Bias wiggle trem w/ cathode bias 6V6s

[Smokebreak]

I've got a cathode biased 6V6 PP power section w/ cathodyne PI, and an extra triode. I figured I'd put a simple oscillator in there. Anyone experimented with where the best point of insertion would be?
I see the GA17RVT inserts at the 220Ks, and also has a 330K to ground? What's that for?
There's a vox that does something similar, but w/o the extra shared resistance. Then there's the vibrochamp, but my NFB inserts at the stage before the cathodyne, so not sure if that's a good spot. Maybe preamp wiggle? I just can't see bias wiggle on the cathode of the 6V6s being that intense, with the cathode bias and all..

[Phil_S]

I am confused here. I own one of these. Nice amp for those who don't know.

This amp already has tremolo. Why do you want to add an oscillator?

I don't think there is an unused triode.
Across the top: Gain 1, Gain 2, Cathodyne
Across the bottom: Reverb input, Reverb recovery, Tremolo.
That is 6 triodes and all there is.

What am I missing? You want to change to a different type of single triode tremolo? Why? IMO, this amp has decent tremolo.

[Leo_Gnardo]

I see the GA17RVT inserts at the 220Ks, and also has a 330K to ground? What's that for?

Ordinarily the output tube grids would be referenced to ground = "grid leak" via resistors say 220K to 470K. The 330K in the schemo keeps a ground reference maintained for those grids while allowing the LFO waveform to be superimposed on a near-zero (with respect to ground) grid voltage, by which it "waves up & down" relative to the outputs' cathodes. As the grids are driven far negative wrt cathodes, bias current is cut choking off amplification, voila! tremolo. If you have some other vibro/trem circuit you'd like to use, no reason you can't apply its modulating signal the same way.

[sluckey]

P/P cathode biased 6V6s respond very well to bias vary tremolo. Plenty of examples out there. Here's an Ampeg circuit that I have built. 12AX7 would work fine too.

[Smokebreak]

I see the GA17RVT inserts at the 220Ks, and also has a 330K to ground? What's that for?

Ordinarily the output tube grids would be referenced to ground = "grid leak" via resistors say 220K to 470K. The 330K in the schemo keeps a ground reference maintained for those grids while allowing the LFO waveform to be superimposed on a near-zero (with respect to ground) grid voltage, by which it "waves up & down" relative to the outputs' cathodes. As the grids are driven far negative wrt cathodes, bias current is cut choking off amplification, voila! tremolo. If you have some other vibro/trem circuit you'd like to use, no reason you can't apply its modulating signal the same way.

Ah I see. So in non trem mode, surely that 330K would affect the tone a bit, eh? Letting more signal through?At least as much as 100K vs 220K leaks go...or a 500K gain pot vs 1M, etc.... small, perhaps subjective change, that is.
The Jet schematic Steve posted(thanks!) , for instance, doesn't have the shared reference....hmmm...good time for an experiment!
Phil, sorry for the confusion, this is a different amp I'm building than the Gibson. I was using it as reference.

edit : duh I see in the gibby there's not a ground reference at all, if the 330K isn't there. In the other designs there's a depth or intensity pot, with a grounded leg. neato

[tubeswell]

I see the GA17RVT inserts at the 220Ks, and also has a 330K to ground? What's that for?...

If you replace that 330k with a 250k or 500k pot, you have a depth control.

Without the 330k, the LFO signal would be clamped by the ground return (i.e. you'd still need to have the pair of 220k ground-referenced - because they are grid leak resistors - in order to avoid the output tubes from going into thermal runaway from excessive grid current)

[Phil_S]

Phil, sorry for the confusion, this is a different amp I'm building than the Gibson. I was using it as reference.

Oh! I didn't read your OP the way it was meant. It happens. Now I see what I missed. I said that I must have missed something. If I had been paying attention, I'd have realized it when you said it's a pair of 6V6's. Thanks for clarifying.

The GA17-RVT tremolo works great. Give it a try!

[Smokebreak]

I sure will Phil! Something I just now noticed is that all 3 cap oscillators are basically the same, they're just always drawn so differently, and I never paid quite enough attention to notice. Now I see it.

http://www.valvewizard.co.uk/trem1.html

This one is feelin pretty good so far. I'll put the trem in on V2b on the term strips.
Also of note is that on this Musical Power Supplies PT275, the 6.3V 3A winding is 20g wire. No need to struggle with those big cables people!
I decided to save 8 bucks on the imp selector but spend a dollar on an extra speaker jack and a mismatch

[Smokebreak]

Got this fired up. Couple of observations and a recap : 3 stages with cathodyne PI. 330V plates on 6V6, I've got 20V across a shared 300R for the outputs. I figured I'd start a little cold to give the trem some room.

The trem is definitely present, but would be what I would call weak. I actually used this vox schematic, but the out put of the signal I used 470K > .1U >500K pot instead of the Vox arrangement. Speed works fine, too. it gets reeeeealy fast if necessary. So I need to modify it. What is interesting is that when I power the amp off, and the caps drain, and the sound fades, I can hear what would be an appropriate intensity of trem. Hmmm. My supply is 275V btw to the oscillator.

The other strange thing is that i fired up with a fresh pair of JJ 6V6, and the amp hummed pretty badly, with just the outputs in. My PT 6.3V has no CT, so I've got 100Rs at 20VDC referenced at the Cathodes of the 6V6s. Also, 1 tube "blue-plated", then appeared to be OK. I swapped in another set of tubes, and the amp hums on power up for 2 seconds, then quickly fades into being dead quiet, as far as the noise floor goes.

[pdf64]

6V6 will need ~twice the trem amplitude as EL84; the Cambridge cathode voltage is 10V, yours is 20V.
So bring the limiting resistor down more, eg 100k, or nothing if that still isn't intense enough.

[Smokebreak]

Thanks man that's exactly what I did! I also upped the pot to 1M and now we have some serious wiggle!

[pdf64]

Great, but 1M seems to be pushing it.
The grid circuit resistance limiting value in cathode bias is 500k http://www.mif.pg.gda.pl/homepages/fran ... /6V6GT.pdf
So when shared, it would be 250k.
Maybe bring the grid leaks down to 100k and use a 250k pot?

[Smokebreak]

Yikes I didn't even think about that. In an effort to smooth out this PI, I also added 47K stoppers to the 6V6 grids, which is also additive to that limit.
So no limiter, 100K leaks , and a 250K(which I believe should be linear) pot is working out great and the trem sounds fab.
Speaking of smoothing the cathodyne, you know how the trick is to put a large stopper at the grid? I put a 1M pot there, to dial it in and there definitely is a little loss of highs between 0 and 470K, and even 470K > 1M. I could hear it in the noise floor of a cranked amp with single coils. Nonetheless, this amp needs a 1M there and it's not harsh.

[trobbins]

Notice that they try to add extra high frequency filtering to reduce oscillator harmonics as much as possible, as they easily extend in to the audio range. The oscillator has a 680k output signal resistor to provide one additional HF filter. The output stage cathode bypasses are not huge, to provide another filter. An there is no feedback around this output stage. The speaker also shouldn't have fantastic bass extension.

Adding addition R-C filtering lowers the depth capability, and again interferes with grid leak resistance, but some amps are known to add a few extra filter stages.

[SeppoL]

P/P cathode biased 6V6s respond very well to bias vary tremolo. Plenty of examples out there. Here's an Ampeg circuit that I have built. 12AX7 would work fine too.

Great!
That would be funny to do. I have 6SL7Gt:s and wanna build ampeg.
Hovewer, can i build it with ppimv?
Does PPIMV master volume work with bias vary tremolo?
Or with vvr?