I recently got lucky and found a fully assembled 9505 Theremax at a thrift store for only a few bucks. It was very crudely assembled, and didn't work at all. After fixing a few soldering mistakes, and replacing Q9 and L5, it worked fine.
The wiring was REALLY messy, so I decided to redo all the wiring. I ran new wires in little mesh harnesses, following the manual as a guide. It then didn't work correctly, sounding distorted, and VCA not working at all.

I then replaced all the transistors and all 4 germanium diodes, and its better now, but still not perfect.
It now sounds incredibly distorted if the main volume is turned down at all. It has to be set to max to sound normal. VCA is VERY hard to tune, and doesn't seem quite right. Velocity, Vol trim and Pitch CV knobs seem to have no effect.

I have quadruple checked all wiring and components for correct locations. Everything looks good to me. I don't have an oscilloscope, Only a digital meter.
I have ordered a new 748 Op Amp, and LM339 Comparator, but wanted to get an opinion here before blindly replacing more parts.

Do these symptoms sound familiar?


Thanks,
JJ

This topic from over in the Technical Bulletins and Assembly Notes, Theremax, section includes a downloadable pdf detailing updates to the circuit over the years.

viewtopic.php?f=11&t=365

Did you use the specified 2N4124 when replacing transistors. If not the biasing might not match up with the substituted ones?

I used a 2N3904 in place of Q9 originally when I had it working perfectly, before re-wiring it. However, since then I have replaced all of the transistors with 2N4124. Ordered from Mouser.

That updates PDF is helpful. I'll look it over later tonight.

If it is of any help with diagnosis, the PCB is labeled 9505C.

Thanks again!
JJ

Here are a couple of pictures on how I have the wiring laid out now. Its difficult to see whats going where, but you can see its tidy, and pulled away from the oscillators.
It was a rats nest before this.

http://www.dashfest.com/theremin1.jpg
http://www.dashfest.com/theremin2.jpg

So reading that document, I realized I replaced 4 CDSH270 diodes with IN34A Germanium's.

Would this cause an issue?

It might be the wiring... You mentioned having had it working perfectly, before rewiring it.

You do have the wiring away from the parts in the oscillator circuit sections, but the wires to the Pitch and Volume Trim-controls are parts of the first and fourth oscillator circuit sections. Them having longer lengths and being bundled with other wires is probably one trouble.

Use this image as a reference for the wiring:

http://www.paia.com/manuals/docs/PTphot ... Repair.JPG

Yes, the CDSH270 diodes are preferred. They give a bit more output for the pitch and volume heterodyne signals.

Using 100pFs at Cs 39 44 and 37, and, adding a 6800ohm on R81-3 to R81-2 provides better waveforms at the Timbre control extremes and improved volume hand-to-antenna response.

Bundling the wiring can increase the tendency for a feedback condition to occur. Adding a 0.1uf to the bottom of the board at solders for the Q6 collector to a nearby ground helps prevent this, but it isn't so likely to occur when wires aren't tightly grouped (and, the audio output level control advanced beyond about 3:00). There's plenty of f-v circuit section pulsing energy and not so much audio signal energy and when things are wide open through the VCA section, it all gets mixed-up.

About the Vol Trim, Velocity, and Pitch CV controls not affecting anything, these probably aren't indicative of the troubles you've encountered. With the volume pair tuned for traditional control response, the Vol Trim doesn't work to change the volume as much as it did for the originally prescribed controller response; however, if you'll set the vol trim about midway and stop with the adjustment of L3 with the background 'pitch' just at the brink of audio/post-audio, you should find that the vol trim works to fine tune the volume about this audio/post-audio region. It's when the pair is tuned just post-audio that the volume cv will be greatest and optimized.

Velocity works to vary the amount of velocity cv that goes to the velocity cv output jack, and, to the VCA section to distort the smooth sounding wave at the ccw Timbre control extreme when the hand to volume antenna control is rapid. The Volume CV control only works to vary the amount of volume cv at the volume cv output connector.

The only difference between the C and D versions of the circuit board are the holes for the i.f. transformers at Ls 1-4. They're not so tight and greatly reduce stress on the legs as the part is inserted making it less likely the wires break off the top of the legs inside the part.

Here's to success!

Thanks so much for the info! I'll work on it more this weekend and come back with results.