De-biasing a Peavey Classic VT (6L6 output stage)

[iep]

I'm repairing an old Peavey Classic VT (212). The Classic VT has a full solid state (op-amp based) preamp and then uses transistors for the phase splitter before the push pull 6L6 output stage.

The output stage uses Grid biasing which is apparently factory set fairly cold so that any matched pair of 6L6 valves should work fine. There is no adjustment potentiometer included. All this can be seen in the attached schematic.

So, my question is this. Could I add a switchable grid bias resistor to futher de-bias the output stage and reduce it from stock ~50W to something more suited for home practice? Would it be as simple as including a switch to adjust the grid bias resistors so that the plate current is reduced by 50% (or even 75%)? Or, would this simply cause the output stage to stop operating or even become damaged?

Thanks,

iep

[potatofarmer]

That won't work. The bias voltage controls the idle current. Too much idle current and the tubes overheat. Too little idle current and you get crossover distortion. Drop the idle current to zero and the tube won't pass signal.

The grid leak resistors on the power tubes have essentially no impact on the bias voltage (assuming they're not far too large).

This amp has a master volume. Probably a good idea to use it.

[TigerAmps]

Here is an idea

You could try the triod switch (similar to what was used in the Marshall JCM 900 amps [low/high] switch)
BUT (big but here) I think the 6L6 sockets are mounted (soldered) onto a circuit board making this difficult to do
without taking that board out and mounting normal chassis mounted tube sockets and moving the circuitry onto a turret board or terminal strip.
This is a lot of work and probably not worth the trouble. Someone did do this in a Peavey Butcher.

[Enzo]

Power is not loudness. If you cut the power in half, the sound will only be three decibels less loud. That is a very small difference. In fact, to get it half as loud, you need 1/10 the power.

Those cool Peavey tube power amps are designed loud, strong, and clean. In other words, I think their contribution to tone is not as great as in some other amps. The tone comes from the preamp, and as was suggested above, just turn down the post control or master.

[iep]

Thanks for the responses.

Potatofarmer, could I ask for a bit of clarification? I had read elsewhere (another forum) that the 47k resistor on the Grid was somehow involved in setting the bias voltage. I'm very new to valve amps so I had just assumed that there was some kind of current input to the Grid pin of the valve (much like the base of a transistor) and so some voltage drop was present over the 47k res. From what you are saying, this is nonsense and the valve operates with little or no input current to the Grid pin (more like a FET). In that case, bias voltage would only be adjustable by altering the potential divider formed by IR8 and IR9 in the power supply? This makes much more sense to me. Have I picked this up correctly?

Thanks also for clarification on the bias current situation. It does indeed sound like my proposed approach would not work. I had simply figured that the 'Plate Voltage*Plate Current=Output Power' formula could be applied blindly. Of course, the valves still need to be kept within their operational range and I was forgetting this.

The amp does indeed have a master volume but I am curious to hear how it would sound with power amp clipping rather than the slightly dated pre-amp distortion sound that the amp currently provides. In fairness, I am a huge fan of the amp's existing sound but wanted to see if there are ways to increase the range of tones available from it.

TigerAmps. You mention the triode switch trick, I'll look it up but could you point me in the direction of anything to read about this? Fortunately, the amp does indeed feature chassis mounted sockets so there is scope for relatively easy modification.

Cheers,

iep

[Enzo]

Power amp clipping only occurs at maximum power output, but you said you wanted to drop the output power to more practice friendly levels. Those are mutually exclusive. Kinda like asking to fly a plane faster than the speed of sound, but you want to do it at a lower safer speed.

The only way you will ever hear power tube distortion is to run them full out. The only way to hear that at less loud levels is to use an attenuator AFTER the amp.

So you have to pick one: lower loudness or power amp clipping.

[Dave H]

It does indeed sound like my proposed approach would to work. I had simply figured that the 'Plate Voltage*Plate Current=Output Power' formula could be applied blindly.

It won't work. Your equation only gives the idle power (= heat) of the tubes not the audio output they can produce. The (idle) plate current could be reduced to almost zero (= zero idle power) and it would still produce full audio power. To use Enzo's analogy, you could reduce the idle speed of your car from 1000 rpm to 100 rpm but it is still capable of doing over 100 mph.

[iep]

Thanks Enzo. This may be down to my lack of understanding of valves but my take had been that, according to the formula below:

Plate Voltage * Plate Current = Plate Dissipation

and

Plate Dissipation = Max Output Power

So, by reducing bias voltage, and so Plate current, the Max Output Power would be reduced and earlier clipping achieved.

Now, it might simply be that my understanding of the relationship between plate current/voltage and output power is wrong. In which case, could you help me understand what I have picked up incorrectly?

Thanks,

iep

[iep]

Dave, thanks. So my issue is that I had understood that Idle Power (Plate Dissipation) was equivalent to output power. I had simply guessed that the output stage was almost operating like Class-A.

So, the Plate Dissipation really sets the ClassA/B bias current.

In which case, the max output power is simply set by the valves max power (assuming power supply and OT are up to the job)?

Thanks,

iep

[Chuck H]

iep,

Just so you know you're thinking on the right track, there IS a system of reducing the output power of tubes by altering their operating parameters. It's most often called power scaling. Though I think that term may be under copy write by the guy most associated with the circuit. It includes a reduction in plate voltage to achieve this. That's what was missing from your notions on the matter because a tubes min/max operating point is dictated by the high voltage on the plate. Not the bias. This circuit alters the plate voltage (with a resulting lower current) to achieve lower output at the onset of clipping. The circuit needs to be made self adjusting WRT the bias. Basically, altering just one parameter of the tubes operation won't do it. You need to alter all the operating parameters.

My point is, you were on the right track. You CAN change the efficiency of a tube such that it clips at a much lower maximum power by altering it's operating conditions. And it's been done. Probably the easiest way for YOU to hear some power tube distortion would be to bias the amp properly for it's current plate voltage and get an attenuator as per Enzo's suggestion.

[potatofarmer]

Thanks for the responses.

Potatofarmer, could I ask for a bit of clarification? I had read elsewhere (another forum) that the 47k resistor on the Grid was somehow involved in setting the bias voltage. I'm very new to valve amps so I had just assumed that there was some kind of current input to the Grid pin of the valve (much like the base of a transistor) and so some voltage drop was present over the 47k res. From what you are saying, this is nonsense and the valve operates with little or no input current to the Grid pin (more like a FET). In that case, bias voltage would only be adjustable by altering the potential divider formed by IR8 and IR9 in the power supply? This makes much more sense to me. Have I picked this up correctly?

R27 and R28 (47k) are referred to as grid stoppers, they form a low-pass filter with the input capacitance of the power tube (albeit probably still a supersonic one) and help to reduce blocking distortion.

IR25 and IR26 are the grid leaks, which are used to bias the grids (or with cathode biased tubes, provide a 0V reference). They also set the input impedance of the power stage.

But to answer your question more directly: there is basically no grid current whatsoever, certainly not at idle. There are exceptions, of course; as Vg1 approaches 0 or if you're trying to push it positive w.r.t. the cathode, it's going to start drawing current. This is the distinction between classes AB1 and AB2.

There's also reverse grid current, which comes from positively-ionized particles being attracted to the grid. When they hit, they make the bias voltage less negative, so the current passing through the tube increases, which leads to more heat and more positive particles (etc.) and eventually into thermal runaway. To avoid this, tube datasheets specify a maximum resistance between G1 and the cathode, which most tube guitar amp designers gleefully ignore for a bunch of reasons.

And yes, IR8 and IR9 are what you'd tweak to adjust the bias voltage.

[iep]

Thanks to all who have contributed here. Really helpful. Particularly appreciate the last posts bu ChuckH and potatofarmer. These have properly clarified my understanding of the valves operating conditions and the purpose of the various resistors used after the AC coupling caps to the first stage. I'd really struggled to find a better succinct description elsewhere.

And yes, it definitely looks like a properly biased amp and a speaker attenuator might be the way to go.

Cheers,

iep

[eschertron]

Thanks Enzo. This may be down to my lack of understanding of valves but my take had been that, according to the formula below:

Plate Voltage * Plate Current = Plate Dissipation

and

Plate Dissipation = Max Output Power

So, by reducing bias voltage, and so Plate current, the Max Output Power would be reduced and earlier clipping achieved.

Now, it might simply be that my understanding of the relationship between plate current/voltage and output power is wrong. In which case, could you help me understand what I have picked up incorrectly?

Thanks,

iep

To add to what others have said: the output power has to do with the transfer efficiency of the tube/OT impedance (impedance matching) and not specifically the dissipation power of the tube. Of course the bigger the tube, the more potential power there is; but a poorly-chosen OT can limit the amount of power that the speaker sees. The OT must be considered as an integral part of the PA when looking at the power that gets to the speakers.

[iep]

On the triode switching idea. Does the image above basically capture what would need to be done? Seems a little too easy but this is based on a bit of background reading I have done today.

Cheers,

iep

[Chuck H]

That's the basic idea, yes. I'm not sure how the amp is arranged inside but there are usually individual screen grid resistors on most amps. One for each screen. These resistors are usually placed between the switch and screen for the mod.