The AM Forum

I built up a small interstage tuner yesterday to experiment with the input of the GSB-201. A basic low power exciter like an ARC-5 with 40 Watts out into 50 Ohms (with a split C match) was the first candidate. I could get enough drive to get about 250 Watts out. Not bad. Switching it in and out gave 8 dB of gain or 40W to 250 W. So this was the gain simply driving into 4 811A's - an untuned input. The macshit that I constructed was a simple Pi filter with a tapped 18 turn 2" Dia. coil made of #10 tinned copper. On each side we have a 2 section heavy broadcast capacitor with a switch to add 470 pF as required. So this is a low Z to Low Z affair. I had heard that the addition of a tuned circuit on the input of the GG linear improves the linearity and reduces distortion and harmonics. What I did not expect was what it would do to the drive requirements. Now the thing was driving to 500 Watts with the 40W. The cable between the tuner and amp is short - 3 feet. The cable from exciter to tuner is 10 feet. Anybody have experience with this input tuner stuff?

You can make a grounded grid amp with a tuned input.
Modern ones might be broad band circuits...

I am using a GSB-101, which has basically the same amplifier circuit.    With my fixed tuned ricebox which has no built in tuner, I am using an LDG autotuner between my IC-718 and the GSB-101.  For 40 and 80M this is needed otherwise the drive is too low to get full output from the linear.   With the tube transmitters that have a PI-output tuning, there is no tuner needed.   The GSB-101 should be heavily loaded on AM and the drive power reduced to that the output carrier is 175W.  The main difference between the 101 and 201 is the cabinet and built in cutoff bias which can easily be added in the 101.

I have a related question.  If you have an exciter with a pi-net output feeding the grid of a plate modulated PA, is the input tuned circuit still necessary?  What if the exciter has a simple inductive-link output?  I'm thinking of a 6AG7 oscillator followed by a 6V6 with no B+ on the coil and the output obtained through a 100 pf mica capacitor into the tuned R/C circuit with a link on the cold end.  Note, I'm not talking about input to a linear amp.

Mike,

This is a new term for me; typo or nouveau-Yiddish?

Yes, a tuned input can/will increase the efficiency of a GG amp. The tube's conduction pulses will be sharper, so higher efficiency.

That ARC5 driving the amp made me sit up and take notice. I have a pair of ARC5s here although I push them a bit harder. I tried the 80M one driving my twin 811A GG linear and there was enuf  of a mismatch I decided it was a bad idea. I have a simplified tuner based on one from the 1940 Radio Handbook (Not ARRL) with  porcupined coil and a variable cap. I use it on my loop mod rigs and had been toying with the idea of trying to use it to match the 80M ARC5 to the unturned input of the twin 811A linear. You post has encouraged me to give it a shot. 
Don

The ARC-5 output which is made for short high capacity wires so low R  high C will match to 50 Ohms if you do a split C match using the loading coil and two external caps, a series 100 pF and Shunt 680 pF, for 80 Meters, for instance. With a short piece of Coax, this will drive your amp better.  If the amp is remote (coax greater than say 5 feet) however, the additional tuned circuit closer to the amp will be the ticket.

Thanks for the tip. I will try it. Coax is long but radio is not cast in cement so I could move it and shorten the coax.
I use a single wire end fed long wire so the stock ARc5 roller induct would load it not bad but not the amp. I have all this stuff in the junk box so will give it a whirl. I run my ARC5s at 750 volts on the PA.
Had to make the PS so figured why not build it for the tubes max rating. Barefoot they aren't bad but on 80m that is still often not enuf.
Don

Yes the roller works great with a longwire that is an odd multiple of a quarter wave. With a half wave you normally need a shunt cap like a 500 pF variable to load it and sometimes you get RF in the shack. That is why I like the split C match or a 1:4 UNUN and a conventional tuner located elsewhere.

With the split C match, you will use 1/3 to 1/2 of the roller depending on the value of C1 which really can be anything from 68 to 150 pF. Make sure you use decent caps with a couple kV rating. Roller Skate caps or Doorknobs would be best but high V ceramics will work. Coupling will be 3/4 to Max.