Trying To Mod An Arion SAD-1, Have Problem....

[Paul Marossy]

So I had an idea to add modulation to mine, that wasn't too hard to do but not real worthwhile without longer delay times. While I was at it, I thought I would see if I could lengthen the delay time so I changed the delay time pot and tweaked a couple of resistors, and it sounds like I want it - but the problem I'm having is that I'm getting clock noise on top of my signal with long delay times, and ring modulatorish sounds on the repeats depending on how the controls are set. I know all of that has to do with clock noise. Right now it sounds like I am getting a delay time of something around 500mS or so. I'd say that the clock noise is in the 4-5kHz range if I had to guess. Tried caps here and there, I can only get it down so much, or I get no oscillation from the MN3102 if I go beyond a certain size cap.

So my dumb question is there a way to "overclock" the BBD and not have the clock noise? I am guessing that I am just stuck with the limitations of the IC chips, but thought I'd ask anyway...  

Schematic for anyone else wanting to participate in an exercise in futility  
http://www.diyguitarist.com/Images/ArionSAD-1.gif

[Fender3D]

You are facing the issues involved with Nyquist-Shannon sampling theorem.
The max frequency allowed to pass, when delayed (and sampled) with BBD should be f=Cl/2 (better, in practice Cl/3).
Then, if your clock is 4-5KHz you should apply a signal no greater than 1.3-1.6KHz.
In order to keep clock noise away, LPFilters should be tuned there...

[Paul Marossy]

You are facing the issues involved with Nyquist-Shannon sampling theorem.
The max frequency allowed to pass, when delayed (and sampled) with BBD should be f=Cl/2 (better, in practice Cl/3).
Then, if your clock is 4-5KHz you should apply a signal no greater than 1.3-1.6KHz.
In order to keep clock noise away, LPFilters should be tuned there...

So how would I do that? The only cap that seems to do anything is the 68pF at the delay pot. If I make it too big, no more delay happens. I can't really go beyond 68pF. Smaller values just take the frequency of the whine up but I can never get rid of it. When I do get rid of it, the delay also stops because I assume the oscillator in the MN3102 stops. Or that's my guess anyway.

[Fender3D]

So how would I do that?...

Both C1815GR are the LPFilters, around the net you'll found filter calculators or similar applets;
compute or simulate what stock filters are doing, and find new values for desired frequency cuts (1.3-1.6KHz).

A fine tuning of the 10K trimmer @ MN3205's pins 3 and 4 might reduce whining also.

Though you won't get totally rid of the whining until you don't tune your LPFs

[Paul Marossy]

Both C1815GR are the LPFilters, around the net you'll found filter calculators or similar applets;
compute or simulate what stock filters are doing, and find new values for desired frequency cuts (1.3-1.6KHz).

A fine tuning of the 10K trimmer @ MN3205's pins 3 and 4 might reduce whining also.

Though you won't get totally rid of the whining until you don't tune your LPFs

I'll try that, thanks. I thought of messing with that trimpot too, just wasn't sure if it would do anything. If I can that whining down to barely audible I would be happy. But right now it's on top of everything and it just won't fly...

[Fender3D]

...But right now it's on top of everything and it just won't fly...

you can always fake you got a fake theremin... 

[Paul Marossy]

Yeah... I didn't think of that LPF being a factor, I can see that I probably have to mess with that too. I'll give it a try and report back in a few days.

[Paul Marossy]

Just to bring this to a conclusion, it doesn't really work to make the delay time longer.  

I changed the caps in the LPF, but it doesn't completely get rid of that clock noise. Adjusting the trimpot helped slighty. With these changes it might have been enough to be useable, but the repeats having a ring modulated sound is the deal breaker.

No big deal, was just an experiment really.