distortion with Class B (with NFB) vs Class AB

Well there's something you don't see every day!

:D

Why would you bias the transistors on? Seems the op-amp can handle the current during the crossover period of maybe 10 millamps and generate a volt across the 100 ohm resistors and turn on the boosters. Shouldn't this circuit be a true class B with zero bias?

--- news://freenews.netfront.net/ - complaints: snipped-for-privacy@netfront.net ---

No, true class B has no bias, in tis, the bias is the normal quiescent curr ent drain ot the IC. the E to B resistors set that bias, and tha tis not th e best idea because quiescent current of an OP AMP IC might not be the most tightly controlled parameter of the device. Why would it be ? If it has a pushpuill output stage, then its bias comes into play, and if it si highter but withint limits, fine. If it is lower within limits fine.

I consider this design pretty much no good. Maybe good for cheapness, but, well, not even for a headphone amp. Piece of shit is what some would say. O thers kust mean it.

If you could trim the internal bias of the OP AMP, maybe. People, evne if i t has zero offset, that doesn't mean that if it is groumded anywhere that t he current drain from the positive and negative rails is going to be equal.

Fact is, the whole design is so shitty I woulldn't even do it with discrete components. there are so many more ways. The problem is tha these websites and shit that supposedly teach you how to design do not. thewy teach you h ow to use ICs.

The IC designers are the omes who design, you jusr mount the tires. I can d esign the IC, in configuration. I canot do anything about the die, wich oth ers can. I can give you an asskicking design for an OP AMP you do with four transistors and about six other components. Nothing with more than three t emrinals.

Wanna see ?

Sure. :)

-J

rrent drain ot the IC. the E to B resistors set that bias, and tha tis not the best idea because quiescent current of an OP AMP IC might not be the mo st tightly controlled parameter of the device. Why would it be ? If it has a pushpuill output stage, then its bias comes into play, and if it si hight er but withint limits, fine. If it is lower within limits fine.

, well, not even for a headphone amp. Piece of shit is what some would say. Others kust mean it.

it has zero offset, that doesn't mean that if it is groumded anywhere that the current drain from the positive and negative rails is going to be equa l.

te components. there are so many more ways. The problem is tha these websit es and shit that supposedly teach you how to design do not. thewy teach you how to use ICs.

design the IC, in configuration. I canot do anything about the die, wich o thers can. I can give you an asskicking design for an OP AMP you do with fo ur transistors and about six other components. Nothing with more than three temrinals.

Sure, I wanna see. But I still don't see why the transistors need to be bia sed on (collector current flowing). From the spec sheet, the op-amp supply current varies from 4 to 8mA which yields 400 to 800 millivolts across the E/B 100 ohm resistors and may or may not turn on the transistors. Why not t rim the E/B resistors to maybe 50 ohms so the E/B voltage is half that and the transistors are definitely off (no collector current)? And at say 15 mi lliamps of op-amp load current, the E/B voltage is 0.75 which is enough to turn on the transistor before the op-amp runs out of gas. Wouldn't that be better for thermal runaway problems?