Homebrew AM Transmitter

I started this Homebrew AM rig several years ago (and still have not finished it) - It will use a pair of 4-125's modulated by a pair of 4-125's. 

The Modulator deck was built first....


Wiring finished, started testing the low level audio stages. 

First problem was 60Hz originating in the first stage. This was solved by grounding the metal sleeve of the tube. 

The phase inverter is not working to my liking. I'm using an ARRL handbook circuit (self balancing phase inverter) with a pair of 6SJ7's. I'm probably going to change it out for a conventional triode phase inverter.

The 4-125 filaments are wired and look pretty cool when they're lit up! I used a string of thermistors in the filament supply to bring up the filament voltage slowly - this works great and should extend the life of the tubes. 

Once I solve this problem I'll apply HV and see how much audio I can get. I'm a little worried that the mod transformer match is not optimum and I might need to run the modulator at a higher voltage (close to 3KV).


The Modulator deck is just about finished now.

I spent a lot of time playing with the phase inverter. I gave up on both the original circuit and the triode based design, both demonstrated unacceptable distortion on one of the two outputs. I almost went with a transformer, it worked great but I was picking up low level hum (60Hz) that I think was coupling from the power transformer. Carefully orienting the driver transformer minimized this but I could not get it low enough to my liking. I ended up using a TL082 dual OP-Amp! it drives the 6SJ7's with about 3V pk-pk and works beautifully. I mounted it inside a cleaned out metal tube so no one will ever know! When I started this project I would never have dreamed of using a modern day OP-Amp but I'm a bit of a perfectionist when it comes to performance.

Those are the screen dropping resistors and regulator tubes to the left of the modulation transformer. Glass window to view the tubes.

Next step is to apply HV and evaluate performance! 


Initial test results were very disappointing but after some work it seems to be working quite well now! 

First problem was the screen voltage was too low (450V). I have increased it to 600V now. l also increased the current into the screen regulators to about 30mA. I was running about 20mA and was dropping out of regulation on voice peaks (still does if you hit it hard).

My mod transformer is not going to allow me to run modulator and PA at same voltage. I'm going to run 3000V on the modulator and about 2250 on the PA. I'll share a 3000V supply and drop to 2250 with a series resistor. You can see my test bed below (don't try this at home). I used a bank of resistors to simulate the PA load. I have a 4uf 4KV cap on the end of the dropping resistor (very important). I can only crank my existing Power Supply up to 2700V so I'll need to change the transformer out or add a second one to boost it up a bit.

I'm probably going to change the PA to run 2 x 4-125A's as well. I'm a afraid that a single 4-400 will not load up to ~8K ohms at 2250v on the anode. 

The tubes need to be cooled and I have been experimenting with different fans. I never really thought about this when I started. Anyhow I think I have something that will work (you can see the cardboard over the mod tubes below). Next step is to make it permanent (I have some glass fiber panel that I will use).


The Modulator deck is finished - at last!

Here's a look underneath. I added a home made spark gap across the mod transformer, it is visible on the left below the filament transformer. Screen regulators and bypas cap are visible below the spark gap. Right section is the low level audio stages and power supply. Lower center is the T-R relay that mutes the RX, turns up the fans in TX and enables power to the HV power supply, antenna relay and RF deck.

Note that I know have 2 Millen HV connectors. 3KV for the modulator and about 2.4KV for the RF deck (mod xfmr secondary).

I added a cover over the Modulator output tubes.  A fan draws air across the tubes. I lowered the screen dropping resistor - visible below the screen regulator tubes (4 x OA2).


Looks pretty cool!

These are Phillips 4-125's. They look a bit more substantial than the Eimacs.

On to the HV Power Supply  now! I picked up a new (to me) old HV transformer that looks good for about 2KW. Should yield a steady 3KV for this project!


Not too much progress to report. I have started on the PSU and should have it finished in a couple of weeks or so. It's going in a BUD cabinet to which I have already installed a set of wheels. I'll sit the modulator/RF deck on top when it's all finished.


I finished up the power supply this weekend. Regulation looks pretty good; 3KV at 100mA load, 2.9KVat 350mA and 2.8KV at 650mA. Its a choke input with a full wave (bridge) rectifier. 32uF cap with a 2K series R and additional 24uF cap for the RF deck HV (remember I had to drop PA voltage because of mod transformer ratio).

I included a step start circuit to limit inrush current and a delay circuit built with a 555 timer ensures that HV is not applied until mod and RF PA tube filaments are good and hot. An external trigger to the 555 will disable HV in the event of high VSWR or other alarm conditions. The front panel meter indicates modulator HV and RF deck HV (switch on lower right). The main on-off circuit breaker is in the center. Indicators show power (upper left), high voltage (upper right) and HV disabled (lower left).


After finishing the PSU and modulator I got sidetracked! First by a 75A2 restoration then by a KWS-1. It's on my list and I will finish it one day!