Tremolo to Panning conversion

[theunrulychef]

Hey all, I'm building a stereo vibro champ sort of thing. Instead of having 2 individual tremolos, I'd rather have a "panning" effect. My thinking is that I'd take the output of the stock vibro champ oscillator, feed it through a cathodyne phase inverter, then to the cathodes of the second triode stage for each channel. In theory (at least mine anyways), that would give me a panning effect, as one side goes into cutoff, the other would be at full volume & back & forth etc.

One thing that I'm trying to wrap my feeble brain around, is what the cathode follower on the stock circuit is actually used for. Should I (A) run the outputs of the phase inverter into cathode followers before hitting the pre amp; (B) run the LFO through a cathode follower before the phase inverter (leaving me with a free triode); or (C) skip the damn cathode follower all together.

I'm wondering if I'm going to run into any pitfalls (signals bleeding between channels etc). Also, if someone can give a decent layman explanation of what that cathode follower is doing there, I'd be much appreciative.

Thanks,
Jay in Philly

[Enzo]

Have you actually listened to that effect for any length of time? Two of my favorite albums are Jeff Beck's Wired and Blow By Blow. BUT. The piano is hard panning left and right with its trem, just as you propose. The thing see-sawing back and forth the whole time makes me seasick. REALLY annoying. I prefer to listen to it in mono.

[theunrulychef]

Have you actually listened to that effect for any length of time? Two of my favorite albums are Jeff Beck's Wired and Blow By Blow. BUT. The piano is hard panning left and right with its trem, just as you propose. The thing see-sawing back and forth the whole time makes me seasick. REALLY annoying. I prefer to listen to it in mono.

Yeah, but were you high?

Seriously though, I've used alot of stereo panning delays & that sort of thing - panning trem pedals & the such. In my more youthful days, my brother & I would set up a big circle of what we used to call "practice amps" back in the 90's (old silvertones, princeton reverb, vibro champ etc, anything cheap & tube under 15 watts). We'd make samples on a bunch of Boss DD5's and Digitech PDS delays, and everything would ping pong around the room - not your standard blues jam. Lots of polyrhythms and drones - I'm sure the term "annoying" would've applied with most people.

Back to the topic. This particular amp will be mainly used for stereo effects, but in a single combo low(er) watt package. If it is too annoying to have hard panning back & forth, I can turn down the depth knob for a more subtle shimmer effect. As with everything I'm building, I'm trying to learn more about tube amp construction & circuitry. Hell, if I just wanted stereo amps ping ponging around, I can plug a pan pedal into 2 amps. That wouldn't get me any closer to knowing how to actually design my own circuits though. I originally thought doing a Magnatone vibrato on one channel would be really cool, like a 12 watt tube JC120. But I realize it's probably way beyond my abilities at this stage to get that circuit working in this amp, so I thought I'd try to figure out how to pan with tubes first. Again, any tips or pointers would be greatly appreciated.

Thanks,
Jay in Philly

[octal]

The cathode follower in the stock circuit is needed as a buffer (much more so than other applications) because in this amp the trem osc needs to be able to source enough current to shift the bias of second gain stage enough to cut it off. The cathode resistor of the second gain stage is 1.5K, and that's a really low impedance load for a tube to drive.

The problem with using a cathodyne PI is that the DC voltages at the top and bottom of the PI are going to be very different, as are the output impedences.

Someone might be able to come up with a more elegant way of doing it, but I would suggest feeding the output of the trem osc (the first tube) AC coupled into a voltage divider. Feed that into a conventional amp stage,(to invert the signal) followed by a cathode follower and then into another 25K trem intensity pot. You would need a double gang pot.

This would be A LOT easier to do with a regular push-pull, bias wiggling amp like a Princeton, as the trem signal is AC coupled to the output tubes and it's a high impedance drive, so no CF would be needed. Trem osc>cathodyne> top into one pair of power tubes, bottom into another pair.

[theunrulychef]

The cathode follower in the stock circuit is needed as a buffer (much more so than other applications) because in this amp the trem osc needs to be able to source enough current to shift the bias of second gain stage enough to cut it off. The cathode resistor of the second gain stage is 1.5K, and that's a really low impedance load for a tube to drive.

The problem with using a cathodyne PI is that the DC voltages at the top and bottom of the PI are going to be very different, as are the output impedences.

Someone might be able to come up with a more elegant way of doing it, but I would suggest feeding the output of the trem osc (the first tube) AC coupled into a voltage divider. Feed that into a conventional amp stage,(to invert the signal) followed by a cathode follower and then into another 25K trem intensity pot. You would need a double gang pot.

This would be A LOT easier to do with a regular push-pull, bias wiggling amp like a Princeton, as the trem signal is AC coupled to the output tubes and it's a high impedance drive, so no CF would be needed. Trem osc>cathodyne> top into one pair of power tubes, bottom into another pair.

Cool, thanks Octal. After reading about that cathode follower a couple times, it's finally sinking in why the extra current is needed. I'm glad you've let me know that the phase inverter idea won't work. I had thought about running the second signal through another gain stage, with the gain set as close as I could to "1", but thought the PI idea might be simpler.

I just looked at the roach trem circuit to see if it would be easier. I see it uses a second gain stage, I'm guessing to get the signal large enough to turn the light bulb on & off? So that wouldn't save me from having to use 2 tubes.

Hmm, more to think about...

Thanks again,
Jay

[loudthud]

A split load phase inverter won't work directly, you would need a cathode follower on each output. You would need to capacitor couple everything but it would work. You might consider using high voltage MOSFETS instead of tubes to lower the cost.

[octal]

A split load phase inverter won't work directly, you would need a cathode follower on each output. You would need to capacitor couple everything but it would work. You might consider using high voltage MOSFETS instead of tubes to lower the cost.

I think capacitor coupling could be problematic in this case- if you cap couple between the PI and the cathode followers, you'd need to bias the cathode followers with a voltage divider, I guess. (Might work.) If you DC couple the CFs off of the PI, the outputs would be at different voltages. AC coupling the output of the cathode followers would be problematic, as you're driving a 1.5K load (the cathode resistor of the gain tube) up and down at 10Hz or something. You'd need huge coupling caps. I'm not saying that it absolutely cannot be made to work, it just seems like there's potential complications.

I think the MOSFET cathode follower is an excellent idea to keep the tube count down.

Nathan

[loudthud]

I would DC bias the inputs of the cathode followers with high value resistors like 4.7meg (make a voltage divider between B+ and ground) and hook up the cathodes just like the Vibro-Champ. Use .1uF or .22uF to connect to the cathodyne. I would also DC bias the input of the cathodyne to insure maximum headroom. About 20% of B+ is the magic point and you have to cap couple that to the oscillator. It's a lot of big caps.

[GibsonLover]

It's nice to see someone else likes stereo - or 2.0 might be a better term. Stereo effects, or separating reverb from your signal, or delays... You get huge, clean sounds with two separete power amps.
Maybe you should look at the Brownface vibrato circuit, like 6G13-A, or any of those. LFO > cathode follower > PI. The last stage mixes the hi pass / low pass separately with each inverted LFO signal. Just change the wiring on the last stage for cathode followers. To experiment, I'd try using IRF820's (high voltage MOSFETS) for the last stages wired as followers, and even the PI stage could be the same device. You could save tons of room and cost - and it's not in the audio path, so just don't tell anyone it's not 'all tube'.

[theunrulychef]

Sorry, kinda dropped out of my own thread here. Been trying to figure out wtf I want to do with this thing before I start building anything & plan it all out for a change. I've got an over-rated PT for this & some extra sockets, so I'm going to go ahead & try to figure out how to do it all tube.

I took another look at the Brownface circuit as GibsonLover suggested. I can't remember why I had decided to originally stay away from that one, but it might be the ticket. I'll have to re-read Tubeswell's current thread about the 6G12a trem & Enzo's replies again to make sure I have a better grasp of what's going on at each point in the circuit.

I was always under the impression that it really just pulses the high & low signals for a complex tremolo, but not vibrato. Now I read some stuff online that people say it does actually pitch shift slightly. It makes me think it might be cool to use this on one preamp section & leave the other one dry for a quasi-chorus effect. I've never heard one of the brown "vibratos" in person that I can remember though. Would it provide enough pitch shift effect to create a chorus sound with the other clean output? Am I correct in assuming the output of the 6G13a circuit would be ok to go directly (coup. cap of course) to the SE 6V6 output stage without the extra gain of the phase inverter (as in the Concert)? There's that unused triode with the inverter...

Doh, gotta get back to work. Thanks again for everyone's help & advice,
Jay