6L6 based VVR regulators like Benson uses??

[Jalmeida]

Hello,

I have been intrigued with Benson’s recent re-introduction of the power tube based VVR and was curious if anyone has done this on their builds? I really like the idea of staying all tube sag and all. 

From what I have read the concept was knicked from the Sho Bun amps that used a 6L6 voltage regulator. But could you in theory use any power tube in the boktage operating range? Like an EL84? Or 6V6? Etc?
 

[tubenit]

Yes, a number of us have done this.  VVR and cathode biased amps (el34world.com)


After having done this on a few amps, I've opted for PPIMV instead as a way of controlling volume. Both VVR and the PPIMV each have their own advantages.


I found recording an overdriven tone using Mixcraft Acoustica to be more easily done with a VVR.  This recording was done at an incredibly low volume that one could easily talk over.  There were no pedals used with this recording. All OD is the amp.


https://soundclick.com/share.cfm?id=9488477


Other than recording at very low volumes at home, I found no advantage for my purposes of a VVR over PPIMV. I found the VVR on one of the amps to become noisy after about a year or two and removed it and replaced it with a PPIMV which resolved the noise. So for me there was a potential reliability issue using a VVR.


I think the PPIMV does not sound as good at a really low volume as a VVR does. Below about "5" or so, the PPIMV's on my amps lack something.


With respect, Tubenit

[Jalmeida]

Yes, a number of us have done this.  VVR and cathode biased amps (el34world.com)


After having done this on a few amps, I've opted for PPIMV instead as a way of controlling volume. Both VVR and the PPIMV each have their own advantages.


I found recording an overdriven tone using Mixcraft Acoustica to be more easily done with a VVR.  This recording was done at an incredibly low volume that one could easily talk over.  There were no pedals used with this recording. All OD is the amp.


https://soundclick.com/share.cfm?id=9488477


Other than recording at very low volumes at home, I found no advantage for my purposes of a VVR over PPIMV. I found the VVR on one of the amps to become noisy after about a year or two and removed it and replaced it with a PPIMV which resolved the noise. So for me there was a potential reliability issue using a VVR.


I think the PPIMV does not sound as good at a really low volume as a VVR does. Below about "5" or so, the PPIMV's on my amps lack something.


With respect, Tubenit

Thanks for sharing. I have found lots searching here for VVR along the lines you posted. What I am looking for specifically is a tube based VVR. In benson’s case for their new amps they used a 6L6 Voltage regulator and made it variable.

[Willabe]

Are you saying that Benson (?) uses a 6L6 instead of a power mosfet for the VVR?

[Jalmeida]

Are you saying that Benson (?) uses a 6L6 instead of a power mosfet for the VVR?

Yes. On the newer models of Monarch(I believe). Taken from a Sho Bud amp apparently. I am curious what other tubes can be used. Seems like a pretty cool idea.

Attached is one I found googling. Just looking for more examples or info.

[separateness]

I have nothing useful to add but man, that is cool.

[separateness]

Well, laying the slight danger of saying something useful aside, I note that on this instance at least
https://www.schematicheaven.net/bargainbin/shobud_compactra_100.pdf
The variably regulated voltage seems to only be used for the power tube screens? I wonder how this thing sounds.

[shooter]

we used a similar circuit in C.T. we called the tube regulator "fast regulation" vs the cap/R-bank as slow regulation.
the "top" 1500vdc of the 140,000vdc was a floating tube regulator.

[separateness]

I must note that upon review the Sho Bud amp regulator is not variable. I am confused as to the benefit of having a regulated voltage at the screens, especially given the extra complexity.  I could understand if it was variable, as I know turning down the screen voltage on a 7199 can give some nastiness that some might like, but why on earth would it be so important to have a dead flat voltage on the screens?

[Jalmeida]

I must note that upon review the Sho Bud amp regulator is not variable. I am confused as to the benefit of having a regulated voltage at the screens, especially given the extra complexity.  I could understand if it was variable, as I know turning down the screen voltage on a 7199 can give some nastiness that some might like, but why on earth would it be so important to have a dead flat voltage on the screens?

I am not sure which Sho Bud amp Benson’s design is from. I am presuming with my limited design experience that making the screen resistor variable is the gist of the idea. But there is likely more computation than I am recognizing.

[AlNewman]

I must note that upon review the Sho Bud amp regulator is not variable. I am confused as to the benefit of having a regulated voltage at the screens, especially given the extra complexity.  I could understand if it was variable, as I know turning down the screen voltage on a 7199 can give some nastiness that some might like, but why on earth would it be so important to have a dead flat voltage on the screens?

It's called a control grid for a reason.  I'm not saying I know exactly what the reason is, but I've seen many designs where the screens are regulated.  My 2 cents.

[separateness]

I am not sure which Sho Bud amp Benson’s design is from. I am presuming with my limited design experience that making the screen resistor variable is the gist of the idea. But there is likely more computation than I am recognizing.

Well, either way the only real difference between the fixed regulator and the variable one is whether you use a couple of fixed resistors or a pot, it seems to me. Probably what you need to know is what range of screen voltages are musically useful, and what grid voltages on the regulator the are going to give you said voltages.  I know on my echo unit which has a pot for the 7199 screen grid, there is a window of good sounding settings and a lot at the fringes which I have no use for.
To one of your initial questions, I would think most any power tube would work, so long as it could take the required voltages (especially the Vhk) and currents (which should be low?). Do NOT take my word for it though. The regulator is triode strapped so you could probably even use a power triode if they weren't so dear.

@AlNewman At least between the schemo I posted and the OP's intended design, there seems to be two different reasons for going about this.  The OP and the design he cites seems to be to manipulate the screen voltage so as to get distorted sounds at lower volumes, a kind of elaborate master volume. In the Sho Bud amp, they seem to just want a rock solid screen voltage, maybe for noise or linearity's sake.  Sorry If I am stating the obvious but I am trying to work out in my head the utility here.

Here is a bit of interesting read I just found by websearching 'regulated screen supply':
https://www.w8ji.com/screen_supplies.htm
He seems to say that regulating the screen voltage is good for linearity and component safety. To wit:

[...]linear stages are especially critical for screen voltage. Even though proper design technique of linear stages dictates a very tightly regulated screen supply, linear amplifiers sometimes cheap out and use non-regulated screen supplies.

[AlNewman]

Yes, and the screen grid resistor is usually the largest in the amp.

[Willabe]

I don't see a single reason to use a power tube in place of a power mosfet for VVR/Power Scaling.

With a power tube, you have more heat, takes more space, you need filament voltage/current.

They made regulator tubes, but I don't know if you could make the tubes circuit adjustable? And they made those tubes long before power mosfets were around.   
 

[Jalmeida]

I don't see a single reason to use a power tube in place of a power mosfet for VVR/Power Scaling.

With a power tube, you have more heat, takes more space, you need filament voltage/current.

They made regulator tubes, but I don't know if you could make the tubes circuit adjustable? And they made those tubes long before power mosfets were around.

My understanding of Chris Benson’s reason was to retain sag characteristics regardless of power setting. Versus solid state. This goes on a lwvel of tube amp design and understanding I am just not at. I was hoping somebody here had played with it.

A friend just sent me a scan from Merlin’s power supply book. It has this schematic for something similar using an EL34. I don’t have the full book, just this scan.

[Willabe]

My understanding of Chris Benson’s reason was to retain sag characteristics regardless of power setting. Versus solid state. This goes on a level of tube amp design and understanding I am just not at. I was hoping somebody here had played with it.

Why using a power tube instead of a power mosfet for a VVR/Power Scaling circuit to retain sag would have to be explained.

Sag comes from the power tubes drawing enough current to load down the power supply rectifier tube mostly, and/or some from a PT B+ wind that is under spec for current, then the B+ dcv sags. 

I don't remember how solid the VVR/Power Scaling circuit is at holding/regulating the B+dcv. Or does it just 'ride' down and back up with the power supply load/sag.

When you turn the VVR/Power Scaling circuit down, you un-load the power supply. The amp draws less voltage and less current. So no sag. 

I don't see how switching the power mosfet for a power tube changes any of that.

[Jalmeida]

My understanding of Chris Benson’s reason was to retain sag characteristics regardless of power setting. Versus solid state. This goes on a level of tube amp design and understanding I am just not at. I was hoping somebody here had played with it.

Why using a power tube instead of a power mosfet for a VVR/Power Scaling circuit to retain sag would have to be explained.

Sag comes from the power tubes drawing enough current to load down the power supply rectifier tube mostly, and/or some from a PT B+ wind that is under spec for current, then the B+ dcv sags. 

I don't remember how solid the VVR/Power Scaling circuit is at holding/regulating the B+dcv. Or does it just 'ride' down and back up with the power supply load/sag.

I don't see how switching the power mosfet for a power tube changes any of that.

You are talking transformer voltage sag.

He is talking about the rectifier sag caused by the rectifier internal resistance change with current draw. My understanding is that it preserves the sag curve characteristics. Sure you can fake sag on a mosfet type VVR. But will it have the same sag curve characteristics?

[tubenit]

My question is whether one could actually tell a difference between a mosfet vs tube in a VVR design with a blind test?


I have not been able to discern any difference in tone between a mosfet and a triode in a CF position.


With respect, Tubenit

[Jalmeida]

My question is whether one could actually tell a difference between a mosfet vs tube in a VVR design with a blind test?


I have not been able to discern any difference in tone between a mosfet and a triode in a CF position.


With respect, Tubenit

Good question. I couldn’t tell you, honestly. Considering the amount of differences in sound characteristics I have heard from changing seemingly insignificant electrical things, I can’t imagine why it wouldn’t make a difference. I am still reeling from how much better the AC10 I built with Carbon Comp resisitors sounds compared to the one I made with metal film resistors. That is what I love about building is trying things that you rarely see on a large production scale and seeing what difference it makes.

[Willabe]

Sag comes from the power tubes drawing enough current to load down the power supply rectifier tube mostly, and/or some from a PT B+ wind that is under spec for current, then the B+ dcv sags.

You are talking transformer voltage sag.

He is talking about the rectifier sag caused by the rectifier internal resistance change with current draw. My understanding is that it preserves the sag curve characteristics.

No I'm not. Please read it again. I wrote rectifier tube.

[Jalmeida]

Sag comes from the power tubes drawing enough current to load down the power supply rectifier tube mostly, and/or some from a PT B+ wind that is under spec for current, then the B+ dcv sags.

You are talking transformer voltage sag.

He is talking about the rectifier sag caused by the rectifier internal resistance change with current draw. My understanding is that it preserves the sag curve characteristics.

No I'm not. Please read it again. I wrote rectifier tube.

I see now what you were saying about rectifier sag. I stand corrected. Misread. My apologies.

To your point, maybe in tube VVR configuration there is more current draw from the rectifier when turned down?

I honestly came here for the answers about the tube VVR as I don’t have them. More hoping to learn from some who has implemented and knows how well they work. Or if they work with regards to sag.

[Willabe]

My understanding of Chris Benson’s reason was to retain sag characteristics regardless of power setting. Versus solid state.

He is talking about the rectifier sag caused by the rectifier internal resistance change with current draw. My understanding is that it preserves the sag curve characteristics.

This is what doesn't make sense "retain sag characteristics regardless of power setting."   

The power setting matters.

You only can get sag from a rectifier tube when the amp draws enough from the power supply that the tube can't supply. You have to turn an amp up high enough/loud enough before their power supply will sag. It's only when the rectifier tube can't supply the current draw being asked of it and because of ohms law the voltage drops to make up for it, the voltage 'sags.'
But when you turn DOWN a VVR/Power Scaling circuit it draws less and less voltage/current (= power). So your un-loading the power supply as you turn down. That means the power supply, it's rectifier tube, can easily supply ALL the voltage/current it's being asked for, so NO sag.   

Sure you can fake sag on a mosfet type VVR. But will it have the same sag curve characteristics?

I guess you are referring to adding a series resistor in the power supply?

[Willabe]

To your point, maybe in tube VVR configuration there is more current draw from the rectifier when turned down?

The tube is being used to drop dcv, I wouldn't think there would be enough added current draw with that extra tube to make up for the drop in demand from the VVR/power Scaling circuit.

I honestly came here for the answers about the tube VVR as I don’t have them. More hoping to learn from some who has implemented and knows how well they work. Or if they work with regards to sag.

Nothing wrong with wanting to learn about this circuit option. It's a good question. I'd like to know too.

I don't recall ever seeing that VVR/Power Scaling circuit used here with a power tube. But, maybe I just missed it?

There are some guys here that should be able to answer you questions about this circuit.

[Jalmeida]

To your point, maybe in tube VVR configuration there is more current draw from the rectifier when turned down?

The tube is being used to drop dcv, I wouldn't think there would be enough added current draw with that extra tube to make up for the drop in demand from the VVR/power Scaling circuit.

I honestly came here for the answers about the tube VVR as I don’t have them. More hoping to learn from some who has implemented and knows how well they work. Or if they work with regards to sag.

Nothing wrong with wanting to learn about this circuit option. It's a good question. I'd like to know too.

I don't recall ever seeing that VVR/Power Scaling circuit used here with a power tube. But, maybe I just missed it?

There are some guys here that should be able to answer you questions about this circuit.

Here is what Benson said about the circuit and theory:

“You might be familiar with power scaling. It is a form of power reduction that is usually done with a silicon power MOSFET to lower voltages in an amp. The new Monarch Reverb Plus can be dialed in from 1/2 a watt to 15 watts.
Everyone else uses a transistor. We set up a real life power tube (6L6) as a regulator. You might ask what the point is, and that is a fair question. Here's the point: the "sag" people enjoy in tube rectified amps is because that rectifier tube both limits current, and has an internal resistance...that translates as compression that responds to your playing... certain great guitarists can play sag like its own instrument.
Putting a tube in the power supply path like this creates sag.
We've also included the option to bypass the tube completely at the highest setting, which helps the Monarch bark like a Marshall, if that's what you want.
Of course the science doesn't mean anything if it doesn't sound as good as the original tube rectified Monarch Reverb combo. It does...check out the Demos in the Dark demo.
With no prompting, Ryan (a longtime Monarch Combo user) asked if it was weird that he preferred the 12 watt setting rather than the (highest) 15 watt setting...! told him it's because that tube is in the power supply at 12 watts... doing its sag thing. It really works!
Credit where credit is due....I was inspired by an early Sho Bud amplifier that uses a tube as a fixed regulator for a screen supply (I imagine to try to get a cleaner sound for pedal steel). We first used a tube regulator in the Vincent in 2017, then made it variable in the Vinny Reverb combo in
2020. Now we're putting it in bigger amps and, even though it's WAY off the reservation for what the tubes were designed for, nothing is failing with it. I earnestly invite other builders to stop using MOSFETS and start using tubes for VVR...it's so fun!”

[tubeswell]

If a triode/pentode is used to supply regulated current, then you need to consider the capability of the regulating tube to supply the required amount of current demanded by the rest of the amp - including the Pmax of the regulator tube. And (of course) its a good idea to have a separate a filament supply for the regulator, which can sit at several hundred VDC.

Some old HiFi amps used tube regulation for the output tube screen supply or screens and pre-amp supply (which only draws a fraction of the whole amp's current) - see examples.

[tubeswell]

Another example - triode regulator used for biasing the PI in Hiwatt

[Merlin]

My question is whether one could actually tell a difference between a mosfet vs tube in a VVR design with a blind test?

I have not been able to discern any difference in tone between a mosfet and a triode in a CF position.

If the DC output voltage from the VVR is the same in each case, you will hear negligible difference. Both tube and MOSFET regs will hold the voltage constant enough to sound the same.

When Benson says "Putting a tube in the power supply path like this creates sag" he is wrong, or is leaving out the technical details. He could be talking about the edge-cases where the reg is turned up to max voltage and it stops regulating altogether, or the tube is deliberately 'undersized' and hits a current limit before a MOSFET would. I haven't seen the schematic though, so I don't know if that's what he means. Even then it would really be an aesthetic choice, since a MOSFET plus a resistor could create the same sag at a fraction of the cost.

*Footnote: A VVR in reasonably sized amp may have to dissipate a lot of power, and putting in a big-enough heatsink could be a design pain. A tube reg could side-step this problem since the tube doesn't need a heatsink. I don't think that's the case here since we're only talking about a little 15W amp.

[Jalmeida]

My question is whether one could actually tell a difference between a mosfet vs tube in a VVR design with a blind test?

I have not been able to discern any difference in tone between a mosfet and a triode in a CF position.

If the DC output voltage from the VVR is the same in each case, you will hear negligible difference. Both tube and MOSFET regs will hold the voltage constant enough to sound the same.

When Benson says "Putting a tube in the power supply path like this creates sag" he is wrong, or is leaving out the technical details. He could be talking about the edge-cases where the reg is turned up to max voltage and it stops regulating altogether, or the tube is deliberately 'undersized' and hits a current limit before a MOSFET would. I haven't seen the schematic though, so I don't know if that's what he means. Even then it would really be an aesthetic choice, since a MOSFET plus a resistor could create the same sag at a fraction of the cost.

*Footnote: A VVR in reasonably sized amp may have to dissipate a lot of power, and putting in a big-enough heatsink could be a design pain. A tube reg could side-step this problem since the tube doesn't need a heatsink. I don't think that's the case here since we're only talking about a little 15W amp.

Thanks for that explanation. A huge help.

To your point, I don’t have his schematics for the amp in question. So I suppose it is a guess at how he supposedly gets sag. Perhaps with a resistor in the same way as a Mosfet? Is that even possible? And then would it sound any different than a mosfet in that case? I would love to see some sort of Oscilloscope data to show sag and see what is talked about.

As with many things guitar related, there is mojo factor or perceived improvement. And trying to substantiate with data this mojo in some things are tough. So could be snake oil. Or not.

[kagliostro]

I agree with those who say the audible difference will be negligible putting a MOSFET VVR vs a Tube VVR

I didn't read all the post of this thread but seems to me that all starts from a circuit where G2 voltage was regulated using a tube, this architecture was used in the past when the intention was to have a G2 voltage more stable or lower than B+ or both things and the cost of Solid State components was very high

One thing may put on a difference, a VVR on the B+ path or in the G2 pat can result in a bit of audible difference if used in alternative, both obtain a power reduction but may be you'll find it to have small audible differences

About tube circuits to achieve voltage regulation I've a small collection of examples but it is on my broken PC so I don't have access to it for some time, if there is interest I can post it when my PC is newly usable

Despite of all there are people that prefer to use OA2 tubes in voltage stabilization because they ..... glow

Franco

[Jalmeida]

About tube circuits to achieve voltage regulation I've a small collection of examples but it is on my broken PC so I don't have access to it for some time, if there is interest I can post it when my PC is newly usable

Despite of all there are people that prefer to use OA2 tubes in voltage stabilization because they ..... glow

Franco

I would love if you could share if not too hard.

[kagliostro]

If you have a bit of patience I'll try to find the documentation and to post here

Unfortunately, as told, my PC is out of service and the thing will be a bit more difficult (but I hope not impossible)

Franco

[kagliostro]

Here I'm


Attached the schematics

Franco

[kagliostro]

Others


Note that one is a Bench Power Supply, if you have interest on those circuit you can find other schematics of this kind having a Google search


Franco

[Jonas]

I've always estimated screen current to me around 5mA, but does the screen current increase as the pot is adjusted to lower screen voltage?