Tainwreck master volume question

[Randall]

Hey, if it worked for Trainwreck.....

[Chuck H]

Well it's just simulations, but...

Using a TW type circuit with a pair of el84's (because that's the circuit I already have constructed) I measured the signal at the power tube grids. The simulation is set up for what would be considered this circuits "sweet spot" by most players.

The top shot is with no master circuit in place. Adding the above master circuit and reducing resistance I started to see signals sharing the same side of 0V (what I'm assuming is "common mode") at about 50k. So the second from top is with the circuit at 40k. Then 20k, 10k and last is 2k. Below that (in red) is the resulting waveform at the amps output.

[Chuck H]

And here are shots at the power tube plates. Top is with no master circuit and the tubes were drawing 150mA of current. Below is with the master set to 2k and the tubes were drawing 173mA of current.

[nickb]

I agree in principle.
The question remains, how high common mode drive signals could actually be. Are there any measurements?

For a 'typical' LTPI with a tail resistor of 22K and with the pot at zero a 1V drive will give you 0.65V on both power tube grids (i.e common mode) and will result in a small output due mainly to mismatching of the power tubes. Seems to remain about the same even when driven hard.

[Chuck H]

Hey, if it worked for Trainwreck.....

Trainwreck amps don't have master volume controls. These mater volume circuits are sometimes attributed to Trainwreck because their diagrams were provided by Trainwreck. Trainwreck didn't design, use or endorse them. The one Ken preferred is called "type 2", but he never installed one in a Trainwreck amp.

[Helmholtz]

For a 'typical' LTPI with a tail resistor of 22K and with the pot at zero a 1V drive will give you 0.65V on both power tube grids (i.e common mode) and will result in a small output due mainly to mismatching of the power tubes. Seems to remain about the same even when driven hard.

Thanks, that's very little and I expect it to be less than the common mode component without the MV as the LTPI has some common mode gain.

[Helmholtz]

The 2k sim shows almost perfect common mode drive of considerable amplitude. Interesting.

[nickb]

And here are shots at the power tube plates. ....

Could you post the schematic you used please?

[Chuck H]

Could you post the schematic you used please?

...

[nickb]

...

{schematic}

Thx. I wanted to check we were on the same page as I wasn't seeing such large currents. To get those numbers I had to drive the PI with 80Vpp. It's VERY interesting. I've avoided this kind of PPIMV but I had never appreciated just how brutal they can be to the power tubes. Almost no power out yet the poor little dears are working their hearts out. I prefer to use a stereo pot.

[g1]

All this and still no schematic of the actual master in or out of circuit.

[Chuck H]

All this and still no schematic of the actual master in or out of circuit.

Ah! Sorry. I didn't save the master to the file when I was done. I literally just placed a resistor across the PI coupling cap outputs.

[Chuck H]

Thx. I wanted to check we were on the same page as I wasn't seeing such large currents. To get those numbers I had to drive the PI with 80Vpp. It's VERY interesting. I've avoided this kind of PPIMV but I had never appreciated just how brutal they can be to the power tubes. Almost no power out yet the poor little dears are working their hearts out. I prefer to use a stereo pot.

I thought the same thing. At first... Then I considered that the actual AC from the plates is a good bit lower when you take that 0C crossing into account. The actual watts the tubes are making is probably less with the master down, though not a lot. To be honest I'm not sure of this, but it seems that way looking at it. There's definitely some validity in Jon's observation of what's going on and I hadn't considered it before. And even if it's not really threatening the tubes the performance at low volumes looks lousy.

[nickb]

I thought the same thing. At first... Then I considered that the actual AC from the plates is a good bit lower when you take that 0C crossing into account. The actual watts the tubes are making is probably less with the master down, though not a lot. To be honest I'm not sure of this, but it seems that way looking at it. There's definitely some validity in Jon's observation of what's going on and I hadn't considered it before. And even if it's not really threatening the tubes the performance at low volumes looks lousy.

The truth is quite the opposite. Taking your cct with the master at 2K, the EL84's dissipate a staggering 34W but just 12W with the same drive level with master wide open. The reason being the cathode to plate voltage is large even with high currents with the MV down.

What is concerning about this is that presumably the intention is to allow you to turn the master down and still get decent breakup at bedroom volumes i.e you will still be driving the output tubes hard. OK it's not 100% continuous so they don't melt, but it seems like this MV arrangement is just mean to them and the power supply.

I think the idea should be banned. I didn't like it before now I really hate it.

[Chuck H]

The truth is quite the opposite. Taking your cct with the master at 2K, the EL84's dissipate a staggering 34W but just 12W with the same drive level with master wide open. The reason being the cathode to plate voltage is large even with high currents with the MV down.

What is concerning about this is that presumably the intention is to allow you to turn the master down and still get decent breakup at bedroom volumes i.e you will still be driving the output tubes hard. OK it's not 100% continuous so they don't melt, but it seems like this MV arrangement is just mean to them and the power supply.

I think the idea should be banned. I didn't like it before now I really hate it.

Ah! I see that now. I was "looking" at the wave forms too fast without thinking and didn't take into account that the waveform isn't crossing at 0V. So yes. That looks horrible.

Sorry Randall. Surely didn't expect this thread to end with "This circuit sucks and shouldn't even be used."