Delay runaway volume clamp: No, seriously, how is this done? :/

[midwayfair]

Can't for the life of me get this to work properly, or maybe it just doesn't work like I expected it to. I've got a delay on the breadboard right now and I put back to back diodes from every point of contact to ground and between every audio pin of the 2399 and none of them prevent the volume from just spiraling out of control. I did manage to find one or two spots where the delay still worked and one spot where it seemed to shift the point of oscillation (but it did nothing to curb the oscillation volume)

So ... (a) where exactly do I stick these diodes on, say, the rebote, to clamp the volume, and (b) does it actually put a ceiling on the volume, or does it just slow it down? If it just slows it down, I might need to find something else.

One thing I know for absolute certain works -- and this is probably what I'll end up doing if I can't get the diodes to work -- is shorting lug 2 of the feedback pot to ground with a trim pot in series. This lets me dial it in for infinite repeats instead of oscillation. But I thought it would be cool to actually get the sound of oscillation without volume increase.

There's one other thing I know probably works, but it involves adding a compression stage to force the repeats to compress their own signal. Not sure I want to go through all that.

[the3secondrule]

Check out the build docs for the echobase. I'm pretty sure it's done with back-to-back diodes across the feedback pot?

[midwayfair]

Check out the build docs for the echobase. I'm pretty sure it's done with back-to-back diodes across the feedback pot?

Yeah, it's lug 2 >> ground there. But it doesn't seem to do much of anything. It moves the oscillation later in the pot's rotation but doesn't actually clamp the volume.

I think this only gives the illusion of taming the volume runaway.

I mean, it's better than nothing.

[hammerheadmusicman]

Is it the whole feedback loop thing where it just gets louder and louder until it sounds like eventually it's going to set your amp on fire?

I had the above problem with both of my cave dwellar builds, what i had to was put a resistor in series with the top of the pot, at the point just before the volume ran away, so the end of the pot's travel was totally usuable, not sure if that's what you are asking, but that was my solution to that problem, i'm guessing if there is a better solution, it will appear on this thread, and i will hear of it soon

Good Luck Jon

George

[the3secondrule]

wondering if diodes with a lower forward voltage would make a diff? Ge's, or Bat41's or something?

Check out the build docs for the echobase. I'm pretty sure it's done with back-to-back diodes across the feedback pot?

Yeah, it's lug 2 >> ground there. But it doesn't seem to do much of anything. It moves the oscillation later in the pot's rotation but doesn't actually clamp the volume.

I think this only gives the illusion of taming the volume runaway.

I mean, it's better than nothing.

[madbean]

1n194s off the wiper of the feedback pot will clamp the feedback volume, although not all the way. I have this option on the new Zero Points, but it is not totally ideal. Lower forward voltage diodes will clamp it further, but choosing ones with too low of a Vf might cause some unwanted clipping under with lower feedback settings.

One thing you could try is some lower diodes like the BAT41 or GE an put a 50k pot in series to ground, wired as a variable resistor. Set your feedback pot to oscillate and play with the new pot until you get the right amount of clamping on the volume. Once you find the right spot, then substitute a fixed resistor for the pot.

On my old Rebote build I think I did something similar to this. I think I also had a cap either in series or parallel to the diodes so there was a bit of a swell on it. I don't really remember exactly how I did it, but it worked (I don't have the pedal any longer).

Or, if you want to get really fancy, use some kind of transistor based switch in conjunction with shorting the feedback pot via a momentary switch. I don't know exactly what this would look like. Maybe something like this. Here the switch connects to 9v, charges the cap and the resistance across Q1 goes down to nearly zero Ohms. So, this (nearly) shorts the two lugs of the FDBK pot together with C1 being the timing cap. I think. Anyway, R2 and R3 would be adjusted to taste.

Point being, this type of idea would give you some kind of control over the volume swell. I think. I should try this out.

[midwayfair]

Or, if you want to get really fancy, use some kind of transistor based switch in conjunction with shorting the feedback pot via a momentary switch. I don't know exactly what this would look like.

This looks like an awesome idea. I've still got some room on the breadboard, I'll try it out as soon as I can make some noise.

[madbean]

You could also put in a 10k-500k trimpot in parallel with the 10uF to have some control over the charge time, I think.

[crw414tele]

I know this is an old thread but I,m looking for a solution to limit the volume  of oscillations too and have tried the diodes and not had much luck. I'm wondering if the transistor method is any good ?
thanks

[midwayfair]

I know this is an old thread but I,m looking for a solution to limit the volume  of oscillations too and have tried the diodes and not had much luck. I'm wondering if the transistor method is any good ?
thanks

I think I couldn't get it to work, but I might have done something wrong.

The LED mod on pin 7 in the Hamlet Delay in combination with a trimpot to dial in the repeats actually ended up being my solution to this problem when doing the infinite repeats footswitch. The LED clamps the maximum output of the PT2399 and the trimpot keeps the repeats from slamming too quickly. Doesn't get all washy and distorted, though.

[crw414tele]

Thanks , I ended up using the trimmer method and found that it works well enough for what I need. I may even use a pot on the outside as I'm doing it on a multiplex and found the feedback/oscillation works differently on each of the delay types.