Pimped Out EHX Frequency Analyzer

[moosapotamus]

This is a work in progress. I've been tinkering with it for a while, now. But I think I've reached a point where some other eyes and minds would be most welcome.
...and, don't we all just love looking at schematics! 

In short, this is the EHX Frequency Analyzer with a TAPLFO and few other mods. I think the power supply is sorted (thank you, Brian!) so, for clarity I've left it out, along with the CV controls for the TAPLFO. As shown, there would be seven pots, one or two rotary switches, 3 to 5 toggle switches and, potentially, 3 more inputs in addition to the basic I/O. So, obviously, some pruning might be on order. Of course, I have some ideas but would be interested to hear others' thoughts.



I think tapping the square wave output from the ICL8038 VCO, in addition to the sine output, is a keeper. It's rougher, harsher sounding than the sine, in a good way. As shown, it is also noticeably brighter sounding than the sine, so I might add a cap somewhere to bring it in line, tonewise, with the sine. I think the triangle wave sounds too similar to the sine wave to make it worthwhile.

The duty cycle (pulse width) of the 8038 can be made variable, as shown. I spent a little time trying to figure out how to get some PWM going on. By just twisting the duty knob while playing, it didn't sound remarkably different than the FM provided by the TAPLFO. So, I think I'd leave that out.

Aside from the LM13700 VCA (U6a) shown bottom right, I have most of this assembled on two breadboards. No surprise, it is a little noisy. I've built a standard EHX FA (madbean Freekout) and it is one of the quieter ring mods I've ever built, played, heard. The faint, fixed frequency of the VCO is almost unnoticeable when not playing. However, modulating the VCO frequency seems to make even the faintest noise or bleed-through stand out a bit more. So, that's why I have been considering adding a VCA. Would this LM13700 config work well? Is it overkill, maybe?

If implementing a VCA, noise gate, it seems typical to use it to silence only the carrier signal. However, when I simply disconnect the VCO from pin 3 of the AD633, I can still hear the VCO at the output, especially with FM going on. I don't doubt that this is at least partially because everything is sitting on two breadboards. But, maybe it would need to go on the final output instead. Are there other thoughts about how to silence this beast? My power filtering is pretty standard, 100u and 100n on both power rails. Maybe more?

This post is starting to get too long, but I'll add one more thought. If the LM13700 VCA makes sense, I will have the other half left over. How about using it as a low pass filter with a pot to control the cutoff, and replacing the existing filter U1b and U3a? Yeah, that could really turn this thing into a beast. Might also change the sound, character. Would hate to loose that crisp, clear bell tone that the stock version is capable of producing. Thoughts on that?

Thanks for making it all the way down here! 

[moosapotamus]

I hadn't really been listening to the clean signal output while working on modifying this ring mod. But, I just noticed that I am getting even more carrier/modulation bleed-through on the clean signal output than on the ring modulator output.

So, I think I need to go back to the drawing board, maybe get a new breadboard and just rebuild it. Will probably leave out the filter next time, eliminating two opamps and simplifying things a little. However, that will still leave an odd number (3) of opamps in the circuit, and that kind of irks me a little.

So, I'm thinking... replace the clean opamp output buffer with a transistor buffer. Something like the LPB-1, which I have used as a clean blend output buffer in a number of other builds I've done. The concern is that this ring mod circuit uses a bipolar +/-15V power supply.

Should, or could the LPB-1 be adapted to run on +/-15V, or would it be fine to just leave it on a single +9v supply?

[antonis]

could the LPB-1 be adapted to run on +/-15V?

Easily..



C3/R6 in parallel with R5 retain original LPB-1 gain..

[moosapotamus]

Nice! Thank you!

[moosapotamus]

Ok. I guess I tried to do too much at once and got a little sloppy with this project. So, I'm going to start over... a little more methodically this time. For re-starters, the simple idea is to just add an Electric Druid TAPLFO to the EHX Frequency Analyzer for carrier frequency modulation, and get that to behave nicely before considering other mods. Aside from it behaving nicely, I was able to do that with the initial fiasco, but carrier/modulation bleed-through became a big problem. My messy breadboard construction was probably part of the problem. But I'm now considering that the power and grounding scheme may have also been less than optimal. So, one thing at a time now...

Going forward, here's my current idea for the power supply...



Gnd-A and Gnd-B would be kept separate. The +5V/Gnd-A would be for the TAPLFO and everything associated with it. The +/-15V/Gnd-B would be for everything in the FA circuit (AD633, ICL8038, Opamps, etc...). I am thinking that 'isolating' the TAPLFO in this way could help minimize bleed-through. Yes/No? I've never really rigged anything up in this way before. So, I guess I'm just looking for a quick gut-check. Am I missing anything? Might there be a better approach to the PSU for this? Thx!