feedforward vbe compensation in class-A

[tifojo]

Its a long time since I've read such tortured logic to complicate
what is essentially a very simple design problem by adding in
engineering terms non-sensible requirements.

sreten:

if there's a flaw to my idea, it's that it's childishly straightforward. if the distortion from a single-transistor emitter follower is from non-linear loading and vbe, let's just make the current in the output transistor constant. it's stupid, it's wasteful, it probably wouldn't sound better, but whatever, how could I know without trying? and all I was asking was, "did someone else try this?" to perhpas save myself the trouble.

and simple? nothing in audio is a "simple design problem."

Concern for dynamic range is fine, inconsistent logic isn't.
IMO 106dB maximum with an inaudible noise floor is better
than 126dB maximum with an all too audible noise floor.

This is not a question of logic, it's a question of physics...will the noise be audible, or won't it?

I have listened to an emitter follower with a resistor load to V- that gives about 0.19 watts into the sennheisers with 0.3 amps of quiescent current and no audible noise. For +-32 V supplies, I can use that same quiescent current in a 100 ohm resistor to V-, giving a negative output swing of 24 volts and my design goal of 1 watt. probably still inaudible noise.

Concerns for transients that simply don't exist on most
recordings is also fairly pointless. The very nature of
setting CD "average" levels means they will be limited,
except for the few brave souls allowing high peak to
average levels on CD, but this by definition means
average resolution is a lot less than 16 bits.

And that's ok? If what you want is good performance for the "average" recording, what you'll get is an "average" amplifier. My goal is to design for the best recording I can find.

Anyway, I've had it up to here with this thread. I'll just use a resistor, waste the 20 W per channel, and forget about you all as I listen to its beautiful sound.

 

[tifojo]

that was a 40 ohm resistor to V- in the other amp, by the way. so maybe the noise is up 6 dB at worst, if it was dominated by current noise in the transistor muliplied by Re. but double inaudible is (hopefully) still inaudible.

 

[sreten]

By all means use 20w diisapation to drive your headphones if you so choose.

The White follower (used by D.Self in a phono preamp, and the
basis of the JLH, Sugden and DoZ class A designs) is a form of
push pull class A but it only modulates the current by 50%, not
the 100% you require.

I'd be very interested in seeing how you achieve constant current
operation for the output transistor whilst providing output voltage
swing via your current sensing arrangements.

sreten.