Switchable Power Driver Sections

[TroyBoy]

I've been scouring the forum to see if anyone here has built an SSS clone with a switchable driver section, between a typical PI driver and a Cathode Driver, and can't seem to find anything.

I think it would be a best of both worlds type of build, and would offer a lot of flexibility.

I'm just waiting on my gear to arrive, and I'm thinking of undertaking something like this, so any advice would be appreciated of anyone has given it a shot.

[jfs322]

As a preface, I'm not a tech. But I think it'd be hard to put something like that on a switch because you'd need a switch rated for handling higher voltages than what the standard Carling style switches are rated for.

[TroyBoy]

I don't think switch ratings would be a problem in this part of the amp.

I think i want to switch with a relay running off of an on/off/on standby switch. I think it would make for a neat solution.

[erwin_ve]

I don't think switch ratings would be a problem in this part of the amp.

I think i want to switch with a relay running off of an on/off/on standby switch. I think it would make for a neat solution.

That would result in a huge thunder plop. If you ever measured certain points around the PI you know that is a super sensitive area. I wouldnt recomend it for switching with a relay.

[TroyBoy]

I don't think switch ratings would be a problem in this part of the amp.

I think i want to switch with a relay running off of an on/off/on standby switch. I think it would make for a neat solution.

That would result in a huge thunder plop. If you ever measured certain points around the PI you know that is a super sensitive area. I wouldnt recomend it for switching with a relay.

That pop can be managed with a cap, and this is only going to switched when an amp is switched out of standby, effectively switching the amp "on" in one of the drive modes.

I'm really just asking if anyone has run a dual drive stage in a build, and to hear their thoughts.

[rootz]

Never seen a solution for your idea's. I think your idea is interesting, maybe rather difficult to implement.

Erwin is right, pops might become an issue. Rapidly (dis)charging of caps is mostly the problem. Those caps in your idea would be the coupling caps leading to the CF or power tube grids. Those are connected to the bias supply, so have a grounding point. Thus I'd keep that connection intact and do the switching on the power tube grid side.

There is one very obvious downside when switching the power tube grids to either CF output or PI output: power tube grids will not have bias voltage on them while switching!! I consider switching like this only safe when on standby.

Other than that, SSS then to have 470k ton ground resistors on the power tube grids IIRC, thus forming a voltage divider with the bias supply when driven from the PI plates. This means dropping negative voltage. That has to be accounted for.

I attached a schematic of your idea. I'd say this is for learning purposes only, as there are no fail safes for when the switch fails or the brief moment without bias voltage during switching.

Please note that in both modes you get an approximation of the real deal. In PI mode there is extra loading from the output tube grid leak resistors, in CF mode there is less loading (2x 1meg). Other than that PI coupling caps are bigger than in say a #005, while the bias resistors are smaller (220k vs 330k as can be found in e.g. the AN Steel string sultan, with is supposedly based on the 005/HPD).

I'll say it again: use this for learning purposes. I'd advise against implementing straight away.

[mojotom]

As I’m building an amp with a second tube for CF and still putting down a proper layout I noticed it could be quite easy to design an amp with the CF stage, have a good listen then quickly rewire the amp and bias section to remove that one and try the amp with a regular PI. Maybe going back and forth (you need to take time with those amps and try to play the amp a while with a change, at least for me this is the case).

You should then be able to choose easily between the 2 and the amp will be easier to build. I have to say I tend to remove a lot of switches here and there as I rarely use them once the amp is finished and tweaked.

[jfs322]

.

[TroyBoy]

Never seen a solution for your idea's. I think your idea is interesting, maybe rather difficult to implement.

Erwin is right, pops might become an issue. Rapidly (dis)charging of caps is mostly the problem. Those caps in your idea would be the coupling caps leading to the CF or power tube grids. Those are connected to the bias supply, so have a grounding point. Thus I'd keep that connection intact and do the switching on the power tube grid side.

There is one very obvious downside when switching the power tube grids to either CF output or PI output: power tube grids will not have bias voltage on them while switching!! I consider switching like this only safe when on standby.

Other than that, SSS then to have 470k ton ground resistors on the power tube grids IIRC, thus forming a voltage divider with the bias supply when driven from the PI plates. This means dropping negative voltage. That has to be accounted for.

I attached a schematic of your idea. I'd say this is for learning purposes only, as there are no fail safes for when the switch fails or the brief moment without bias voltage during switching.

Please note that in both modes you get an approximation of the real deal. In PI mode there is extra loading from the output tube grid leak resistors, in CF mode there is less loading (2x 1meg). Other than that PI coupling caps are bigger than in say a #005, while the bias resistors are smaller (220k vs 330k as can be found in e.g. the AN Steel string sultan, with is supposedly based on the 005/HPD).

I'll say it again: use this for learning purposes. I'd advise against implementing straight away.

Hi Rootz,

Really elegant design.

A relay or FET could be used break B+ supply to the power amp should the bias drop, and when switching. This would not only act as a safety device to prevent run-off in case of bias loss, it will eliminate the pop when switching, as B+ has been removed momentarily when switching.

Mesa Boogie, along with many others, use a similar idea to mute signals when switching, so I think it would be easy to achieve with some slight adaptation.

Thanks again.