Active tone controls?, how much positive feedback?

[Satamax]

Hi Everybody!

Well i'd like to make a kind of baxandall active preamp. With tone controled negative feedback from plate to grid of V1, then the signal out of that first valve would be also feed to the second valve and since second valve's plate is in phase with first valve's grid, i would have positive feedback giving me amplification at the frequencies cut by the bax. Or am i completely out of my mind regarding the active tone controls. My big question is, how much positive feedback am i allowed?

Bye.

[Don Moose]

Well, I wouldn't feed back to the grid of the very first gain stage - that would do some odd things to your input impedance, and send signal back up your instrument cable.

Next, tone-controlling plate-to-grid feedback on a triode, and sending that plate signal to another triode is not positive feedback - in phase or not. It is an active tone control, though.

I wonder if you need to wire the pots in reverse - they're going to work backwards. Adding treble in the feedback reduces treble in the output (you knew that).

Anyway - hope this helps!

[Steve Conner]

Have you seen my Baxandall tone controls? @ http://scopeboy.com/baxeq.html

[Satamax]

Thanks a lot guys for your replies.

Well, i was thinking more along thoses lines.



The main problem is isolating between positive and negative feedback. Because i don't realy want negative feedback on V2 grid.

[Old Tele man]

re: "The main problem is isolating between positive and negative feedback. Because i don't realy want negative feedback on V2 grid."

...add a bypass capacitor across V2 cathode resistor, of course it'll approximately double V2's gain, but 'tapping" the output from its plate (instead of across the whole Ra) will enable you to scale back the added gain.

[Satamax]

re: "The main problem is isolating between positive and negative feedback. Because i don't realy want negative feedback on V2 grid."

...add a bypass capacitor across V2 cathode resistor, of course it'll approximately double V2's gain, but 'tapping" the output from its plate (instead of across the whole Ra) will enable you to scale back the added gain.

Thanks a lot Old tele man for your reply.

Could you explain?I don't have enough tech knowlege to see why using a cath bypass cap would eliminate the litle neg feedback aplied to V2 grid. Doubling the gain would be intresting for fighting against the negative feedback. But anyway, i was thinking the litle negative feedback i would have on the grid would may be counteract a bit the tendency of going into oscilation of the whole circuit, due to the positive feedback aplied to V1 grid. I know my schemo is incomplete, there's no pot on V1 negative feedback loop Which would be needed for the cut control. Obviously the boost control is the pot from V2 plate to V1 grid. and obviously i need a resistor in series between the pot and the grid of V1. I'm wondering how much positive feedback i can apply to V1 grid. And i also willingly ommited the two cathode bypass caps just to save on the parts count of superspice, but i was thinking of using some.

Thanks a lot again.

Bye.

Max.

[Old Tele man]

...an "unbypassed" cathode resistance introduces negative-feedback voltage (NFB) which greatly reduces the normal tube/circuit gain...sometimes up to 60% but usually about 50%.

...thus, "adding" a bypass capacitor across the cathode resistor (which STILL generates a cathode bias DC-voltage) eliminates the NFB (if capacitor is sufficiently big enough to "fully bypass" Rk) and the 50-60% loss in tube/circuit gain is restored.

...because a 2X increase in gain might be too much, the "tapped" plate load suggestion provides a way to reduce the gain back down to where you want it to be...how much THAT is I have no idea, it's your design...was just a suggestion.

[Satamax]

Thanks a lot Old tele man.

Anybody else on the isue about how much positive feedback i can use in an active tone control system like this one?

Thanks a lot.

Max.

[Steve Conner]

I'd love to answer, but I have no idea how your circuit works or even if it would work. I can't even see where the input is meant to be.

[Satamax]

Sorry Steve, i know the circuit is badly drawn. Basicaly input is on the grid of V1 at the left. And output is taken from the plate of V2 on the right.

The principles. The baxandall filters some frequencies taking them somewhat out - i can't remember how much for a bax. The frequencies that it lets through then go normaly to V2, but also go back to V1's grid in a frequency controled NFB. I forgot to put a pot on that line, but it can completely eliminate the frequencies that the bax has let through. But obviously the volume isn't too big, and you need some volume. The next part, the second valve receive whatever is left through the bax and first valve "filter" amplifies it, and then it's going to the next stage. But to be able to amplify even more thoses frequencies can be fed back to the first valve's grid, a verry litle amount obviously, as grid of V1 and plate of V2 are in phase, and it would turn into a squealing oscilator if i'd feed back too much. Realy it's to fight against the losses created by the "tone controls" which are in between the two valves. IIRC it's been done before. And regenerative feedback seems to have been common practice in radio receiver to amplify signal too.

Am i making this understandable?

Thanks for your interest.

Bye.

Max.

[Steve Conner]

Well, it's very nicely drawn, I'm just not sure if it's well thought out! :P

The Baxandall tone stack can also be used as a regular passive tone stack with no feedback. Look at the 12AX7 channel of the original Ampeg SVT for an example.

http://users.aol.com/portaflex/schems/svtpream.gif

If you use it in this way, you'll have the same Baxandall response, and as much gain as possible, since there's no negative feedback anywhere. If you then put another tube amplifying stage after it, you get even more gain!

If that still isn't enough, you can then start playing with positive feedback from the plate of the second tube to the grid of the first one. According to the Nyquist criterion for stability, the maximum amount you can apply is roughly equal to the product of the two tube stage gains, divided by the lowest possible insertion loss (both bass and treble knobs at 10) of the Baxandall stack.

From a simple inspection of the Bax circuit, the insertion loss approaches 0dB (a gain of nearly 1) at very high and low frequencies, so if each tube stage has a gain of 50, you can't use more than 1/2500 worth of positive feedback.

[Satamax]

Thanks a lot Steve.

That's the first time i hear about nyquist criterion. I'll read furter about this whenever my left eye gets better, as i've had a litle prob with cast iron grinder dust today. Making a tenoning machine http://satamax.free.fr/alive.jpg If only i could be as dedicated making amps and and learning electronics as i am with mechanics or wood working

Thanks again, you're a star.

Bye.

Max.

[Satamax]

Steve, i've modified the schemo.