven 2 meter amp and the grid strip line being integrated into the
tube socket. That would present some challenges getting the grid to
look like it wasn't grounded at 50 MHz. The simple solution was to
leave that alone and drive the cathode. Since I took that approach
the amp could be easily converted back to 2 meters if desired.
73,
Paul N1BUG
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|