Current feedback Mosfet Circlotron

Sébastien, I deleted 33 p and 100 p. The square wave just got better. All parameters remain the same. Now listening to Patricia Kaas. I will spend several days observing the parameters and listening to music.
Great circuit!!!: :up::cheers:
Square wave - 32 kHz:
 

Attachments

  • DSC02317.JPG
    DSC02317.JPG
    108.4 KB · Views: 337
Member
Joined 2001
Paid Member
Joe what do you think about this?

Very interesting result, thanks for sharing. This is just the kind of detail you can only sort out by building and listening, I think.

In theory, the cap is there to keep common mode supply noise out of the front end. However, I've seen in simulations that removing the cap actually reduces even-order harmonic distortion. That could be what you're hearing, or it could be that removing the cap cleaned up the sound by removing supply noise that the cap was coupling into signal ground.

Either way, I'm OK with removing the cap if it improves the sound, as long as it doesn't result in common mode oscillation. You can check for that by watching either output terminal to ground. If you see oscillation, you could try jpatay's suggestion of using a smaller cap value in place of the 100 uF.
 
Member
Joined 2001
Paid Member
It won't work as shown because the current from +32V returning to ground through R17-R18 puts a positive DC voltage on the output rails, which turns off J1-J2. You may be better off with a completely separate front-end supply, which is really a different design approach.