Good location for reverb mixing resistor in bassman/marshall like preamp?

[elipsey]

I have built a homemade amp that I'm playing around with, and I would like to try reverb in a marshall style preamp circuit. The amp is a cathode biased el84 combo. The reverb tank is driven by a 12at7, with half a 12ax7 driving the return. Does it seem resonable to use a mixing resistor as follows (shown circled in red), if I omit the other "bright" channel?



Would it be better to put the mixer after the tone stack, for example in series with a master volume like this?

[Mike Sulzer]

I have built a homemade amp that I'm playing around with, and I would like to try reverb in a marshall style preamp circuit. The amp is a cathode biased el84 combo. The reverb tank is driven by a 12at7, with half a 12ax7 driving the return. Does it seem resonable to use a mixing resistor as follows (shown circled in red), if I omit the other "bright" channel?

[ATTACH=CONFIG]39835[/ATTACH]

Would it be better to put the mixer after the tone stack, for example in series with a master volume like this?

[ATTACH=CONFIG]39837[/ATTACH]

Reverb is not my thing, but in general with something like this I would want the guitar and the added "effect" to both be affected the same way by the tone control, which means combining them before the tone control.

[elipsey]

Thanks -- it seems like even if it's after the tone controls, the reverb input signal is affected by the tone controls, and, thereby, the output?

I'm a little worried about whether having a small signal into the reverb and a lot of gain after the return will make the reverb more subject to feedback or mechanical noise, and also whether the reverb will distort with the preamp volume turned up high and the master turned down. Surely this has all been tried before, but I couldn't find any account of it...

[Mike Sulzer]

Thanks -- it seems like even if it's after the tone controls, the reverb input signal is affected by the tone controls, and, thereby, the output?

I'm a little worried about whether having a small signal into the reverb and a lot of gain after the return will make the reverb more subject to feedback or mechanical noise, and also whether the reverb will distort with the preamp volume turned up high and the master turned down. Surely this has all been tried before, but I couldn't find any account of it...

I am not understanding the whole picture from your schematics. I think I am missing something.

[uneumann]

You can get decent results in many insert/mix points. My experience has favored tapping and inserted reverb late in the signal chain - the PI input is a good point IMO.

Tapping is relatively easy, but mixing with resistors will cut your dry signal level. Equal resistors will cut it in half. You can favor the dry signal a bit since reverb is usually a smaller signal component, and you can boost the reverb signal level in your recovery amp to a high level. That way it's OK to favor the dry signal in the resistor mix and minimize it's attenuation.

[pdf64]

Does the power amp have an LTP phase splitter / global negative feedback loop?
If LTP and no loop, send the reverb into the 'other' LTP input, in the same way that 60s Watkins / 18 watt and Vox amps used both LTP inputs.
The relative polarity of the reverb wet signal (with regard to the dry signal) is immaterial, as it is delayed etc by the tank.

A Vox type tone cut control (between the LTP plates) would then satisfy Mike's requirement, ie it would affect the combined wet/dry signals.

[elipsey]

I am not understanding the whole picture from your schematics. I think I am missing something.

Sorry if I didn't explain well. The schematics are entirely hypothetical -- I just want to have a design discussion. If I like where it goes, I might try implementing a circuit in an amp I already have, so the configuration of that amplifier probably places a few constraints on what is practical or convenient to actually try. For example, it's already an el84 push pull, so it would be nice to keep that part.

[elipsey]

You can get decent results in many insert/mix points. My experience has favored tapping and inserted reverb late in the signal chain - the PI input is a good point IMO.

Tapping is relatively easy, but mixing with resistors will cut your dry signal level. Equal resistors will cut it in half. You can favor the dry signal a bit since reverb is usually a smaller signal component, and you can boost the reverb signal level in your recovery amp to a high level. That way it's OK to favor the dry signal in the resistor mix and minimize it's attenuation.

Thanks. This amp already has way more gain than is needed to drive el84's, so I'm using a voltage divider at the master volume anyway. This is formed by a resistor in series with the master volume pot. The master volume is the last thing in the signal chain before the PI, so it seems like a good place for a reverb mixer resistor, but the layout is problematic and might require spaghetti wiring. I'm mostly wondering if having the mixer resistor at the volume/gain pot right after v1a (as shown in the first diagram) would be really terrible somehow compared to doing it later in the signal chain. If so, I'll try to make the layout work for a mixer near the master volume, otherwise, it's probably not worth trouble, and I would put it ealier in the chain.

[elipsey]

Does the power amp have an LTP phase splitter / global negative feedback loop?
If LTP and no loop, send the reverb into the 'other' LTP input, in the same way that 60s Watkins / 18 watt and Vox amps used both LTP inputs.
The relative polarity of the reverb wet signal (with regard to the dry signal) is immaterial, as it is delayed etc by the tank.

A Vox type tone cut control (between the LTP plates) would then satisfy Mike's requirement, ie it would affect the combined wet/dry signals.

Thanks. It has an LTPI, with feedback.

This is a cool idea, but I think the layout would be difficult to fit with what I already have, and also, I kind of like having the NFB on this one. I might try something like this on my next attempt if this one doesn't work out though, so it's a nice idea to have in the back pocket...

[uneumann]

Thanks. This amp already has way more gain than is needed to drive el84's, so I'm using a voltage divider at the master volume anyway. This is formed by a resistor in series with the master volume pot. The master volume is the last thing in the signal chain before the PI, so it seems like a good place for a reverb mixer resistor, but the layout is problematic and might require spaghetti wiring. I'm mostly wondering if having the mixer resistor at the volume/gain pot right after v1a (as shown in the first diagram) would be really terrible somehow compared to doing it later in the signal chain. If so, I'll try to make the layout work for a mixer near the master volume, otherwise, it's probably not worth trouble, and I would put it ealier in the chain.

"Really terrible" is probably not going to happen for any option, but you may want to think about whether you want V2 preamp distortion after the reverb or not. If you run clean, there's probably not much difference. If you use gain pedals, or high volume settings (since you have a MV), V2 distortion can get high. It's personal taste to some degree. My taste runs to thinking of reverb as "enhanced room reverberation" so I prefer to put it late in the chain. Ideally, you tap the speaker signal (or mic it) to run to a separate verb/amp. There have been threads on this topic before so some browsing of prior discussions will get you a fuller picture.

Lastly, not clear why layout should be an issue with a shielded wire in/out of the tap point.

[elipsey]

(since you have a MV), V2 distortion can get high. It's personal taste to some degree. My taste runs to thinking of reverb as "enhanced room reverberation"

I agree -- distorted reverb doesn't seem like what I'm looking for.

Lastly, not clear why layout should be an issue with a shielded wire in/out of the tap point.

I know what you mean, but the reverb driver tubes have grid resistors, cathode resistors, bypass caps, coupling caps and a few other bits like the reverb transformer and tank jacks that all have to go somewhere, preferably not too far from the tubes themselves, no?

All of that stuff is closer to the begining of the amp now, and there are other dumb layout things already, like the cathode follower pair will have to be split across two 12ax7, with the reverb driver in between them. Maybe I need to start over

[tubeswell]

I did a '5F6B+' with reverb and bias vary trem a few years back (using 6 pre-amp tubes in all). The reverb circuit was a cross between BF/Allen Old Flame and was inserted between the tone stack and the LTP. Sounded awesome.

[uneumann]

I did a '5F6B+' with reverb and bias vary trem a few years back (using 6 pre-amp tubes in all). The reverb circuit was a cross between BF/Allen Old Flame and was inserted between the tone stack and the LTP. Sounded awesome.

This makes me realize my post above could be misleading... The reverb insert should be before the trem effect. I'm sure it's been said before, but I didn't mention it. If you have a tremolo in the amp, reverb is best inserted before the trem modulation. If the trem is first, the reverb mushes it up and diminishes the trem effect. So the position of a trem effect limits where you can put the reverb mix point.

[Chuck H]

This schem has two "channels". The "British" channel uses a very Marshall/Vox TB type topography. The reverb uses a 12dw7 with the AU7 side as the driver. It works great with just a little less wet mix capability than a BF Fender type circuit. It may be a little trick tracing the circuit via the channel select switches, but it's all here. The "virtual earth" type summing stage is arranged to balance the effect by actually allowing some amplification of the wet signal. Isolation is maintained via the dual ganged reverb volume pot arrangement.

[Chuck H]

Here's the same amp with the "Fender" channel removed so that it's easier to see the signal/reverb interactive circuits. As you can see, if you only use one Marshall/5f6a type channel you can reserve a triode out of the three 12ax7 circuit for summing. That allows this reverb circuit to be implemented by adding a single 12dw7 tube to the chassis. Tremolo on my particular amp, which adds another tube not withstanding.

Cheers