I was about to answer the OP (Class A appears popular among the Nelson Pass crowd, such designs are easy to make and have low parts count even though some parts are really big like the heat sink), but I just noticed it's from 15 years ago.
I was under the impression that a great number of opamps work in class A when lightly loaded. How lightly is easy to approximate from the current consumption.
a small fraction of a ma is typical, people even brag when they use as much as 500 ua output Q bias current in monolithic op amps
there are a few DSL driver op amps that use several ma, or have output bias current adjust
generally if you want more you have to pull on the monolithic op amp output with an external css
there are a few DSL driver op amps that use several ma, or have output bias current adjust
generally if you want more you have to pull on the monolithic op amp output with an external css
Last edited:
I've seen circuit designs using IC op-amps in which the designer recommends loading the op-amp output with a resistor to V- to keep the NPN output transistor ON and the PNP output transistor OFF. The goal is to make the output operate class A. I don't know how effective it is. I am not recommending this, just reporting it.
I like to compare class "A" to driving your car with the engine at high rpm while using the clutch to control the speed. The result is heat.
Tom
I like to compare class "A" to driving your car with the engine at high rpm while using the clutch to control the speed. The result is heat.
Tom
Complementary Symmetry stage topologies preclude class-A operation. In fact, that is the most common bias class in all audio amplification stages aside from those of power amplifier final output stages. In addition to half the output impedance (twice the transconductance) complementary symmetry stage topologies also inherently suppress even-order distortion.
Last edited:
Most drum machines use samples, and probably high def, there not synthetic. There recordings of great drums.
yes that's one that brags about 500 ua in the output Q
much of the rest is input Q bias for the nV noise, multiplied by folded cascode, bootstraped mirror VAS
much of the rest is input Q bias for the nV noise, multiplied by folded cascode, bootstraped mirror VAS
Thanks for this! Always wondered what fraction of the total current goes towards the o/p. Would this make a good argument towards forced SE operation?
> Do not push-pulls introduce crossover distortion
No.
Getting out of Class A makes crossover. You can easily design push-pull to never leave A (at rated output/load).
But much of the "advantage" of push-pull is so you CAN slip into B, have low idle current with high peak current.
And read Doug Self. He argues (I tend to agree) that properly done near-B (very small A bias) can be *cleaner* than class A. It is not a black/white question.
No.
Getting out of Class A makes crossover. You can easily design push-pull to never leave A (at rated output/load).
But much of the "advantage" of push-pull is so you CAN slip into B, have low idle current with high peak current.
And read Doug Self. He argues (I tend to agree) that properly done near-B (very small A bias) can be *cleaner* than class A. It is not a black/white question.
I think you're consfusing two things. There is "single end vs. push-pull" and "Class A vs. Class AB". They are two different issues. To produce a linear output, a single ended amp has to be class A. But a push-pull amp can be class A, too, or it can be class AB. Push-pull class A has some technical advantages over single ended class A, but a lot of people seem to be into single ended these days.
Nearly all those circuits, even many using ICs will actually be operating in class A mode when driving light loads. Class AB amps operate in class A until the output current called apon is high enough then the operation transitions into class AB. For example drive about any line level output into 100k ohm load like some preamps or power amps present, and with a 1Vrms (high) audio output sinewave level, the peak output current would be all of 14 microamps! Pretty much any class AB output stage is going to be biased with more idle current than that (and it only needs half that for push-pull to stay in class A).
Edit: oop! I now see there were already an other page of replies!
Nearly all those circuits, even many using ICs will actually be operating in class A mode when driving light loads. Class AB amps operate in class A until the output current called apon is high enough then the operation transitions into class AB. For example drive about any line level output into 100k ohm load like some preamps or power amps present, and with a 1Vrms (high) audio output sinewave level, the peak output current would be all of 14 microamps! Pretty much any class AB output stage is going to be biased with more idle current than that (and it only needs half that for push-pull to stay in class A).
Edit: oop! I now see there were already an other page of replies!
Last edited:
why-class-so-popular
The opener's question way back was "why is classA so popular?
Well it's not.
If one were to survey all the amplifiers past and current and find which ones are in Class A, I'll take a guess that less than 0.01% of the world population of power amplifiers (including headphone driver amps) are ClassA.
That means that >99.99% of low and high power amplifiers are NOT classA !
I suggest that (imaginary survey) proves that ClassA is NOT popular.
We buy and use ClassAB and ClassD and switching amplifiers in massively greater numbers.
The opener's question way back was "why is classA so popular?
Well it's not.
If one were to survey all the amplifiers past and current and find which ones are in Class A, I'll take a guess that less than 0.01% of the world population of power amplifiers (including headphone driver amps) are ClassA.
That means that >99.99% of low and high power amplifiers are NOT classA !
I suggest that (imaginary survey) proves that ClassA is NOT popular.
We buy and use ClassAB and ClassD and switching amplifiers in massively greater numbers.
Class A is a waste of time and money, just go for AB it's powerful, very high performance and easy.
EDIT: As for the reason I think it's popular. You need to remember in this forum that peoples main hobby is likely to be making the gear rather than sound quality so folk lore, mystery and making a rod for their own back is high on the list of priorities. This is so their hobby is sustained; because audio electronics is easy it's possible to make a very good sound system that is entirely satisfactory in not a lot of time this defeating the point of the hobby in the first place. In audio people seem to be attracted to the magic because I think it makes them feel a lot better when something turns out to sound good. I prefer working hard, learning about physics and actually using my brain to construct something that works and then the pay off for me is huge.
EDIT: As for the reason I think it's popular. You need to remember in this forum that peoples main hobby is likely to be making the gear rather than sound quality so folk lore, mystery and making a rod for their own back is high on the list of priorities. This is so their hobby is sustained; because audio electronics is easy it's possible to make a very good sound system that is entirely satisfactory in not a lot of time this defeating the point of the hobby in the first place. In audio people seem to be attracted to the magic because I think it makes them feel a lot better when something turns out to sound good. I prefer working hard, learning about physics and actually using my brain to construct something that works and then the pay off for me is huge.
Last edited:
A. Everything else being equal, class A amplifiers distort less than amplifiers of any of the other classes. They especially generate far less high-order distortion (high-order distortion becomes audible at lower levels than low-order distortion, and when it does, it also sounds worse). Class A amplifiers also have no severely distorted currents flowing anywhere, so there is no need to worry about magnetic coupling from the wires carrying those currents to other wiring in the signal path.
B. Stating essentially the same the other way around, for a given distortion level, there is far less need for error correction techniques (negative feedback, error feedforward, whatever) in class A amplifiers than in the other types of amplifiers. Extremely simple amplifiers can therefore produce reasonably acceptable results.
My guess is that point B is what attracts DIY audio enthusiasts most.
B. Stating essentially the same the other way around, for a given distortion level, there is far less need for error correction techniques (negative feedback, error feedforward, whatever) in class A amplifiers than in the other types of amplifiers. Extremely simple amplifiers can therefore produce reasonably acceptable results.
My guess is that point B is what attracts DIY audio enthusiasts most.
I support your view fully. Nothing out there sounds like a single ended class A. I have experience of audiophiles hearing their first performance of a Single Ended, low power class A and don't want to give it back.
Furthermore, I tend to listen at much lower volume using an SE ClassA than I did with other amps since this kind of am just resolves so much more material in the performance that you do not have to turn the volume to earsplitting levels to hear the essence of the performance.
In my opinion John Lindsley Hood offering of a Pushpull class A was the closest thing I heard to single ended. If you build and listen to one you will donate your current amplifier to someone in the learning.
You Dutchies will understand my synergy, dit verskil is soos kak en kaas! There is no better definition.
Single Ended Class A when done right, sounds natural, engaging, non-fatiguing. I prefer it to all my other amps (and I have about three dozen). The difference is night and day and once you hear it - you won't want to go back. And if you haven't heard it, you can't imagine what you are missing.
I took Mooly's virtual listening room test a while back, I don't know how valid the simulations are, but of the three amps I prefered the JLH. I like the simplicity of the design and I think if I did build another amp I'd try one
Doesn't everyone remember how basic radios would sound great at extremely low volume?
I was so much smarter before my teenage years...
Thanks, I may try to put together a JLH sometime, and check it out.
I was so much smarter before my teenage years...
Thanks, I may try to put together a JLH sometime, and check it out.
Here's another virtual audition made with a mic and different amps.
http://www.diyaudio.com/forums/solid-state/295286-virtual-audition-very-simple-quasi-mosfet-amp.html
> The opener's question way back was "why is classA so popular? Well it's not.
If one were to survey all the amplifiers past and current and find which ones are in Class A, I'll take a guess that less than 0.01% of the world population of power amplifiers (including headphone driver amps) are ClassA.
With you up to "past".
1920-1963, 99% of home radios, phonos, and TVs were one Class A power bottle. Even at only one radio per home, that's many-many millions of class A outputs. True, post-1963 history is not only transistors, 99% working near B, but also stereo and 5-channel, and multiple radios/etc per home. Still I suspect that "all the amplifiers past and current" run nearer 10% class A.
A is popular when you can barely afford ONE output device. As when selling 5-tube radios at $9.98.
If one were to survey all the amplifiers past and current and find which ones are in Class A, I'll take a guess that less than 0.01% of the world population of power amplifiers (including headphone driver amps) are ClassA.
With you up to "past".
1920-1963, 99% of home radios, phonos, and TVs were one Class A power bottle. Even at only one radio per home, that's many-many millions of class A outputs. True, post-1963 history is not only transistors, 99% working near B, but also stereo and 5-channel, and multiple radios/etc per home. Still I suspect that "all the amplifiers past and current" run nearer 10% class A.
A is popular when you can barely afford ONE output device. As when selling 5-tube radios at $9.98.
- Status
- Not open for further replies.