Clipping diodes on master volume?

[EricKnabe]

Quick question: how feasible would it be to design a diode clipper circuit that connects somehow to a pre PI master volume, and set it up in such a way so that as the master is decreased, the clipping threshold of the diodes also decreases so that the output stage overdrive (or an emulation of it rather) is always present in the circuit?
Basically, the less the PI and output tubes clip, the more the diodes do.

[FiveseveN]

My first thought is it would be easier to vary the signal to be clipped rather than the threshold, and that would be trivial with a dual-ganged pot. Or just keep the saturation constant and mix it in like this:


But I think you'll find the Fletcher–Munson effect more impactful than saturation. How much are your PI and output tubes typically clipping anyway?

[EricKnabe]

Well I'm not actually sure yet. I'm designing a Marshall(ish) clone and I don't want the master volume to affect the amount of overdrive I get, I just need it to control volume and volume only. So I want something to compensate for the loss of PI overdrive when I turn the master volume down. Or I suppose I could introduce more negative feedback so that the power amp is completely clean, or as clean as possible.

[swamphorn]

Add a master volume control between the preamp and output/power amp. The original control becomes your gain control.

[EricKnabe]

Well that's kind of what I'm talking about. That's what the 2203 did and still, there's less overdrive if you turn that master down because of the output stages clipping. So I'm kinda looking for a way to eliminate that, whether it be the diode clipping or by adding more NFB to make the power amp cleaner and an extra gain stage to make the preamp dirtier.

[swamphorn]

A pair of diodes will give you consistent distortion but they won't give you power amp distortion, which is much more complicated than soft clipping the signal. If it's power amp distortion you want, one way to keep it and turn the volume down is with an attenuator between the power amp and speaker. These range in sophistication from simple high-wattage voltage dividers to reactive dummy loads.

[FiveseveN]

A pair of diodes will give you consistent distortion but they won't give you power amp distortion

Heh, debatable. Being transformer-coupled to a reactive load might give rise to "complicated" effects, but otherwise that elusive PP power amp distortion is just symmetric hard clipping.

Eric, it might benefit you most if you just stuck to one thread pertaining to this project, for starters. At least you wouldn't have to explain the context every time you have a question.
Now, what's the nature of this "clone"? Have you settled on tubes after all or is it going to be solid-state? Do you have access to a "real" 2203? Have you worked on anything like this before? Mostly I'd advise you to build a working 1:1 clone first, before trying to "fix" it with things like fake power amp clipping.

[willienillie]

Heh, debatable.

Not really.  Maybe similar on paper, but diode clipping doesn't sound (or feel) anything like overdriven power tubes.

BTW, do you carry a 5.7?

[FiveseveN]

Are we comparing transfer functions or some ill-defined notion of how an amp "behaves"? I guess a fair evaluation would involve something like strapping some appropriately-high (tens of Volts) threshold diodes across the speakers, or at least across the OT primary. But that's hardly ever what we think of when we talk about SS poweramp distortion simulators and I'm sure it's beyond the scope of this thread.

BTW, do you carry a 5.7?

The Belgian pistol? Since Romania has some of the strictest gun laws in Europe and therefore the world, no. But I've been making my own toys since middle school. Also, I admire it for many reasons but cost-effectiveness isn't one of them, even if I needed and could get a gun.