My differential input 49830

[Phil Olson]

I basically copied their schematic but added a differential input. I decided to do four pairs for possible sub amp use but would start with one pair and add them if power requirements warrant.

Alothough I have the IR FET's there I am considering the Toshiba's mentioned in the 4702 app note.

One thought I had was to drop back to two FET pairs and parallel two boards for the sub. I would just swap the differential pair of one to basically do a bridged design. That's what the BPA-300 design essectially does with seemingly excellent results.

Any suggestions for improvement?

 

[SpittinLLama]

Some of the final values will depend on the PCB layout. The compensation capacitor and gate resistors, for example. Be prepared to adjust them for full stability. The bias may also need to adjust depending on where you want to bias your design. Be sure the drain supply bypassing is close to the FET as possible.

-SL

 

[Phil Olson]

Thanks! I was counting on doing some tweaking. My guess is that going from one FET pair to two or more may also require tweaking.

 

[Mitch Alsup]

Zobel network/filter at the output.

 

[Phil Olson]

Thanks, I forgot to add that.

If that's the worst of it I'll have to pat myself on the back!

 

[pitone75]

Very good circuit!!!

 

[djk]

The source resistors aren't big enough to do any real good.

I would consider using an FET for the bias thermal compensation device, it will have a better tempco than a BJT for this circuit. C14 needs to be much bigger than 10pF, I would prefer 10µF.