The way I understood amplifer classes from way back, at least the text books I used in school and am familiar with states it like this...
class A = outputs conduct 360 degrees.
class B = outputs are bias off so they conduct less than 180 and there is a dis-continuity at the zero current crossing.
class AB = outputs are bias above cutoff at an optimal level to acheive continuous output stage transconductance for both large and small signals.
class C = output is less than 180 degrees and is typically used in radio not for audio
class D = outputs are digital PWM switching
class G = outputs are cascode with a higher tier that tracks the audio signal.
class H = outputs are cascode with a higher tier that switches to one or more higher tier rails.
I think it depends on where you are in the world as to what class refers to which operation. Unfourtunately this means that a discription must be given along with or instead of just naming a class of amplifier.
Also there is a lot of marketing notations out there that are invented to sell products and it creates lots of confusion. It is like temperature coefficient notation. Positive temp co is when the device conducts more when heated, negative temp co is when the device conducts less when heated. I see these noted as the other way around sometimes as well.
class A = outputs conduct 360 degrees.
class B = outputs are bias off so they conduct less than 180 and there is a dis-continuity at the zero current crossing.
class AB = outputs are bias above cutoff at an optimal level to acheive continuous output stage transconductance for both large and small signals.
class C = output is less than 180 degrees and is typically used in radio not for audio
class D = outputs are digital PWM switching
class G = outputs are cascode with a higher tier that tracks the audio signal.
class H = outputs are cascode with a higher tier that switches to one or more higher tier rails.
I think it depends on where you are in the world as to what class refers to which operation. Unfourtunately this means that a discription must be given along with or instead of just naming a class of amplifier.
Also there is a lot of marketing notations out there that are invented to sell products and it creates lots of confusion. It is like temperature coefficient notation. Positive temp co is when the device conducts more when heated, negative temp co is when the device conducts less when heated. I see these noted as the other way around sometimes as well.
Last edited:
@CBS240 You forgot Class I = Balanced Current Amplifier. I am just doing a simulation on this and I LOVE this topology. This is clearly THE future topology for Multi-Kilowatt, since high voltage switches and diodes can be used w/o the usual problem of body diode / reverse recovery / dead time getting worse with higher voltage devices.... I am thinking of a 2 x 200V Bus powered BCA design using CoolMos and SiC diodes....
I'm no expert.. I got the audio bug a little over a year ago now. I have read a LOT though and have certainly came to the conclusion that class d is something to take seriously these days. Certainly its the way forward that i am interested in, compact and efficient are things that attract me. I'm not to sure about the fact that class d usually seems to involve a lot of surface mount stuff and what this means in regards leaky caps 10 years down the road etc.....
As the the sub topic regarding class definitions i don't have any comments.
As the the sub topic regarding class definitions i don't have any comments.
I started reading this thread in hope it would shed some light on the issue of how the different designs tend to sound, but I see there is no consensus at all. Another I like this vs I like the other argument...
For what it might be worth, I live with a class D power amplifier (Tact SDA2175) for more than 5 years now and I have to say it' s been a revelation. Before that I used to have a variety of tube and AB amps. Soon I will be sporting a DIY F5 as well (for the first time a class A).
The "digital" one simply sounded the best of anything I've had before. I will not be changing this amplifier for the years to come. Anyone who has ever listened to music through my system has been happy. Noone ever complained about a headache or went away "shaved" by shrill highs...
Every time I took him to another place the comments were favorable too. Noone ever sais its the best amp they've heard but everyone likes it. They sit and listen to it for hours. They keep changing their cds to see how "that" would sound. I always come home with a smile...
For what it might be worth, I live with a class D power amplifier (Tact SDA2175) for more than 5 years now and I have to say it' s been a revelation. Before that I used to have a variety of tube and AB amps. Soon I will be sporting a DIY F5 as well (for the first time a class A).
The "digital" one simply sounded the best of anything I've had before. I will not be changing this amplifier for the years to come. Anyone who has ever listened to music through my system has been happy. Noone ever complained about a headache or went away "shaved" by shrill highs...
Every time I took him to another place the comments were favorable too. Noone ever sais its the best amp they've heard but everyone likes it. They sit and listen to it for hours. They keep changing their cds to see how "that" would sound. I always come home with a smile...
CENTRAL:
You've made a good choice in the F5... it's an extremely good amplifier.
Once you get it, take some time to give it a good comparison to your class D amp. Put your ear near the tweeter while no source is playing, and compare the noise floors. Once you're comfortable with the character of the noise, go back to your listening position and see if you can tell the difference between the two. It should be obvious from up close, and I'd wager the F5 is a good 20-30dB lower noise. With source material, pay particular attention the top octave, and listen for any differences in level and distortion. The F5 will be perfectly flat in this region, amplifying music precisely the way it was supposed to. The class D will not be flat in this area, and could roll off or peak depending on the load. It could be as little as 1dB, or as much as 5-6dB. 1dB might be very hard to hear, 5-6dB should be easy to hear.
If your speakers are efficient enough to play loud with the 25 watts from the F5, try to listen for differences in power bandwidth. Listen at higher levels (just below clipping on the F5 and pay very close attention to impulse and decay of high frequency transients. Depending on how much power output the class D amp has, there might not be much difference here, but it's worth a try. The power bandwidth on the F5 will be significantly larger than the class D.
Report back here when you get a chance to compare!
Cheers,
Owen
You've made a good choice in the F5... it's an extremely good amplifier.
Once you get it, take some time to give it a good comparison to your class D amp. Put your ear near the tweeter while no source is playing, and compare the noise floors. Once you're comfortable with the character of the noise, go back to your listening position and see if you can tell the difference between the two. It should be obvious from up close, and I'd wager the F5 is a good 20-30dB lower noise. With source material, pay particular attention the top octave, and listen for any differences in level and distortion. The F5 will be perfectly flat in this region, amplifying music precisely the way it was supposed to. The class D will not be flat in this area, and could roll off or peak depending on the load. It could be as little as 1dB, or as much as 5-6dB. 1dB might be very hard to hear, 5-6dB should be easy to hear.
If your speakers are efficient enough to play loud with the 25 watts from the F5, try to listen for differences in power bandwidth. Listen at higher levels (just below clipping on the F5 and pay very close attention to impulse and decay of high frequency transients. Depending on how much power output the class D amp has, there might not be much difference here, but it's worth a try. The power bandwidth on the F5 will be significantly larger than the class D.
Report back here when you get a chance to compare!
Cheers,
Owen
Thanx Owen,
I will certainly compare and your specific directions are very welcome.
If things go as planned, I will have the F5 ready in two weeks time. Most of it depends on the Cnc process since all the rest is ready and waiting to be put in place.
I have always had the curiosity to host a class A amplifier. And I believe he will mate well with my Spendors S8e...
I will certainly compare and your specific directions are very welcome.
If things go as planned, I will have the F5 ready in two weeks time. Most of it depends on the Cnc process since all the rest is ready and waiting to be put in place.
I have always had the curiosity to host a class A amplifier. And I believe he will mate well with my Spendors S8e...
Deaf and particular as I am getting older
I am rapidly getting there I think. I need stereo's that reproduce the high freq well, very well, and I can hear the mp3's sound fragment ... so I venture to guess that deafness and need for better stereo equipment go hand in hand
BTW I know one guy who was in his late 80's who sold his sx950 cos "he couldn't hear it no more". I look around and spot a marantz 2325. I ask him about that, and he goes you can get that when you pry it from my cold dead hands. I told him I'll call him next week then. That was over 2 years ago and 3 month ago, well, I still dont have my 2325 lets just say.
Cool.
Srinath.
I am rapidly getting there I think. I need stereo's that reproduce the high freq well, very well, and I can hear the mp3's sound fragment ... so I venture to guess that deafness and need for better stereo equipment go hand in hand
BTW I know one guy who was in his late 80's who sold his sx950 cos "he couldn't hear it no more". I look around and spot a marantz 2325. I ask him about that, and he goes you can get that when you pry it from my cold dead hands. I told him I'll call him next week then. That was over 2 years ago and 3 month ago, well, I still dont have my 2325 lets just say.
Cool.
Srinath.
If they use an output filter, then they do, and almost everything out there uses a filter. Most publish graphs with 8 and 4 ohms (and sometimes 2), and generally, those are acceptable, but the problem starts to show with no-load and higher 16-32 ohm loads. This is also not unrealistic since most drivers with voice coils have steadily increasing impedance at higher frequencies. People using full range drivers could be looking at even higher impedance out at 20kHz.
In addition to that, you have the problem of HF rolloff. Even the ncore amps are down nearly 0.5dB at 20kHz and they're down nearly 3dB at 50kHz. The F5 can make it's way to 800kHz before it rolls off. Sure, nobody can hear out to 800kHz, but it does hint at the problem of power bandwidth, which can be easily audible in the right circumstances.
I would imagine that the ncore amps would sound very good. They are the only class D I know of that technically rival the best of class A and class AB. I would also agree that if you really need all that extra power, then they might be a better choice than the F5.
If you only need 25-50 watts though, I know where I'd spend my money. That, however, is what you'd call a difference of opinion, and everyone's free to have their own. The important thing is to measure and listen, and decide for yourself.
Cheers,
Owen
Hi Pieter
Saying that is like saying "What's the point of HiFi when 99% of today's digital content is sold as MP3"
More than ever before, we're seeing a strong push to content at higher and higher sample rates. If you visit HDTracks or any of the other good high-resolution download sites, you'll see content at 192k and even some at 384k if you have the USB-I2S and the DAC to support that.
That would give you an effective audio bandwidth from DC up to nearly 200kHz which is pretty incredible.
You can also buy good ribbon tweeters with output up to an beyond 80kHz. Don't you think it's a bit absurd to have your power amplifier as the limiting factor in your frequency response? I sure do...
Where are you claiming the 20kHz filtering is happening? In the recording phase, or on the output of the DAC?
Regards,
Owen
Saying that is like saying "What's the point of HiFi when 99% of today's digital content is sold as MP3"
More than ever before, we're seeing a strong push to content at higher and higher sample rates. If you visit HDTracks or any of the other good high-resolution download sites, you'll see content at 192k and even some at 384k if you have the USB-I2S and the DAC to support that.
That would give you an effective audio bandwidth from DC up to nearly 200kHz which is pretty incredible.
You can also buy good ribbon tweeters with output up to an beyond 80kHz. Don't you think it's a bit absurd to have your power amplifier as the limiting factor in your frequency response? I sure do...
Where are you claiming the 20kHz filtering is happening? In the recording phase, or on the output of the DAC?
Regards,
Owen
Yes,....."a strong push to content at higher and higher sample rates", but the push is not that strong actually. We still live in a CD redbook world, 20 Hz - 20 kHz.
Besides the claims for the "better" formats like SACD are seldom met in practice with respect to sound quality; IMO most of the quality is determined in the recording process.
I am aware of the "DC to daylight" concept but never heard a system with a DC-daylight amplifier sounding better than a system with a high quality transformer coupled tube amplifier, or bandwidth limited SS amplifier.
Maybe it's more absurd to strive for 80 kHz in a loudspeaker system (unless you want to frighten the cat....). There are much more important parameters in loudspeaker technology responsible for sound quality than flat reproduction up to two octaves above audio.
Besides the claims for the "better" formats like SACD are seldom met in practice with respect to sound quality; IMO most of the quality is determined in the recording process.
I am aware of the "DC to daylight" concept but never heard a system with a DC-daylight amplifier sounding better than a system with a high quality transformer coupled tube amplifier, or bandwidth limited SS amplifier.
Maybe it's more absurd to strive for 80 kHz in a loudspeaker system (unless you want to frighten the cat....). There are much more important parameters in loudspeaker technology responsible for sound quality than flat reproduction up to two octaves above audio.
It's as pieter says the bandwidth requirements for a system are only necessary up to the point of the most limiting part. Where modern hifi is concerned this is 99.9% of the time going to be our ears (ignoring any claims of our ability to perceive anything meaningful at ultrasonic frequencies).
In the case of lots of analogue audio equipment you end up creating a wide bandwidth device simply by maximising its potential but this doesn't mean that it is in any way necessary. In fact bandwidth limiting is often employed to help maintain stability and as a means of reducing issues with EMI.
It is a fact that our ears don't do much if anything above 20kHz and in most of us the figure is significantly lower. It is also true that the majority of tweeters don't do much above 30kHz and most recording mics don't go up particularly far either.
Now having a system that can reproduce anything up to and beyond about 20kHz might seem like a fun thing to do, but it doesn't actually avail you of anything, you can't hear it, the mics can't often pic it up and most standard tweeters won't reproduce it either.
Now of course if wide bandwidth should happen be a side effect of good engineering then you might as well keep it if it doesn't do anything detrimental, because limiting it will inevitably cost you more. Having a diamond dome tweeter which is pistonic far above 20kHz, with its primary bell mode resonance @ 80kHz, isn't useful because of the extra bandwidth it gives you, but it's useful for pushing the resonance so far up that it renders any artefacts related to the resonance a non issue.
Now I like wide bandwidth as much as the next person, providing it is there as the product of good design. But automatically damning something simply because it doesn't go up much beyond where we can hear up to is just silly.
In the case of lots of analogue audio equipment you end up creating a wide bandwidth device simply by maximising its potential but this doesn't mean that it is in any way necessary. In fact bandwidth limiting is often employed to help maintain stability and as a means of reducing issues with EMI.
It is a fact that our ears don't do much if anything above 20kHz and in most of us the figure is significantly lower. It is also true that the majority of tweeters don't do much above 30kHz and most recording mics don't go up particularly far either.
Now having a system that can reproduce anything up to and beyond about 20kHz might seem like a fun thing to do, but it doesn't actually avail you of anything, you can't hear it, the mics can't often pic it up and most standard tweeters won't reproduce it either.
Now of course if wide bandwidth should happen be a side effect of good engineering then you might as well keep it if it doesn't do anything detrimental, because limiting it will inevitably cost you more. Having a diamond dome tweeter which is pistonic far above 20kHz, with its primary bell mode resonance @ 80kHz, isn't useful because of the extra bandwidth it gives you, but it's useful for pushing the resonance so far up that it renders any artefacts related to the resonance a non issue.
Now I like wide bandwidth as much as the next person, providing it is there as the product of good design. But automatically damning something simply because it doesn't go up much beyond where we can hear up to is just silly.
Class I can be temptating, but is basically just multiphase - which of course deserves a new letter....
Take care, aside multiple obviously temptating properties it has less obvious down sides:
- The two paralleled stages must have excellent DC-properties, otherwise you will get unpleasant DC balancing currents.
- The switching residuals that are being injected back into the modulator through the feedback have double frequency of the modulator carrier and the resulting distortion mechanism appeared difficult to cure, while I found a simple but extremely effective method for traditional designs.
Both topics are not necessarily show stoppers. It is more a question of your target applications.
For very high power the extra efforts for the DC-balancing won't bother in costs and complexity. Also nobody will panic, because of slightly higher THD - and of course there might already exist appropriate cures of these specific double-fs-feedback/modulator distortions.
Those are all valid points, and I'll leave it up to the individual to decide how important content above 18kHz is... I personally think there might be something to it, but I would agree that it's probably not much.
Perhaps I'm approaching this the wrong way. I keep talking about areas where class D amplifiers are technically deficient, but perhaps we should focus on what possible advantages there can be to class D amplifiers purely in terms of sonic performance. After all, for an audiophile, this should be numero uno.
I'm not talking about weight, size, efficiency, or operating temperature, but rather pure sonic performance.
In what ways does a class D amplifier offer better sonic performance than class A and class AB?
Please stick to technical justifications, and not "I own X brand of class D amplifier and it sounds the best"
Regards,
Owen
Perhaps I'm approaching this the wrong way. I keep talking about areas where class D amplifiers are technically deficient, but perhaps we should focus on what possible advantages there can be to class D amplifiers purely in terms of sonic performance. After all, for an audiophile, this should be numero uno.
I'm not talking about weight, size, efficiency, or operating temperature, but rather pure sonic performance.
In what ways does a class D amplifier offer better sonic performance than class A and class AB?
Please stick to technical justifications, and not "I own X brand of class D amplifier and it sounds the best"
Regards,
Owen
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.