• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Class A vs AB -- Do They Really Sound Different?

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,460
Likes
3,628
Location
San Diego
There is one thing that has not been mentioned on why some Class A amps may "sound different/better" than some Class AB amps. Traditionally Class A amps do not have any thermal protection because they are dissipating maximum power all the time where as most Class AB amps do have thermal protection. Many of these thermal protection circuits do not work that great and unannounced to the user would "protect" the output transistors but distort the sound as they engaged and disengaged especially when playing loud. I am not sure how much better the "computer powered" protection circuits are but this was an issue in the past and could potentially lead to a Class A amp sounding different than a Class AB of similar power.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,882
Location
Santa Fe, NM
There is one thing that has not been mentioned on why some Class A amps may "sound different/better" than some Class AB amps. Traditionally Class A amps do not have any thermal protection because they are dissipating maximum power all the time where as most Class AB amps do have thermal protection. Many of these thermal protection circuits do not work that great and unannounced to the user would "protect" the output transistors but distort the sound as they engaged and disengaged especially when playing loud. I am not sure how much better the "computer powered" protection circuits are but this was an issue in the past and could potentially lead to a Class A amp sounding different than a Class AB of similar power.
The microcontroller based protection circuits we use currently do not have a mechanism where they can 'protect' without something being very obviously wrong. When they do trip, the audio will be muted by disabling the current sources, typically for 10 seconds or until temperature returns to a safe level so it will be obvious that something has happened. When temperatures (or other fault other than a DC fault) are restored to normal, the amplifier will resume operation.
 
Last edited:

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,078
Likes
377
@MakeMineVinyl , you can build absolutely perfect solid state class A of some 30W/8Ohm with distortion better than -130dB and negligible noise. No SE tube would compete neither in noise nor in distortion. And your horn issue is fixed. Please stop fairy tales about <1W tube advantages.
I am interested in such a project for driving tweeters only any further info?
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,882
Location
Santa Fe, NM
I am interested in such a project for driving tweeters only any further info?
I'm sure if you do a search for 'active crossovers' you will find abundant information. Most of these will focus on solid state amplification, but you could just as well go the tube route. Unless you have efficient horn tweeters (>100dB/watt), you will need more than a couple watts - more like 20W for an average dome tweeter.

I would be extremely suspicious of posts like the one you originally quoted. In that example, this is a person I've never met, a person who has never heard my system, a person who has no idea of my background or qualifications, and a person who is totally clueless of the logic I used in assembling my system and my goals. How could he possibly judge something of which he has no direct knowledge? ;)
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,460
Likes
3,628
Location
San Diego
I am interested in such a project for driving tweeters only any further info?

I use these amps to drive the tweeters in a tri-amp system https://neurochrome.com/products/modulus-86. They are very quiet and low distortion although not Class A (they are class AB) but they work very well and the kits and documentation and support are great. Click on the specifications tab to see detailed specs.
 
Last edited:

Ken Tajalli

Major Contributor
Forum Donor
Joined
Sep 8, 2021
Messages
2,234
Likes
2,029
Location
London UK
I am new to this thread, so be patient with me.
1- am I right that some people are confusing class A amplification with single ended amplification?
I mean class A can have a balanced output stage, does not have to be single ended (as the amp circuit by @MakeMineVinyl )
2- The difference between class A and class AB is only in the biasing of the final stage. Class A is absolutely superior in sound quality and in technical terms, it just "will cost you". Not only in money, but in efficiency, power output, heat dissipation, power supply oomph etc.
But superior, it is beyond doubt!
BTW, a class A amp, wastes more energy at rest than when actually working!
Class AB can only sound reasonable if it is highly biased into class A. But the more you bias it to class A region, the less total power output you get and higher heat dissipation.
BTW, to drive a tweeter, you almost must use class A amp to get a clean sound, but as tweeters are usually efficient, no problem! a low powered class A will do very nicely.
I have built a few amps in my time, Tube based and solid state.
Tube is harder, more expensive to get right, but worth it at the end. Tubes have a larger linear operating region.
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,934
Likes
3,167
Location
Ichihara City, Chiba Prefecture, Japan
Hello Ken Tajalli,

I essentially agree with you; throughout my long and intensive amplifier exploration in my multichannel multi-driver multi-amplifier project, I actually experienced and (subjectively) confirmed almost the similar perspectives on amplifier selection as you indicated.

As for my actual evaluation of class-A (Accuphase A-36) and quasi-class-A (Sony TA-A1ES) amplifiers, you would please refer to my posts #265 through #311 on my thread; based on these intensive evaluation in my system at my listening room/environments, I finally selected quasi-class-A Sony TA-A1ES to dedicatedly and directly (without LC network) drive the Be-tweeters (and metal horn super tweeters). You would please find my latest system configuration in my post #416.

I hope your web browser would properly translate this comprehensive page on TA-A1ES and also this spec page into English.

Sony TA-A1ES, rather warm though, behaves in class-A mode in my system for driving only tweeters, and I found and confirmed your point of clean high Fq sound as well as very nice phase characteristics/perspectives giving wonderful 3D sound stage with "complete disappearance of all the SPs".
 
Last edited:

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,882
Location
Santa Fe, NM
I am new to this thread, so be patient with me.
1- am I right that some people are confusing class A amplification with single ended amplification?
I mean class A can have a balanced output stage, does not have to be single ended (as the amp circuit by @MakeMineVinyl )
2- The difference between class A and class AB is only in the biasing of the final stage. Class A is absolutely superior in sound quality and in technical terms, it just "will cost you". Not only in money, but in efficiency, power output, heat dissipation, power supply oomph etc.
But superior, it is beyond doubt!
BTW, a class A amp, wastes more energy at rest than when actually working!
Class AB can only sound reasonable if it is highly biased into class A. But the more you bias it to class A region, the less total power output you get and higher heat dissipation.
BTW, to drive a tweeter, you almost must use class A amp to get a clean sound, but as tweeters are usually efficient, no problem! a low powered class A will do very nicely.
I have built a few amps in my time, Tube based and solid state.
Tube is harder, more expensive to get right, but worth it at the end. Tubes have a larger linear operating region.
Class A is just the method of biasing an output stage. You can have balanced class A or single ended. Class A isn't necessarily superior in sound quality, but in some instances such as in my situation, it is vital to avoid crossover distortion artifacts. Class A is definitely more wasteful of AC power.
 
Last edited:

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
22,122
Likes
22,400
Location
Canada
Class A is just the method of biasing an output stage. You can have balanced class A or single ended. Class A isn't necessarily superior in sound quality, but in some instances such as in my situation, it is vital to avoid crossover distortion artifacts. Class A is definitely more wasteful of AC power.
Linear output class A is great. If going class A don't bother with non-linear amps. Even a moderately powered amp of 75W@8R becomes 300 @ 2. Mileage may vary. :D When using decent speaker(s) and a USA made class A amp I thought the system had more depth and detail with the class A amp when wound up louder. Not sure what you do with your tubes but I just solve issues with more power.
 

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,078
Likes
377
I use these amps to drive the tweeters in a tri-amp system https://neurochrome.com/products/modulus-86. They are very quiet and low distortion although not Class A (they are class AB) but they work very well and the kits and documentation and support are great. Click on the specifications tab to see detailed specs.
How do you regulate the gain for each driver pair?I was thinking of using something like the DENON POA 2800 specs are 123db S/N and 0.002 thd:
https://www.audio-high-store.com/denon-poa-2800/
 
Last edited:

Ken Tajalli

Major Contributor
Forum Donor
Joined
Sep 8, 2021
Messages
2,234
Likes
2,029
Location
London UK
Hello Ken Tajalli,

I essentially agree with you; throughout my long and intensive amplifier exploration in my multichannel multi-driver multi-amplifier project, I actually experienced and (subjectively) confirmed almost the similar perspectives on amplifier selection as you indicated.

As for my actual evaluation of class-A (Accuphase A-36) and quasi-class-A (Sony TA-A1ES) amplifiers, you would please refer to my posts #265 through #311 on my thread; based on these intensive evaluation in my system at my listening room/environments, I finally selected quasi-class-A Sony TA-A1ES to dedicatedly and directly (without LC network) drive the Be-tweeters (and metal horn super tweeters). You would please find my latest system configuration in my post #416.

I hope your web browser would properly translate this comprehensive page on TA-A1ES and also this spec page into English.

Sony TA-A1ES, rather warm though, behaves in class-A mode in my system for driving only tweeters, and I found and confirmed your point of clean high Fq sound as well as very nice phase characteristics/perspectives giving wonderful 3D sound stage with "complete disappearance of all the SPs".
You've been having fun! nice.
Reminds me of myself 30 years ago, when I made my 2nd active speakers.
I used Meridian 105 monoblocks for my two bass drivers perside, and a Meridian 103D (D= dual PSU) stereo power amp to drive my midrange and tweeter. To complete the electronics, I designed and built a simple three way electronic crossover to feed the amps.
The amps were heavily biased into class A operation.
I made a pentagon column pair of speakers with midrange driver having an enclosure of its own at top (Kef B110).
It worked, and sounded good, It is still working at a friends house in Yorkshire!
The Sony amp uses a novel (to me) bias arrangement based on volume control setting, to alter the bias of the output stage according to how much power it assumes it needs - for what purpose? - save on electricity and heat generation!
As cute and clever that it is, I rather not! lets think about it:
- let us assume we are listening to some music with a dynamic range of 30dB (let's keep it real and easy) at an average listening level if 95dB.
- let us assume our speakers have an efficiency of 89dB for one watt (again we keep it real)
- This means we need an average power of 4 Watts from the amp.
- Nick Mason comes along and starts to kick his bass drum on "Run like Hell" from "the Wall" !
- momentarily we may need 120+ Watts to reproduce the transients, but the amp sees that we are on level 4 on the volume control and has adjusted the bias accordingly!
- Amp goes into class B mode on transients.
- Now when Nick hits those hi-hats hard, sound will defo. gets grainy.
When I want sound quality, I WANT sound quality! I will pay for it, in electricity bill, heat built up, equipment cost and wife's nagging!
No compromises!
My OTL tube amps burn over 400 Watts at rest EACH! while producing a max. of 50 Watts output, but on average I am on about 10 Watts per channel.

Sony's approach is cute, class AB is a useful topology, but isn't it horses for corses?
If I need 200 Watts from a moderate size amp, class AB does nicely, if I need 2000 Watts class D does nicely, but at home when I need an average of 10 Watts, only class A will do.
 
Last edited:

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,483
Likes
5,402
2- The difference between class A and class AB is only in the biasing of the final stage. Class A is absolutely superior in sound quality and in technical terms, it just "will cost you".
I don't know about "absolutely superior" - clever design work can get very good class AB amps that are way more efficient, generate a lot more output power, and still sound excellent (and measure pretty well, to boot). In fact, I'm of the opinion that class A makes very little sense as a power amp because of the amount of current you burn just idling. Small signal stages (preamps and what have you) sure, but as power amps? Nah.
BTW, to drive a tweeter, you almost must use class A amp to get a clean sound, but as tweeters are usually efficient, no problem!
I'm... pretty sure that's not the case. Speakers I'm currently listening on use a TDA7293 for the tweeter, and it sounds just dandy.
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,934
Likes
3,167
Location
Ichihara City, Chiba Prefecture, Japan
You've been having fun! nice.
Reminds me of myself 30 years ago, when I made my 2nd active speakers.
I used Meridian 105 monoblocks for my two bass drivers perside, and a Meridian 103D (D= dual PSU) stereo power amp to drive my midrange and tweeter. To complete the electronics, I designed and built a simple three way electronic crossover to feed the amps.
The amps were heavily biased into class A operation.
I made a pentagon column pair of speakers with midrange driver having an enclosure of its own at top (Kef B110).
It worked, and sounded good, It is still working at a friends house in Yorkshire!
The Sony amp uses a novel (to me) bias arrangement based on volume control setting, to alter the bias of the output stage according to how much power it assumes it needs - for what purpose? - save on electricity and heat generation!
As cute and clever that it is, I rather not! lets think about it:
- let us assume we are listening to some music with a dynamic range of 30dB (let's keep it real and easy) at an average listening level if 95dB.
- let us assume our speakers have an efficiency of 89dB for one watt (again we keep it real)
- This means we need an average power of 4 Watts from the amp.
- Nick Mason comes along and starts to kick his bass drum on "Run like Hell" from "the Wall" !
- momentarily we may need 120+ Watts to reproduce the transients, but the amp sees that we are on level 4 on the volume control and has adjusted the bias accordingly!
- Amp goes into class B mode on transients.
- Now when Nick hits those hi-hats hard, sound will defo. gets grainy.
When I want sound quality, I WANT sound quality! I will pay for it, in electricity bill, heat built up, equipment cost and wife's nagging!
No compromises!
My OTL tube amps burn over 400 Watts at rest EACH! while producing a max. of 50 Watts output, but on average I am on about 10 Watts per channel.

Sony's approach is cute, class AB is a useful topology, but isn't it horses for corses?
If I need 200 Watts from a moderate size amp, class AB does nicely, if I need 2000 Watts class D does nicely, but at home when I need an average of 10 Watts, only class A will do.

As you may kindly understand, in my latest configuration of the multichannel multi-driver (multi-way) multi-amplifier system, the sound signal is crossovered by digital software crossover EKIO in upstream, and Sony TA-A1ES receives only high Fq signals above ca. 4 kHz (high pass -12 dB/Oct filter at 6 kHz) from CH7+CH8 of OKTO DAC8PRO. Consequently, the bass kick drum low Fq sound never goes into TA-A1ES.

Even though I rather prefer pure class-A Accuphase A-36 to quasi class-A Sony TA-A1ES in this configuration, the cost performance of TA-A1ES plus the results of actual intensive listening comparison between the two, however, were the major deciding factors in my recent amplifier selection.
 

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,078
Likes
377
I don't know about "absolutely superior" - clever design work can get very good class AB amps that are way more efficient, generate a lot more output power, and still sound excellent (and measure pretty well, to boot). In fact, I'm of the opinion that class A makes very little sense as a power amp because of the amount of current you burn just idling. Small signal stages (preamps and what have you) sure, but as power amps? Nah.
I'm... pretty sure that's not the case. Speakers I'm currently listening on use a TDA7293 for the tweeter, and it sounds just dandy.
It makes sense as a tweeter amp because class d amps was found to interfere with some dacs and produce unwanted high frequency distortion.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
8,024
Likes
16,969
Location
Monument, CO
It makes sense as a tweeter amp because class d amps was found to interfere with some dacs and produce unwanted high frequency distortion.

How does the amp interfere with the DAC? I can see where a DAC with excessive image output (e.g. a non-filtered output from the DAC) might cause problems for any wideband amp, but do not understand how the amp can interfere with the DAC... If anything I would expect it to be the other way around.
 

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,078
Likes
377
How does the amp interfere with the DAC? I can see where a DAC with excessive image output (e.g. a non-filtered output from the DAC) might cause problems for any wideband amp, but do not understand how the amp can interfere with the DAC... If anything I would expect it to be the other way around.
Amir has had such problems with a Denon AVR if i recall correctly due to insufficient filter from the dac. After a bit of searching i found a good solution for driving tweeters and control its level in the form of the NAD 326BEE.The previous model measured quite good here and it has some features like dual sub out power amp direct etc that i find interesting.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
8,024
Likes
16,969
Location
Monument, CO
Amir has had such problems with a Denon AVR if i recall correctly due to insufficient filter from the dac. After a bit of searching i found a good solution for driving tweeters and control its level in the form of the NAD 326BEE.The previous model measured quite good here and it has some features like dual sub out power amp direct etc that i find interesting.

OK. I would call that the DAC interfering with the amp, not the amp interfering with the DAC.
 

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,078
Likes
377
I use analog active crossovers which have adjustable gain, you can do the same with digital active crossovers.
The crossover allows for direct driving the tweeters without bypassing it so an active crossover is unnecessary and besides most of them measure poorly. Going full digital needs serious investment...
 
Top Bottom