Gated sound in a tube-modded Big Muff

[kissack101]

I modded my USA EHX Big Muff by replacing the clipping diodes with two 12ax7 tubes (and also converting it to true-bypass). The problem is now the output sounds gated, especially when either the sustain or volume knobs are turned down.

I was hoping someone here might be able to suggest why this is happening? I''ve checked and re-checked the wiring and everything looks fine. The adaptor i'm using is rated as 300ma so I was wondering whether the draw from the heating elements in the tubes could be somehow starving the rest of the circuit? Or could it be that the input signal needs to be boosted before the first clipping stage?

I don't know a great deal about tubes or electronic theory which is probably why i'm having as much trouble with this as I am. Any suggestions on possible solutions would be greatly appreciated.

Adam.

p.s. when the sustain and the volume are cranked the fuzz sounds really sweet! Once i've got it sorted i'll post some pics in the pictures! thread...

[earthtonesaudio]

How are the tubes connected?  Can you post a schematic?

I'm no tube expert by any means, but most people I've talked to have advised against using tubes as clipping diodes because they operate at such high voltages.  The output of the effect will barely cause the tubes to conduct, let alone get anywhere near the clipping threshold (or so I've been warned).

If it was me, I'd record it as-is, then take the tubes out of the circuit completely, then record again and do an A/B comparison.  My hunch is that you're hearing the circuit more so than the tubes.  But again, I don't know much about tubes.  Maybe someone who knows more will reply. 

[kissack101]

I just removed the 4 clipping diodes from the board and replaced them with the 12ax7s, connected up as detailed in this thread:  http://www.diystompboxes.com/smfforum/index.php?topic=41416.0 with one tube for each pair.

I was thinking that the forward voltage (?) of the 12ax7's might be the problem which is why I was thinking that a booster in front might help sort the problem, but as I don't really understand the theory I'm just guessing really...

[kissack101]

...anyone?

[earthtonesaudio]

I'm right there with you, trying to figure this out.  I recently stumbled across a decent sized stash of tube diodes, and I've been toying with the idea of incorporating them as clipping diodes (along with the AMZ "saturation" controls). 

I've talked it over a bit with my boss (an experienced electronics tech and guitar guy as well), and here's the gist of what he had to say:

Solid-state diodes work well for clipping because they have that "knee" where they don't conduct, and then suddenly they totally conduct (simplification).  This gives you hard-limiting and a waveform with the tops lopped off. 
Tubes, on the other hand, have a nice, gentle slope (fairly linear) from 0 volts up to hundreds... not exactly what you want in a "hard" limiter.

This pretty much squashed my hopes for "tube-like" clipping... for a while. (Hey, if this was easy, everybody would be doing it)
...Until he mentioned that tubes are non-linear near cutoff.  So rather than try to bias them into saturation (requiring tons of power), just bias them into cutoff.  The sound will probably be quite different from the clipping/clamping diode sound, but it might be usable.

I also started another thread on this topic, and got a good explanation of solid state diodes from R.G.  Search the forum, or my posts.

[kissack101]

Yeah I noticed your other post, still trying to get my head around the concepts involved though, I'm hoping someone much more knowledgeable will be able to help me (us?) out on this one, i'm not sure what 'biasing them into cutoff' would involve? Instead of 0v ground having some maybe a higher voltage as "zero" volts and running the tubes at a higher voltage still? Problem is because I don't have the theortic grounding i'm just guessing, might have to hit the seach function I think...

[earthtonesaudio]

Here's my basic knowledge on biasing into saturation vs. cutoff.

In the regular way of using clipping diodes, in order for them to clip, you have to feed them a large enough voltage to exceed their forward voltage drop (approximately). 
If you were feeding them a large signal, they would be clipping a lot, and you could say they were biased into saturation.

Biasing into cutoff, in this case, is where you just feed them a very small signal, which just hovers around zero (as far as I can guess, anyway).

With solid state diodes, this would not sound like anything.  But with tubes, they are very non-linear at small signal levels, so by feeding them a tiny signal, you'd get some sort of signature tone out of them.  Might not sound good, but it would be unique (I think).

So for your Muff, I would suggest putting a volume pot before the clipping diodes at the end of the circuit, to dial the output down real low.  That might do something, but of course you'd have to amplify the signal again to be able to get a good output from it.
As you can probably tell, at this point I haven't really found a satisfactory explanation for what we're looking for... but I'm still looking.

Let us know if you get it doing something cool.


p.s. I have a feeling that the gated sound might have something to do with the diodes in the feedback loop, not the ones on the output.  Hence the suggestion to mess with the output clippers.

[earthtonesaudio]

http://music-electronics-forum.com/showthread.php?p=47663

4th post says:

Whenever the signal exceeds 0 V, they conduct. Therefore, putting one forward and one backward results in your signal slamming between +0V and -0V...in other words, you get no signal -- silence. That's not what you want at all.

I take it that you were getting more than zero volts out... as a "gated" sound is just a tad different from "silence."

Still no answer... back to Google I go.

edit: farther down, same post is a schematic for the Fender Pro-Roc 700.  It's using a twin triode as back-to-back diodes to ground, like you are doing...
But it's funky...

First off, the grids are tied to the plates. (Did you hook yours up like this?)

V1a has its plate and grid tied to signal, and its cathode goes to +2 Vdc.
V1b has its cathode tied to the signal, and its plate/grid get -2 Vdc

This looks weird to me, because all the circuits I've seen have the positive voltage going to the anode/plate, and the cathode is always at a lower potential.
Maybe there is something to reverse-biasing them to get them to clip right.  I'm clueless.