In need of 2x6L6GC power amp expertise

[pdf64]

...Frequency doubling would be my guess. The 'dip' in the lower sine wave peak mirrors the negative peak of the upper sine wave. This is probably due to the hot bias of the LTP, which would increase the onset of grid current limiting. Merlin discusses this on p172-173 of the 2nd edition pre-amps book. If you make Rb=470R, this may well disappear.

Yes it looks like that but I don't see how in this case (plate loaded by crossline MV) the load value would be invoking the effect; surely it's would just be down to the signal level at the grid, and the bias?

[tubeswell]

...Frequency doubling would be my guess. The 'dip' in the lower sine wave peak mirrors the negative peak of the upper sine wave. This is probably due to the hot bias of the LTP, which would increase the onset of grid current limiting. Merlin discusses this on p172-173 of the 2nd edition pre-amps book. If you make Rb=470R, this may well disappear.

Yes it looks like that but I don't see how in this case (plate loaded by crossline MV) the load value would be invoking the effect; surely it's would just be down to the signal level at the grid, and the bias?

On P 173 Merlin's explanation is that - "the increasingly negative signal peaks that are passed via the cathode to the slave triode cause it to draw grid current too, draining charge from its decoupling capacitor".

If the cross-line MV worked perfectly, then there'd be a gradual and equal reduction in both signals all the way to '0' on the dial. But we see uneven signals at '1'. Is this simply because one LTP output has slightly more gain? or because there is something else happening in the LTP? (I suspect a combination of both).

Whereas, if you dimed the MV, and drove the LTP really hard, you would get red-plating on the side of the output stage that is being fed by the master triode, because the duty-cycle of the LTP would become more uneven, because the LTP bias is hot. This leads me to think the hotter than usual bias must also be playing a role in what is seen on the scope here.

Thinking about this a bit further, when the master triode is subject to grid current limiting, its grid-to-cathode voltage becomes clamped because the grid voltage can't change but there's still increasing current at the grid. This current is sunk through the LTP tail, increasing the tail voltage (while the master triode is in grid current limiting). This would push not only the slave triode plate up by about the same amount, but also could 'hold-up' the master triode plate voltage somewhat. Also, the two grids are connected (in a cathode biased LTP) via the grid leak resistors, therefore some of the forward grid current from the master triode would be sunk through this pathway, affecting the charging/discharging cycle in the slave triode grid decoupling cap. Being a differential amplifier, the LTP amplifies the difference between both grid signals.

However, because the effect of grid current limiting on the master triode plate may be softened somewhat further by the increase in voltage across the tail, then depending on how hard the grid in the master triode was being driven, you might get a point at which you still had what resembled a sine wave at the master triode plate, but the beginning of grid current limiting in the slave triode, causing a small signal output discrepancy, which may be picked up by the cross-line MV at lower settings.

This is all somewhat speculative, and I may be overthinking things. It would be interesting to see both LTP grids and LTP tail signals on a scope while all this is going on.

[Bergheim]

I got to play a little more with the scope today. Tried some different bias settings again, and it shows the same symptoms no matter the bias resistor value. However, it looked like it got worse the more I offset the PI balance, which supports my theory that the dominant PI side signal bleeds across the mv pot at low settings. Also, no stages including the PI are driven into overdrive/distortion at this point, so I can't see how this can be frequency doubling due to grid current limiting. I tried to hook a probe directly to the PI input to see if the grid was clamped but I think it got loaded down by the scope which made some really ugly distortion so no luck there.

[Tony Bones]

You can probe the shared cathode without problem. You should be able to see from there if the PI is being driven into grid current. But I don't think that it is.

In the end, that style of MV is just not well suited to cutting the signal so much. A pre-PI MV might work better.