Crazy idea: two different headroom (not just power) levels on a footswitch in a single amp

[Badside]

So yeah, this is a bit over the top, some would argue totally unnecessary.
But I was talking with a guitarist friend who is running an amp head that I believe is sort of a Blackface front end into a pair of cathode biased EL84 (haven't opened it up yet, but it's 2 12AX7 and 2 EL84 into a 8.4k OT, and it has Bass and Treble control and no MV).
He loves this amp for lead (and it does have a lovely compression at band volumes) but he also wishes he could get more headroom for rhythm work or clean tones.

This got me thinking... Now, switchable power levels is not a new thing, it's been done before. Either through a Pentode/Triode switch, by changing between Fixed and Cathode bias, by changing the plates or screen voltages, etc. But every amp I've tried with multi watt options always sounds the best on full power. Because all these strategies fail to address one big difference between a small and a big power amp: the output transformer and the power supply!

When you switch a 50W head into a 10W mode of some kind, you're still going through a 50W OT and getting the full capacitance of the 50W power supply.

Yes, it distorts at a lower volume, but it just turns the power amp into a distortion box. It does not get you that "amp is struggling to keep up" lovely compression where single note lines because as large as power chords.

Amplifier switchers are a thing, they're actually very simple: see if you mute one amp then you can just send it's output to ground (or some largish resistor) because there is no sound coming out. The OT only carries AC, not DC. So if the amp is muted, the OT barely cares what's on the other side. A ground reference is preferable of course, but at best there will be a tiny low level noise going through. There's no need to dissipate watts of power.

What if I translate that into a single amplifier, with a shared preamp and power transformer, but two different power amps. A low power cathode bias EL84 pair (like his current amp), and a higher power fixed bias 6L6 or similar power amp.

So here's where I'm at (ignore the preamp for now, I'll tweak that later)

So what we have is a low-ish voltage PT with high current capability. Using the 125VAC primary, B+ should be around 345-350VDC (based on my 272FX getting 375). This is a nice voltage for 5881, so reconducting the power amp from my "Brownvibe" build for the high power side. Should net around 30-35W with a firm low-end. My Brownvibe gets a lot louder than I expected it to be (but it does have an extra 30VDC to work with)

Now for the low power side, take that 345VDC and run it through an EZ81 to bring the voltage down to safe levels for the EL84 and add the desired "sag". Combined with much lower filtering.

How the switching works:
Two relays are necessary, one of which has to be rated for high current (around 6A if the high power side is ran at full tilt)

Relay 1A - Takes the preamp out and either send it to the High Headroom PI or the Low Headroom PI (both PIs have a reference to ground to prevent popping)
Relay 1B - Ground the PI input of whichever side isn't being used (this is to ensure almost nothing comes out the other end)

Relay 2A - Takes the OT output of whichever side is in use and sends it to the speaker output
Relay 2B - Ground the OT output of whichever side isn't in use (again, because that power amp is muted, there is no AC to dispose of)

CHALLENGES:
What I'm worried about though is that during the switching, at some point the speaker output is lifted, and it's at that point a pop is most likely to occur which could send a big jolt through the OTs. That's why I put a 120R resistor across each OT secondaries, so they never go full open. But unsure if this is enough...

Maybe if I can find a "make before break" relay, that could work (the OTs would be in parallel for a fraction of a second, but that is not really an issue).

If I put two preamps (which might be preferable actually, but adds a 9th tube), I could just shunt to ground one preamp at a time without having to hard switch the output, which would prevent any popping.

Also, I really wish I could get the High Headroom side into the 400+ volts range, but that would require dropping a lot more Vs for the Low Headroom side, which means dumping a bunch of heat. Short of getting a custom wound PT (or putting in a 2nd one), I guess I have to make that compromise.

Last crazy idea: kinda want to figure out a way where if I plug in a 2nd cabinet, then I can use both sides at the same time (switch between just Low and Low + High)

 

[2L man]

Very interesting challenge!

Perhaps you could connect both OT secondarys series and then relay short the secondary which is not used?

There are power transformers which have for example something like 325-275-0-275-325 and getting two different outputs is easy.

 

[Peegoo]

Interesting ideas you have there.

Maybe if I can find a "make before break" relay, that could work (the OTs would be in parallel for a fraction of a second, but that is not really an issue).

Because a speaker acts as an inductor, one way to possibly solve the issue would be to include a snubber network of flyback diodes to handle the transient power spike.

 

[Badside]

Very interesting challenge!

Perhaps you could connect both OT secondarys series and then relay short the secondary which is not used?

There are power transformers which have for example something like 325-275-0-275-325 and getting two different outputs is easy.

Hmmm... That's an interesting idea

As for the transformer, I know Hammond (I'm in Canada so that's where I go looking first) makes a couple of those, but none that have the necessary 6A rating on the 6.3VAC tap.

Also, current draw at idle would be around 200mA because both amps are on at all time.

The 270HX is a rare bird that does all of this.

 

[Beebe]

I've heard you can pull out power tubes on a Dual Showman amp to get less headroom. I wonder if you could switch them in and out of the circuit?

 

[Badside]

Interesting ideas you have there.



Because a speaker acts as an inductor, one way to possibly solve the issue would be to include a snubber network of flyback diodes to handle the transient power spike.

That makes sense to me in principle, but I'm not sure where exactly the diodes go.

Are we talking about the diodes that go from anode to ground and cathode to plates that Fender puts on new amps? That would be a simple extra protection that makes sense.

I need to look into what Mesa is doing that mutes the whole amp while switching is happening. Or if I do dual preamps, then I can do something with like transistors and caps where the preamp being switched out is muted instantly but the one being switched in is delayed by a few microseconds.

 

[Badside]

I've heard you can pull out power tubes on a Dual Showman amp to get less headroom. I wonder if you could switch them in and out of the circuit?

I'm looking at more than just reducing output. I also want the sound of a smaller output transformers and a more saggy power supply.

Merely dropping half the tubes is something pretty easy, Mesa does it on a per channel basis. All it does is let you get distortion at a lower volume, but its not the same as a real smaller amp.

I did say this was a bit over the top and probably unnecessary. But that's where the fun is

 

[Peegoo]

Are we talking about the diodes that go from anode to ground and cathode to plates that Fender puts on new amps? That would be a simple extra protection that makes sense.

Yes, something similar to that.

You could use an IGBT or MOSFET to do the switching delay.

 

[printer2]

I am assuming this is for playing live. Will changing OT's really going to give that much difference? I see a dangerous situation, switching under high power. Timing is everything. It is one thing for a factory to muck around until they get a working solution, a fried transformer or two might just be chalked up to a consumable expense in development. For the average person it might be a disappointment.

 

[FenderLover]

Since you are adding a second OT, maybe just cut to the chase and a second PT also. I would think separate preamps for each output would be a plus, but then you are just building two amps in one chassis and switching between them. Not as much fun as a universal design, but allows less interactivity between sides and greater tweakability of each.

I suspect this exercise is why we see multi-preamp designs feeding one power amp, and not one preamp feeding multi power amps. People got lazy and used a power amp to make the preamps louder instead of exploiting power amp characteristics. In the end, a two amplifier system is what we want to emulate, but easier to just use two amps.

 

[schmee]

I've thought of taking an old Twin and making it two amps in one, one more gainy and less headroom, like a tweed deluxe etc. Two OT's. But really:
Take a BF/SF Fender amp with two channels. Set the channels different settings. Add an A/B/Y switch from the guitar output to the channels. Add a volume pedal to the lower headroom channel if you want.... the more you back off the volume pedal, the more it sounds like a low headroom amp.
Or the same rig with two amps.
?

 

[Badside]

I am assuming this is for playing live. Will changing OT's really going to give that much difference? I see a dangerous situation, switching under high power. Timing is everything. It is one thing for a factory to muck around until they get a working solution, a fried transformer or two might just be chalked up to a consumable expense in development. For the average person it might be a disappointment.

I do find that smaller OTs combined with less filtered PSUs make a lot more difference than just turning off a couple tubes or switching to triode. Most power scaling/switching features just sound stiff and dull in the lower power modes, they're just tool to generate more distortion. At least, in my opinion and experience.

But you're right, this is a risky operation. There's some interesting ideas flying around, but my analysis tells me the key is completely muting everything a few nano seconds before and after the switching happens so there's no surge of current with nowhere to go.

I don't have the means to burn a few OTs on my way to figuring this thing out, but I'll keep thinking about it. Mesa does a thing where the little voltage spike when you activate a relay can be used to shunt down a signal just before the relay actually switches (transistors will turn on faster than a relay can switch) so that's one thing to investigate. Someone suggested shunting one OT at a time, which is safer than hard switching their output. Maybe we use octocouplers so there's a smooth transition from one to another instead of a hard switch (both OTs being in parallel for a microsecond should not be an issue).

 

[Badside]

Since you are adding a second OT, maybe just cut to the chase and a second PT also. I would think separate preamps for each output would be a plus, but then you are just building two amps in one chassis and switching between them. Not as much fun as a universal design, but allows less interactivity between sides and greater tweakability of each.

I suspect this exercise is why we see multi-preamp designs feeding one power amp, and not one preamp feeding multi power amps. People got lazy and used a power amp to make the preamps louder instead of exploiting power amp characteristics. In the end, a two amplifier system is what we want to emulate, but easier to just use two amps.

At that point I'd just build a 2nd amp. Stack it on top, use a Radial Headbone VT to switch them.

A shared preamp could still be in a play, e.g. Let's say a Super Reverb combo, add an FX loop. Build a lower power head that is just a power amp (with an "FX loop return" type input) and put it on top of the combo. You can use the Radial Headbone to switch between them while always using the preamp from the Super Reverb.

 

[FenderLover]

Yeah, I didn't want to sound like I was trashing your idea, because I'm doing something unconventional as well. @Peegoo has posted a photo of an SE tweed with two 8's fairly recently, and I'm currently doing something similar, but with dual SE power sections (dual volumes) and dual 8's in a V-front cab. It all fits nicely on a 5E3 chassis. Will it be a waste of time? I'll find out soon enough, but it's all fun to me.

 

[Peegoo]

it's all fun to me.

That is why I do this stuff!

 

[Ricky D.]

I’m just a simple minded old guy.

The challenge is an interesting technical exercise, but awfully complicated.

Why not just run two amps with an A/B switch?

 

[Timbresmith1]

So yeah, this is a bit over the top, some would argue totally unnecessary.
But I was talking with a guitarist friend who is running an amp head that I believe is sort of a Blackface front end into a pair of cathode biased EL84 (haven't opened it up yet, but it's 2 12AX7 and 2 EL84 into a 8.4k OT, and it has Bass and Treble control and no MV).
He loves this amp for lead (and it does have a lovely compression at band volumes) but he also wishes he could get more headroom for rhythm work or clean tones.

This got me thinking... Now, switchable power levels is not a new thing, it's been done before. Either through a Pentode/Triode switch, by changing between Fixed and Cathode bias, by changing the plates or screen voltages, etc. But every amp I've tried with multi watt options always sounds the best on full power. Because all these strategies fail to address one big difference between a small and a big power amp: the output transformer and the power supply!

When you switch a 50W head into a 10W mode of some kind, you're still going through a 50W OT and getting the full capacitance of the 50W power supply.

Yes, it distorts at a lower volume, but it just turns the power amp into a distortion box. It does not get you that "amp is struggling to keep up" lovely compression where single note lines because as large as power chords.

Amplifier switchers are a thing, they're actually very simple: see if you mute one amp then you can just send it's output to ground (or some largish resistor) because there is no sound coming out. The OT only carries AC, not DC. So if the amp is muted, the OT barely cares what's on the other side. A ground reference is preferable of course, but at best there will be a tiny low level noise going through. There's no need to dissipate watts of power.

What if I translate that into a single amplifier, with a shared preamp and power transformer, but two different power amps. A low power cathode bias EL84 pair (like his current amp), and a higher power fixed bias 6L6 or similar power amp.

So here's where I'm at (ignore the preamp for now, I'll tweak that later)
View attachment 1062821

So what we have is a low-ish voltage PT with high current capability. Using the 125VAC primary, B+ should be around 345-350VDC (based on my 272FX getting 375). This is a nice voltage for 5881, so reconducting the power amp from my "Brownvibe" build for the high power side. Should net around 30-35W with a firm low-end. My Brownvibe gets a lot louder than I expected it to be (but it does have an extra 30VDC to work with)

Now for the low power side, take that 345VDC and run it through an EZ81 to bring the voltage down to safe levels for the EL84 and add the desired "sag". Combined with much lower filtering.

How the switching works:
Two relays are necessary, one of which has to be rated for high current (around 6A if the high power side is ran at full tilt)

Relay 1A - Takes the preamp out and either send it to the High Headroom PI or the Low Headroom PI (both PIs have a reference to ground to prevent popping)
Relay 1B - Ground the PI input of whichever side isn't being used (this is to ensure almost nothing comes out the other end)

Relay 2A - Takes the OT output of whichever side is in use and sends it to the speaker output
Relay 2B - Ground the OT output of whichever side isn't in use (again, because that power amp is muted, there is no AC to dispose of)

CHALLENGES:
What I'm worried about though is that during the switching, at some point the speaker output is lifted, and it's at that point a pop is most likely to occur which could send a big jolt through the OTs. That's why I put a 120R resistor across each OT secondaries, so they never go full open. But unsure if this is enough...

Maybe if I can find a "make before break" relay, that could work (the OTs would be in parallel for a fraction of a second, but that is not really an issue).

If I put two preamps (which might be preferable actually, but adds a 9th tube), I could just shunt to ground one preamp at a time without having to hard switch the output, which would prevent any popping.

Also, I really wish I could get the High Headroom side into the 400+ volts range, but that would require dropping a lot more Vs for the Low Headroom side, which means dumping a bunch of heat. Short of getting a custom wound PT (or putting in a 2nd one), I guess I have to make that compromise.

Last crazy idea: kinda want to figure out a way where if I plug in a 2nd cabinet, then I can use both sides at the same time (switch between just Low and Low + High)

A snubber might help get rid of the “Pop”.
Also, perhaps a shorting relay (“Make-before-break” type) can be utilized)

 

[mretrain99]

Some amps have switchable negative feedback resistors for more or less headroom. I’ve thought about having that footswitchable.

 

[Badside]

Some amps have switchable negative feedback resistors for more or less headroom. I’ve thought about having that footswitchable.

My main stage amp has that, it's basically a 2204 JCM800 with a totally indépendant AB763 preamp channel. When I switch to the Fender clean side, NFB is increased and Presence and Resonance controls are defeated. Surprisingly, it tends to only pop the first time I use it after firing up the amp (and not that loudly).

But this does not address what I'm talking about. It does not provide more of the transformer saturating/power supply struggling to keep up thing that smaller amps do.