how to calculate feedback resistor

[Lico]

The other day i put my amp in pentode mode by moving the strapping resistor to B+ (250v.) It doesn't sound all that great, but I wasn't expecting much out of it. I'd like to try adding some negative feedback but I'm not sure exactly how to do it or what values to use. I drew in what I think is the right circuit based on other schematics i've seen. Is there an equation as to how much resistance or is it just trial and error?

[Sch3mat1c]

Easiest is trial and error with some nice equipment. Even so it's a can of worms each for LF, HF and stability parameters when you add NFB. However you may not be helpless here, first of all what are the 1st tube and OPT?

Tim

[cogsncogs]

Hello

To use NFB you have to make some cicuit changes. As shown in your schem one side of the FB resistor is connected to ground and that won't work. I'm posting your diagram re-configured to use feedback. I would say a 10k FB resistor is about as low as you could go. I would start at 22k. A 50k pot would be good to use here to adjust the feedback, along with a scope and so on. Trial and error seems to be your only hope here!

Cheers
Wayne

[Eli Duttman]

Before adding NFB, improve the linearity of the "final". When the screen grid is connected to B+ via a resistor, it has to be grounded via a cap. Even then, the circuit is not particularly linear, as there are times when the instantaneous plate potential is less than that of the screen grid. "Pentodes" are most linear when the screen grid B+ is tightly regulated at about 50% of the plate rail. A 105 V. gas regulator tube will do nicely for the screen grid B+.

If you use single driver speakers, NFB might not be necessary. As soon as crossovers enter the picture, the poor damping factor of "pentodes" indicates the use of NFB. Notice that the NFB is being used primarily to lower the O/P impedance; distortion control is a secondary benefit.

[cogsncogs]

Originally posted by Eli Duttman:
Even then, the circuit is not particularly linear, as there are times when the instantaneous plate potential is less than that of the screen grid.

That's one of the reasons why I don't like using pentodes SE. But it can be done. Remember those old tube table clock-radio's using a 50C5? Got one on my front porch that still works! And yeah I know the sound isn't first rate, buzzz hmmmm.

Wayne

[Sch3mat1c]

Originally posted by Eli Duttman
Before adding NFB, improve the linearity of the "final". When the screen grid is connected to B+ via a resistor, it has to be grounded via a cap.

What's peak screen current, 20mA? Even if it were as high as 50mA, it would experience only 5V sag. This 2% change would go completely unnoticed by the tube.

Even then, the circuit is not particularly linear,

No duh, it's a pentode!

as there are times when the instantaneous plate potential is less than that of the screen grid.

Duhhhhhhhhhh?

"Pentodes" are most linear when the screen grid B+ is tightly regulated at about 50% of the plate rail. A 105 V. gas regulator tube will do nicely for the screen grid B+.

Ok, ok, so how do you increase plate current in return? Positive grid voltages? (Class A2) Erm...maybe not...

Consider that the 6V6 was BUILT to run at 250/250, and it does it with aplomb.

Tim

[Eli Duttman]

Consider that the 6V6 was BUILT to run at 250/250, and it does it with aplomb.

That is not the only "textbook" operating point. 315 V. on the plate, 225 V. on the screen grid, and -13 V. on the control grid yield 34 mA. of plate current and 5.5 W. power O/P.

When lowish regulated screen B+ is employed, the plate rail voltage has to go up to get an appropriate plate current.

[Sch3mat1c]

And that still leaves a range of 190V in which the plate is below screen potential. Alas, few tubes have such perveance that screen voltage can be taken down as low as saturation voltage, without grid current. (4X150A is one such tube. It saturates at 300V.)

When lowish regulated screen B+ is employed, the plate rail voltage has to go up to get an appropriate plate current.

If you drop screen voltage within the realm of a gas regulator tube, that is, 150V, plate voltage will have to be increased dramatically, in fact past ratings, to regain the power output. Impedance will also suffer dramatically (can you say 10k SE OPT?). Alas, it is notoriously useless to try to increase plate current by increasing plate voltage.

Tim