Adding reverb to a Peavey Classic 20 circuit?

[elipsey]

Hi folks,

How would you approach adding tube reverb to the Classic 20 circuit, in a scratch build? Note that I do not have a Peavey Classic 20 -- I just want to discuss the circuit. If it is reasonable, I would prefer to stick with 12AX7 or 12AT7 for the driver and recovery circuits, and a 4AB3C1B style tank.

How about using R5 or R24 as a mixing resistor, adjusting it's value as necesarry, and perhaps compensating for any change in gain by adjusting a plate resistor?

Cheers

peavey_classic20.pdf

[printer2]

R24, that way you don't have to fiddle with the reverb level if you change the volume.

[elipsey]

R24, that way you don't have to fiddle with the reverb level if you change the volume.

I'm not sure I understand this. R5 is after the volume control between stages 2 and 3 as in a Princeton Reverb, for example. It seems like the volume control VR1 preceeds R5 in the signal path and would affect the rever driver input volume and therefore the reverb output volume as well.

[Enzo]

Exactly. If the reverb feed came off before the volume control, then it would be full up drive all the time. Imagine turning the volune down to 1 or 2 but the reverb remains at 9. Take the feed after the volune control and now as you turn down teh amp, the reverb gets turned down as well.

[teemuk]

Take the feed after the volune control and now as you turn down teh amp, the reverb gets turned down as well.

And at lower volume settings the tank is driven very weakly and the reverb doesn't sound all that great. Thin sound with short reverberant effect.

You want to drive the tank so that you get full bore output from it. It sounds best that way. But while splitting the signal path to parallel "wet" and "dry" branches you also want to keep gain of the "dry" branch fixed. If you don't, every gain adjustment needs corresponding adjustments to "reverb" control for right balance of wet and dry.

If gain of dry signal path remains fixed you only need one control to balance wet and dry signal levels. This can be accomplished in various ways: Typical scheme involves mixing varying degrees of "wet" signal in with the dry. Volume control following in the signal path will control amplitude of the entire signal mix. Another scheme is parallel volume and reverb controls. The scheme can mix wet and dry in various degrees, but also pass 100% complete wet signal. Or signal that is primarily "wet". Third scheme is a mixer that "crossfades" between dry and wet signals. It can only works as balance control between wet and dry signals, but it can not control overall volume level.

[Chuck H]

OTOH all the old Fenders fed the driver tube analogous to the signal level. It's the effect we're use to and makes some sense because with the amp playing smaller and weaker it might sound odd to have the reverb big and fat.?. JM2C

And FWIW I might just use a digital reverb module for that amp. Adding a tube reverb comes with all the cursory problems like filament current. On the Peavey it may be a filament VOLTAGE matter. I don't know if that model uses a series filament string. And then there's the issue of finding a good sounding, non microphonic tank. That doesn't usually happen on the first purchase IMHE. Tuning the frequency range on the circuit to avoid microphonics can also become an issue. As can situating the pan for lowest hum, which you probably won't eliminate completely. Working out a tube reverb added to an existing amp just has a lot more bugaboos. The digital add ons sound good and avoid most of the pitfalls.

[elipsey]

Exactly. If the reverb feed came off before the volume control, then it would be full up drive all the time. Imagine turning the volune down to 1 or 2 but the reverb remains at 9. Take the feed after the volune control and now as you turn down teh amp, the reverb gets turned down as well.

Right. What I meant was that the stated concern seems to be irrelevant to either of my proposed solutions, as far as I can understand.

[elipsey]

OTOH all the old Fenders fed the driver tube analogous to the signal level. It's the effect we're use to and makes some sense because with the amp playing smaller and weaker it might sound odd to have the reverb big and fat.?. JM2C

And FWIW I might just use a digital reverb module for that amp. Adding a tube reverb comes with all the cursory problems like filament current. On the Peavey it may be a filament VOLTAGE matter. I don't know if that model uses a series filament string. And then there's the issue of finding a good sounding, non microphonic tank. That doesn't usually happen on the first purchase IMHE. Tuning the frequency range on the circuit to avoid microphonics can also become an issue. As can situating the pan for lowest hum, which you probably won't eliminate completely. Working out a tube reverb added to an existing amp just has a lot more bugaboos. The digital add ons sound good and avoid most of the pitfalls.

Interesting idea. I assume that by "existing amp" you mean existing design (as stated earlier I don't have a Peavey, just the schematic). If I build this, I'll get a transformer with an adequate filament supply. I have done OK with both the new Accutronics and Belton tanks, which I believe are now made in Korea. I got one from Webber once that I thought maybe wasn't quite as good as those, but it worked well enough. I like to put the tank in foam inside a bag. This seems to keep them adequatley isolated for my purposes, meaning anything short of sloppy wet reverb with the amp cranked to 10.

I do actually have a Belton Digilog module lying around. I tried this in another amp once, mainly as a learning exercise. I got it to basically work after a little bit of trial and error but my result was inelegant and not really a "product", in the sense that I wouldn't have liked to send that amp out into the real world to be used by a player. I built the power supply and op-amp driver circuits point-to-point on a DIP prototyping board with a bunch of jumper wires, and I think it both lacked adequate protection, and also wasn't physically road worthy. It was fun to try, but I ended up pulling it from the amp. If you want to look at couple of pictures, and see my rookie mistakes, here's the thread. Anyway, if I was going to do that again, I think the right way to go would be to cowboy up and design a proper carrier board on PCB. I haven't got around to that yet, but it's on the bucket list.

Anyway, we're talking about a 15 watt combo here, and my personal experience with small combos and reverb tanks has been pretty good. Sure, they can get a little funky at high volumes, but it's just a fun little amp for jamming and stuff. The reverb doesn't have to work for playing stadiums, you know? If it doesn't behave perfectly sitting next to a kick drum with the volume and reverb turned up to 10, the player can turn the reverb off, or use a pedal.

[66 Kicks]

R24 looks like it could be made to work, but you may need to use a two-stage recovery with the NFB coming in like that. Another idea is to tack the classic Fender reverb in front of the Peavey circuit.

[elipsey]

Looks good. Thanks for taking the trouble to make that schem just for me!

[printer2]

I'm not sure I understand this. R5 is after the volume control between stages 2 and 3 as in a Princeton Reverb, for example. It seems like the volume control VR1 preceeds R5 in the signal path and would affect the rever driver input volume and therefore the reverb output volume as well.

Master volume that is. If you want to put it after VR1 then no worries about changing the input volume and messing up the wet/dry balance.

[Chuck H]

Interesting idea. I assume that by "existing amp" you mean existing design (as stated earlier I don't have a Peavey, just the schematic). If I build this, I'll get a transformer with an adequate filament supply. I have done OK with both the new Accutronics and Belton tanks, which I believe are now made in Korea. I got one from Webber once that I thought maybe wasn't quite as good as those, but it worked well enough. I like to put the tank in foam inside a bag. This seems to keep them adequatley isolated for my purposes, meaning anything short of sloppy wet reverb with the amp cranked to 10.

I do actually have a Belton Digilog module lying around. I tried this in another amp once, mainly as a learning exercise. I got it to basically work after a little bit of trial and error but my result was inelegant and not really a "product", in the sense that I wouldn't have liked to send that amp out into the real world to be used by a player. I built the power supply and op-amp driver circuits point-to-point on a DIP prototyping board with a bunch of jumper wires, and I think it both lacked adequate protection, and also wasn't physically road worthy. It was fun to try, but I ended up pulling it from the amp. If you want to look at couple of pictures, and see my rookie mistakes, here's the thread. Anyway, if I was going to do that again, I think the right way to go would be to cowboy up and design a proper carrier board on PCB. I haven't got around to that yet, but it's on the bucket list.

Anyway, we're talking about a 15 watt combo here, and my personal experience with small combos and reverb tanks has been pretty good. Sure, they can get a little funky at high volumes, but it's just a fun little amp for jamming and stuff. The reverb doesn't have to work for playing stadiums, you know? If it doesn't behave perfectly sitting next to a kick drum with the volume and reverb turned up to 10, the player can turn the reverb off, or use a pedal.

I think I underestimated your goal. I took the OP to mean that you were going to build a Classic 20 amp and then just kludge a reverb into it. As it happens you plan to design a reverb into it. My bad for underestimating your skill and intentions.

I agree that R24 looks like it could work as the isolation/mixer resistor. You may need to bump it up to 470k and lose a little more dry signal gain to make the reverb "pop". I might try bumping R24 to 470k, feeding the reverb from the tone stack through a 1n cap and return it through a 470k on the PI side of R24.

There's going to be some loss by bumping R24 to 470k. Not much. You won't get it all back, but most of it by bumping R4 to 220k.

I know this approach was already in the works. JM2C