Phase-4 power control system.

[Otis]

I checked out tone kings new metro amp on you tube and was pretty impressed with what I heard. On their web page they talked about a new power control they dubbed the Phase-4 power control system. Claim it's unlike anything else out there. On the you tube demo it does seem as though the amp sounds remarkable at all volumes, something I have not heard from amps using the usual attenuators, triode switching,master volume , ect. Anyone out there have a clue as to what they might be doing different?

[Otis]

OK, here's a reply Mark Bartel from Tone King was kind enough to send me.

"Thanks for your interest in the Metropolitan !  The phase-4 system is a lot different than the usual master volume control.  It's based on the fact that overdrive and distortion in a tube amp of this sort is actually generated by and interaction between the grid clamping of the output tubes and the phase inverter.  The phase-4 circuitry makes this effect happen at varying levels, effectively controlling the output power produced by the output stage. "

My level of understanding the dynamics of tube amplification is basic at best, I know the PI splits the signal in two for the push pull power tubes, but I thought the sweet distortion is occuring by pushing those tubes. What do you suppose he ment by "interaction between the grid clamping of the output tubes and the phase inverter" ?

[tubenit]

interaction between the grid clamping of the output tubes and the phase inverter"

I don't know what he meant, but I would view a PPIMV doing "grid clamping" meaning dropping signal to ground.

With respect, Tubenit

[The_Gaz]

That's really nice he wrote you back, and his amps are some of the nicest of there. I'd like to know how the Mid-Bite control on his amps work.

But here's the thing about the Phase-4... There's really only so many ways one can lower the power of an amp, and his amps have a big honkin' MOSFET on the top of the chassis so I assume he's using power scaling in  one form or another. His comments about the PI and output tubes would lead me to believe that he's simply scaling the PI and output tubes. I also think this because in my experience scaling the whole amp to very low volumes sounds really bad, while doing just the PI and power tubes still sounds pretty good. Also, just scaling the output tubes lone sounds pretty bad when turned way down as well unless you have a PPIMV of some kind to throttle the level. The Tone King's have no other power amp level control, which is what makes me think he's just scaling the PI and output tubes. Doing so also keeps the drive to output tubes relatively the same, and the PI get slammed harder contributing to the 'grid clamping' effect.

What I find most interesting is that he thinks grid clamping is a good thing to a certain extent, while most people are trying to abate it. Personally, I like to design a little in to vintage style amps while limiting the bass a bit to keep it from getting out of control, but still leaving power tubes grid resistors small enough that the amp sounds like it's on the brink of implosion when maxed out :)

Thanks for sharing!

[Otis]

Thanks for the replys, Gaz I think you nailed it.

[HotBluePlates]

... his amps have a big honkin' MOSFET on the top of the chassis so I assume he's using power scaling in  one form or another. His comments about the PI and output tubes would lead me to believe that he's simply scaling the PI and output tubes. I also think this because in my experience scaling the whole amp to very low volumes sounds really bad, while doing just the PI and power tubes still sounds pretty good. ...

What if you had a single control that scales down the power amp and phase inverter, while a second pot section also reduces the drive signal from the preamp to the phase inverter?

You may not see them often, but there are dual-pots available, even if you need a 2w rated pot for your power-scale/VVR. They could even be dissimilar sections (probably by special order).

When using power scaling (as I understand it), if you don't scale the whole amp, you have to have a master volume control between the scaled and unscaled parts of the amp. The amp is designed to have proper signal level before scaling; once one part of the amp is scaled to be "smaller" the unscaled signal looks "larger". You have to use a master volume to reduce the unscaled signal to keep from getting a too-distorted sound, or to allow a clean sound at the lower volume level. Otherwise, you can get less power, but more distortion is the only sound available.

... Doing so also keeps the drive to output tubes relatively the same, and the PI get slammed harder contributing to the 'grid clamping' effect.

What I find most interesting is that he thinks grid clamping is a good thing to a certain extent, while most people are trying to abate it. ...

You might be equating "grid clamping" to "grid blocking." Or maybe they're the same; I really don't know if there are set definitions for some of these things.

What I think he's saying is there is a point on the volume of typical amps when the preamp/phase inverter is overdriving the output tubes to some extent, and hanging out in that region is a good thing. Less drive means no distortion/less touch sensitivty, while much more drive could be a ragged-sounding distortion (or in extremes, grid blocking).

What he may be calling clamping is this:
If you don't drive a tube grid positive (relative to the cathode), it looks like an extremely high impedance; the phase inverter can deliver its signal as intended. Once the grid is more positive than the cathode (positive peak input signal exceeds the bias), the grid starts collecting some of the current intended for the plate. It is no longer an extremely large impedance. If the bias is -35v, and you apply a 40v peak signal and that causes 1mA of grid current to flow, the grid looks like a 40k resistance.

The phase inverter is no longer lightly loaded by the grid of the output tube, but see a load drawing current that it was not designed to deliver. The phase inverter can't cope, and it's gain drops (we can demonstrate this with simple gain equation for a typical preamp tube stage). So, due to the output tube being overdriven, the phase inverter also has its output capability sapped. One could look at this as the output tube clamping the maximum signal the phase inverter can deliver.

There are amplifiers designed to drive output tube grids positive, for greater output power from the available tubes. Those are the class AB2 or class B2 amps, where you normally see the "1" in class AB1 omitted. You'll usually see an output tube (say, 6V6 or 6L6) as a driver for a transformer that performs the phase splitting, which is able to deliver the power (voltage and current) demanded by the grid of the real output tubes. Almost every production guitar amp runs class AB1, and only requires drive voltage, not drive power.

[Otis]

Hmmm, some thoughtful info to consider HotBluesPlate. Especialy on the AB2 analysis. The amp in question is rated .01 to 40W using two 6v6 cathode biased. Since both channels have independent control of the ability to power scale,(really cool feature), I assumed it was basically two amps in one, but maybe they're used in the manner you describe as best to my elementary knowledge.

[phsyconoodler]

Marketing talk for VVR of the power amp.Maybe a few twists here or there.

[The_Gaz]

What if you had a single control that scales down the power amp and phase inverter, while a second pot section also reduces the drive signal from the preamp to the phase inverter?

You may not see them often, but there are dual-pots available, even if you need a 2w rated pot for your power-scale/VVR. They could even be dissimilar sections (probably by special order).

When using power scaling (as I understand it), if you don't scale the whole amp, you have to have a master volume control between the scaled and unscaled parts of the amp. The amp is designed to have proper signal level before scaling; once one part of the amp is scaled to be "smaller" the unscaled signal looks "larger". You have to use a master volume to reduce the unscaled signal to keep from getting a too-distorted sound, or to allow a clean sound at the lower volume level. Otherwise, you can get less power, but more distortion is the only sound available.

Well, I'd already seen gutshots, so I knew that he's only using a single pot. In practice scaling just the PI and output tubes without a master volume sounds just fine, and the increase in overdrive is not only negligible, but desirable to a lot of people who want the cranked at bedroom level deal. The point where things start to sound funny is at whisper quiet level, but how could they not?! Scaling the whole amp works to a certain level, but starts to sound so strange when the preamp tubes get down to too low a voltage. You can go so much lower in volume just scaling the PA before things get weird.

It's kind of like how a PPIMV messes with the feedback ratio. In theory, it gets all screwy and should be compensated for. In practice it's not so bad, especially in guitar amps where not a huge amount of feedback is used in the first place.

I think the tone change both examples is tolerable because at such low volumes it's hard to keep things in perspective anyway, especially because the speaker stop doing its thing as well. And honestly, I stop caring when I can hear the pick hitting the strings over the amp (Jimi rolls in grave).

You might be equating "grid clamping" to "grid blocking." Or maybe they're the same; I really don't know if there are set definitions for some of these things.

It's my understanding they are the same thing, and I do believe he likes the way the unscaled signal looks larger as you described - kind of a convenient side effect to scaling just the PA.

[HotBluePlates]

You might be equating "grid clamping" to "grid blocking." Or maybe they're the same; I really don't know if there are set definitions for some of these things.

It's my understanding they are the same thing, and I do believe he likes the way the unscaled signal looks larger as you described - kind of a convenient side effect to scaling just the PA.

Grid blocking, though, happens when you apply an extremely large input signal to an output tube, while also having a too-long time constant formed by the grid reference resistor and the coupling cap. The grid current caused can charge the coupling cap, counteracting the signal and "blocking" the grid.

The other effect I described likely happens every time you drive the output tubes hard enough to distort. But check me... scope a grid while applying a signal big enough you can hear the output tubes are distorting.