Pentode Push-Pull

[Johan Potgieter]

On a different point of interest, I'm wondering about screen stoppers and whether these detract to any significant extent from screen regulation. I tend to think that stoppers are a good idea, for both the control grid and the screen, as a safeguard aginst parasitic oscillations. However, assuming one has a regulated supply for the screens, stoppers would seem to be acting against the regulation. Perhaps there's a happy medium, i.e. high enough resistance to be effective in preventing parasitics but also low enough not to have an adverse effect on the regulation?

Ray_Moth,

You will recall that this point was raised before on the "6L6p.p. power tranny" thread (my post#16 there). To re-state here, with a 1K screen serie resistor (which to me is the upper limit) the voltage drop variation is some 16V at between zero/max. current for a beam tube (quite smaller for a pentode). That might make some difference - I have never tested, but would say very small. Here in South Africa 10V on say 400V can be the difference encountered just from mains variation. If one takes the more normal screen stopper value of 100 - 200 ohm though I am sure the influence will be negligible.

But as one also finds, many UL designs are without any screen stoppers, so I have an idea their presence is often more pious than really necessary. Still, I include them (i.e. the lower values) just to be safe. But define "safe" - next time I hook something up I will have a look at supersonic frequencies to see whether there is a response difference (tendency to peak) with/without them, or someone else could.

I think I also surmised that since our tubes are good up to well into the MHz region, the inductance of even short hook-up wiring and the G2 capacitance to somewhere might set up the scene for a nice r.f. oscillator. Even a few tens of ohms will swamp the necessary Q though.... so, intuitively, I would suggest that we keep them in!

What sayeth others?

Regards.

 

[gingertube]

Screen resistors

Yep Johan has it right.

Screen stopper will occassionally be REQUIRED to suppress oscillation. Back in the 60's and 70's the British method was to use a common screen resistor feeding g2 on both sides of the Push Pull. This acted to keep the g2 voltage reasonably constant (As one side drew more screen current the other side would be drawing less).

More modern practice is to use separate screen resistors. This gives added protection against tube failures in that it tends to limit g2 dissipation and acts as a fuse for g2 to cathode (or g1)shorts.

The sonic effect is to add a bit of NFB. Guitar Amp guys are quite familiar with using screen resitors to manipulate sound. If you want a "full tilt" guitar sound (power amp distortion and compression) at a lower volume then the way to achieve it is to use quite high values for the individual screen resistors - examples: 4K7 or even 10K for EL84, 2K2 (1K minimum) for EL34.

So for HiFi - use individual screen resistors for a little tube protection BUT don't go so high in value that you turn it into a guitar amp.
Example PPP EL34 Ultralinear with VDV2100 - It oscillated with 150R and was stable with 1K ( I did'nt try values in between).

Oh! and what happened to the Music Angel EL34 I was modifying on an earlier post in this thread - I abandoned Pentode Mode and have wire the thing for Triode Mode - why? So that I could avoid using any global feedback.
Cheers,
Ian

 

[Miles Prower]

I think I also surmised that since our tubes are good up to well into the MHz region, the inductance of even short hook-up wiring and the G2 capacitance to somewhere might set up the scene for a nice r.f. oscillator. Even a few tens of ohms will swamp the necessary Q though.... so, intuitively, I would suggest that we keep them in!

What sayeth others?

Regards.

Indeed. Stray and device capacitance, and parasitic inductance can, and does, lead to RF oscillation. I had that very problem: a damped oscillation at some 20MHz before applying any feedback on signal peaks. A plate stopper wasn't quite enough, and it took the installation of screen stoppers to prevent that. After adding these, there was no problem with RF instability with both local and global NFB.

Even if stoppers sort of undoes screen regulation, that's better than having a full-blown RF oscillation.

 

[ray_moth]

You will recall that this point was raised before on the "6L6p.p. power tranny" thread (my post#16 there).

I didn't recall it. Sorry, my fault, I should have.

Back in the 60's and 70's the British method was to use a common screen resistor feeding g2 on both sides of the Push Pull. This acted to keep the g2 voltage reasonably constant (As one side drew more screen current the other side would be drawing less).

That's a clever observation! It explains why Mullard, for instance, insisted on a shared screen resistor in data sheets.

Even if stoppers sort of undoes screen regulation, that's better than having a full-blown RF oscillation.

Yes, indeed!

Thanks Johan, Gingertube and Miles for your answers, which pretty well agree with what I thought. I don't think the same issue applies to UL connection, though. The stabilised screen thing appears to be solely a pentode-mode concern.

 

[Johan Potgieter]

I don't think the same issue applies to UL connection, though. The stabilised screen thing appears to be solely a pentode-mode concern.

...... yes, when one comes to think of it. Utilising G2 as a sort of control grid, which action causes the plate impedance to drop "all the way" from pentode value to almost that of triode, would certainly diminish any propencity for activity in the radio wave regions; some Miller effect and all that.

Miller effect! There is some of that in a UL stage in contrast to pentode (fixed Vg2) operation. What are the relative frequency responses of pentode, UL and triode operation under the same conditions? Effect of g1-stopper value related to the same? Never investigated that - more homework for me!

But then (referring to my 1st paragraph) the Williamson used g2-stoppers even for triodes. Did he find it necessary, and where are we then? Help!!

Regards.