My end goal is:
A variant on the Big MOFO or similar with a frontend would be perfect except it has a low output impedance. I could wrap a current feedback loop around it to raise the output impedance but if I've understood correctly I'd have to use a lot of feedback to get my desired output impedance. And in addition to that I don't need the distortion reduction of the feedback, it already performs good enough. But could I do the F7 trick but in reverse? That is I use a modest amount of positive voltage feedback to raise the output impedance by a lot in addition to a normal negative current feedback loop to prevent instability.
I read somewhere that Nelson wrote that the F7 positive current feedback was 2 dB and reduced the output impedance by 30x. If the same trick works in reverse with positive voltage feedback then I could get by with less overall feedback to get my desired output impedance which would be great.
And so I threw together this circuit. If I've understood correctly the drain pins are inverted relative to the gate but the source is not. Hence I believe that the circuit will apply negative current feedback and positive voltage feedback. Unless I messed up somewhere which is pretty likely 😀
And as I write in the example I'm not sure if Q2 should be connected to Gnd or the current sense. My gut says if it has the same ground as Q1 it won't apply any positive current feedback but I'm no expert so I might be wrong =)
- Single ended with choke CCS
- 40W into 16 ohm @ 1% THD.
- 100 ohm output impedance or more
- At most 100W dissipation
- As little feedback in loops as I can get away with
A variant on the Big MOFO or similar with a frontend would be perfect except it has a low output impedance. I could wrap a current feedback loop around it to raise the output impedance but if I've understood correctly I'd have to use a lot of feedback to get my desired output impedance. And in addition to that I don't need the distortion reduction of the feedback, it already performs good enough. But could I do the F7 trick but in reverse? That is I use a modest amount of positive voltage feedback to raise the output impedance by a lot in addition to a normal negative current feedback loop to prevent instability.
I read somewhere that Nelson wrote that the F7 positive current feedback was 2 dB and reduced the output impedance by 30x. If the same trick works in reverse with positive voltage feedback then I could get by with less overall feedback to get my desired output impedance which would be great.
And so I threw together this circuit. If I've understood correctly the drain pins are inverted relative to the gate but the source is not. Hence I believe that the circuit will apply negative current feedback and positive voltage feedback. Unless I messed up somewhere which is pretty likely 😀
And as I write in the example I'm not sure if Q2 should be connected to Gnd or the current sense. My gut says if it has the same ground as Q1 it won't apply any positive current feedback but I'm no expert so I might be wrong =)
Attachments
Last edited:
I don't know if it's a good idea having positive feedback between 3 stages. Is there any reasons that you just do not build a simple positive feedback loop from output to Q2 source?
Probably not.
And on positive feedback to Q2 source mostly that I was not sure if it would work. If it does then it is probably a more simple solution.
This sort of thing can very easily and quickly be simulated in LTspice. Get your answers in an hour, max, and play with configurations and values.
You should really try that.
Jan
You should really try that.
Jan
For starters, you want to look at your polarities. The feedback to Gate of
Q1 is positive voltage feedback. To Source of Q1, is negative current
feedback. To Source of Q2 is positive current feedback. For completeness
maybe you want some negative voltage feedback to Gate of Q2 -
perhaps you could adjust all the values until they all cancel each other.
