Need help finding a circuit

[MrGuitardeath]

I've been trying to hunt down a simple push pull amp schematic that will also yield a high level of harmonic distortion. Preferably 10%.
I've had absolutely no success beyond finding a few simple push pull schematics.

Is what I'm looking for out there?

[amplifierguru]

Most amplifier designers try to reduce THD to well below this level (typ 1-2%) before feedback so with 40dB NFB end up with 0.01%.

However an amplifier such as my Simple Killer Amp without it's linearising and open loop gain boosting bootstrap C's and substantial emitter degeneration so there is virtually no NFB available (except at DC for stable operating point) could maybe come close at 1-2%. Best I can do. 10% is a lot - how about a bulb in series with the speaker.

I think NP knows more about generating sympathetic distortions than I do.

Cheers,
greg

[theChris]

simplist way to do what you want is to add diodes into the feedback path. this gives compression and will give higher THD specs when used at high volume. you could do something similar to give a higher gain at low levels to give high distortion over the entire band.

or class C bias.

keep in mind none of these are overly "good" ideas. people tend to avoid building amps with intentionally high THD. the only time i can think of where you'd intentionally do this would be in guitar amps in which case i'd suggest going to some tubes and running them on lower then reccomended supplies to degrade performance enough.

[djk]

A push-pull pair of MJ21193/94 have only 0.5%THD with no feedback, less than 0.05% if you match their gains.

You will want to go single-ended with no feedback to get what you want.

The single-ended design will tend to have more second harmonic distortion (what you are looking for) vs a push-pull having mostly odd harmonics.

It will be very inefficient, so you will want a small amp with high voltage swing to drive a clean push-pull follower with no global feedback.

A switch adding global feedback could be used for a 'clean' sound.

The gain curve on this FET is convex below 100mA, giving it a transfer curve like a triode tube.

http://www.supertex.com/pdf/datasheets/DN2540.pdf

Cost from Mouser is under $2 for the FET. A chasis, supply, and power follower parts for a 100W/4R amp will run about $1 a watt if you can scrounge some stuff.

It would be much cheaper to buy a distortion pedal, but this type of amp would sound very different, and you might have some fun too.