Class-A Opamp with output stage

[EWorkshop1708]

Here's a schematic I made and I'm about to construct the actual circuit.

It will be used as a part of a preamp for a diy sound system. I added an output stage to drive low impedance loads without bogging down the OpAmp. It will be driving a 560 ohm load (driving inverting input on power amp)

To preserve clarity, for my stereo L & R channels, and have less %THD, I want to do class-A preamp, with a good op-amp that has low THD as well.

Any opinions or tips, or questions about the circuit, feel free to post.

-Eworkshop1708

[jaycee]

Should work quite well. Also works well as a headphone amp.

[jcx]

you'd have less AC load on the op amp output if you connected the bias R to the output Q emitter - the Vbe will be relatively constant and "bootstrap" the R to appear as a much higher impedance load to the op amp output - even after lowering the bias R value to give the desired current with the Vbe drop instead of the negative supply V

[EWorkshop1708]

Originally Posted by jcx

you'd have less AC load on the op amp output if you connected the bias R to the output Q emitter - the Vbe will be relatively constant and "bootstrap" the R to appear as a much higher impedance load to the op amp output - even after lowering the bias R value to give the desired current with the Vbe drop instead of the negative supply V

Interesting.........Thanks for the tip, I've made the change to the circuit.

I lowered the Class-A bias some on the OpAmp, but it should still remain in Class-A still because the 470 should still bypass enough current to keep the OpAmp biased on, even if the transistor is not loaded, because of the constant DC load from the CCS Class A output.

[Bonsai]

Nice! I did somee work on a class A opamp based GP output stage a few months ago, and managed to get some measurements done on an AP. The results were really good, so I can vouch for this approach. You can read about my results here:-

Audio Amplifer Designhifisonix.com | hifisonix.com

see 'A Universal Small Signal Class A Buffer'

Good luck with your project :-)

[discrete]

Originally Posted by EWorkshop1708

.
I lowered the Class-A bias some on the OpAmp, but it should still remain in Class-A still because the 470 should still bypass enough current to keep the OpAmp biased on, even if the transistor is not loaded, because of the constant DC load from the CCS Class A output.

Why not use a dedicated (and adjustable) CSS for the op-amp like in Bonsai's circuit, or this oldie from Walt Jung (Qn/D2/R?)?

http://waltjung.org/PDFs/WTnT_Op_Amp_Audio_2.pdf

(That is also a well performing op-amp buffer, able to run in class A. The four articles of the series are here:

Services )

[AndrewT]

I like the bootstrapped resistor (ClassA single ended opamp). I had not thought of that as a bootstrap, but seems so obvious when one is told.

The 470r draws ~1.2 to 1.4mA through the opamp. Will all opamps, those with high output bias and those with low output bias, perform similarly with that amount of single ended output current?

[Dr.EM]

Looks nice in a quick Spice sim with 3904s and 32Ohm load, 2Vpp. How can I perform a distortion analysis? I think this will be useful as a headphone amplifier.

[jcx]

some "audio" op amps appear to have ~ 100 uA output bias

some early generation DSL driver op amps may have a few mA - they were competing on MHz distortion specs - Scott Wurcer claims newer DSL chips are accepting a higher distortion number as adequate and trying to cut power dissipation

how "good" the Class A performance for a given bias can be easily calculated assuming op amp internal emitter follower and added buffer Q base current demand

[Conrad Hoffman]

I've used a very similar circuit for a headphone amp with great results. If you use a MOSFET for the pass element, the offset voltage on the gate means that the opamp output doesn't pass through zero until you get to larger signal levels, IMO a good feature if you can stand the slight loss of headroom. Heatsink the circuit well, bias it high enough, and you can even drive small speakers.