Why "Improved" Master Volume?

[miallen]

I'm building a kit and I want to add a post PI master volume. I was thinking of just replacing the pull-down resistors with a dual 250K pot. But it seems there is an "Improved" method that adds a dual 1M pot and 1u caps in front of the pull-down resistors.

However it is not obvious to me why this method is "Improved". If the point is only to eliminate DC on the wiper so that it doesn't make a scratchy noise when turning the pot, I don't think I care. There should be very little grid current and not a lot of amplification at that point so I don't suspect the scratchy noise would be that bad. Can someone speak to this?

Also, it seems to me if you add a big 1u cap in series with the grid and the grid start drawing current when overdriven (aka grid current limiting), those caps are going to charge up (aka bias excursion) and you'll get a bias shift that would totally change sag dynamics. Right?

I know this has been covered a hundred times but I can't seem to find a really good technical explanation about the different PPIMV implementations.

 

[ampmadscientist]

I'm building a kit and I want to add a post PI master volume. I was thinking of just replacing the pull-down resistors with a dual 250K pot. But it seems there is an "Improved" method that adds a dual 1M pot and 1u caps in front of the pull-down resistors.

However it is not obvious to me why this method is "Improved". If the point is only to eliminate DC on the wiper so that it doesn't make a scratchy noise when turning the pot, I don't think I care. There should be very little grid current and not a lot of amplification at that point so I don't suspect the scratchy noise would be that bad. Can someone speak to this?

Also, it seems to me if you add a big 1u cap in series with the grid and the grid start drawing current when overdriven (aka grid current limiting), those caps are going to charge up (aka bias excursion) and you'll get a bias shift that would totally change sag dynamics. Right?

I know this has been covered a hundred times but I can't seem to find a really good technical explanation about the different PPIMV implementations.

Regular master volume:
between preamp and phase inverter input. Does not allow the phase inverter to be over-driven.

PPIMV: between phase inverter output, and power tubes input.
Allows phase inverter to be over-driven. (more distortion and overdrive)

Coupling caps: the value was increased
to pass more accurate frequency response.

No, the caps will not charge up and interfere with the bias voltages. It's OK to do it that way.

BOTH these designs are a compromise. There is no perfect master volume.

Ideally, the overdrive would be from the OUTPUT TUBES.
But this causes very loud volume levels.
Preamp distortion is a compromise, to get overdrive at lower volumes.

Ideally, there would be NO master volume.
The output tubes would be over-driven, to produce the best possible tone.
Ideally, you would use a speaker, that has variable magnetic strength, to control the final loudness.

That's why there is this:
About FluxTone Guitar Speakers

The amp is cranked wide open to get the output tube distortion...
and the speaker is less efficient, to control the loudness.
That "IS" the state of the art solution.

 

[miallen]

No, the caps will not charge up and interfere with the bias voltages. It's OK to do it that way.

That is not consistent with my understanding. See page 46 of Blencowe Designing Valve Preamps regarding blocking distortion.

In short, when a tube (pre or power) is overdriven, as the grid voltage approaches the cathode voltage, electrons are attracted to the *grid* instead of the plate. This causes current to flow into the grid. So the grid goes from being very high impedance to relatively low impedance. This is called "grid current limiting". Because it is only for postive swings on the grid, the coupling cap acquires a transient net charge. The bigger the cap, the longer it takes to discharge. This is called "bias excursion".

Here is an interesting link that includes a calculator that tells you how long the bias excursion will affect the circuit depending on the size of various components:

Grid Bias Excursion Calculator

As you can see, if you change the coupling cap from 22n to 1u, that has a huge effect on bias excursion.

Note that this would only be a problem if the power tube was actually being overdriven which would not be the case if the master volume were turned down (how much I have no idea). But in the maximum postion, I think long bias excursion would definitely occur.

Disclaimer: I have not observed the above issue myself. It is theory. I have very little experience building guitar amps. But I do have a lot of experience with conventional non-tube small signal audio circuits like pro-audio gear.

 

[ampmadscientist]

That is not consistent with my understanding. See page 46 of Blencowe Designing Valve Preamps regarding blocking distortion.

In short, when a tube (pre or power) is overdriven, as the grid voltage approaches the cathode voltage, electrons are attracted to the *grid* instead of the plate. This causes current to flow into the grid. So the grid goes from being very high impedance to relatively low impedance. This is called "grid current limiting". Because it is only for postive swings on the grid, the coupling cap acquires a transient net charge. The bigger the cap, the longer it takes to discharge. This is called "bias excursion".

Here is an interesting link that includes a calculator that tells you how long the bias excursion will affect the circuit depending on the size of various components:

Grid Bias Excursion Calculator

As you can see, if you change the coupling cap from 22n to 1u, that has a huge effect on bias excursion.

Note that this would only be a problem if the power tube was actually being overdriven which would not be the case if the master volume were turned down (how much I have no idea). But in the maximum postion, I think long bias excursion would definitely occur.

Disclaimer: I have not observed the above issue myself. It is theory. I have very little experience building guitar amps. But I do have a lot of experience with conventional non-tube small signal audio circuits like pro-audio gear.

I don't follow you...
because if the grid voltage approached the cathode voltage...
the tube would burn up.

 

[Micky]

if the grid voltage approached the cathode voltage...
the tube would burn up.

Or at least run up into oscillation and redplate...

 

[miallen]

I don't follow you...
because if the grid voltage approached the cathode voltage...
the tube would burn up.

No, that can't happen because of grid current limiting as described above. Unlike transistors, tubes cannot saturate because the grid current limits how "on" a tube can be.