bachiano said:
Class T is just another Class D. Guess the T was chosen to distinguish themselves from the others.
Best regards
Gertjan
Re: Re: What is the difference between d-class and t-class?
Not at all!
Yes, class-T is a switching class amp, but is is substantualy different from class-D.
Class-D generally uses a PWM triangle wave to modulate the switched output signal. Switching frequencies are usually not far above the audio band, e.g. 100kHz.
Class-T is much more sophisticated. It uses a square wave that changes duty cycle to modulate the signal. Resting duty cycle is 50%. The switching frequencies in class-T are also much higher than in typical class-D. The lowest frequency used in class-T is 600kHz, and is typicaly 700~1000kHz . The switching frequency changes according to the audio power and a few other factors.
Higher switching frequencies make filtering the class-T output easier than class-D. The spread spectrum of the class-T helps efficiency as well. There are other things going on, like precise control of the switchng transistors, but those could apply to class-D as well.
Class-T is unique to Tripath, you won't find it elsewhere. I don't know what they are doing inside the chip to process the audio, but they've done a fantastic job. The results are plain to hear.
Neither of these are really a digital amp. Maybe they should both be sub-classes of a class-S.
ghemink said:
Not at all!
Yes, class-T is a switching class amp, but is is substantualy different from class-D.Class-D generally uses a PWM triangle wave to modulate the switched output signal. Switching frequencies are usually not far above the audio band, e.g. 100kHz.
Class-T is much more sophisticated. It uses a square wave that changes duty cycle to modulate the signal. Resting duty cycle is 50%. The switching frequencies in class-T are also much higher than in typical class-D. The lowest frequency used in class-T is 600kHz, and is typicaly 700~1000kHz . The switching frequency changes according to the audio power and a few other factors.
Higher switching frequencies make filtering the class-T output easier than class-D. The spread spectrum of the class-T helps efficiency as well. There are other things going on, like precise control of the switchng transistors, but those could apply to class-D as well.
Class-T is unique to Tripath, you won't find it elsewhere. I don't know what they are doing inside the chip to process the audio, but they've done a fantastic job. The results are plain to hear.
Neither of these are really a digital amp. Maybe they should both be sub-classes of a class-S.
@panomaniac: You've got this definition from Tripaths Application Note AN-2, right? No offense meant, but there is a lot of marketing-bs in this note..
Ghemink is right with his definition: Class-D is used for all switching audio amplifiers. Class-T is just one possible implementation. Others include the fixed-frequency amps using a triangle wave to generate the PWM signal you already mentioned or self-oscillating amps like the UCD or ZAPpulse.
Best regards,
Florian
Ghemink is right with his definition: Class-D is used for all switching audio amplifiers. Class-T is just one possible implementation. Others include the fixed-frequency amps using a triangle wave to generate the PWM signal you already mentioned or self-oscillating amps like the UCD or ZAPpulse.
Best regards,
Florian
Tripath uses either some form of hysteresis modulator or a "classic" DS modulator. Both of them cause a dropping switching frequency with increasing amplitude.
This has advantages and disadvantages, although Tripath wants us to believe that it only has advantages.
But the marketing guys went one step further and boldly call it spread-spectrum.
But the Tripath amps are definitely class-d although not of the PWM type.
Regards
Charles
This has advantages and disadvantages, although Tripath wants us to believe that it only has advantages.
But the marketing guys went one step further and boldly call it spread-spectrum.
But the Tripath amps are definitely class-d although not of the PWM type.
Regards
Charles
flo said:
I got it from the app notes, yes, and also from looking at the signals and trying to build my own class-D amps.
There certainly is marketing hype in the Tripath claims, of course. Fundamentaly the classes are similar, but in the details they are different enough to warrent the new name, IMO.
As I said, perhaps these should be sub-classes of a larger Class-S (switching).
But we give different names to class A & B, and even a so-called class AB. Those are no more different than Class-D and Class-T.
What's in a name? 🙄
Hm...
Come on guys, spell it right and there wont' be any confusion:
Class T "TM" for trademark!~
It's a marketing gimmick, nothing more.
How about class N?
Same thing.
I wish people on the forum would stop using the term.
The only switching frequency imposed on class d is caused by the components themselves, if you could make it happen at 10Ghz it would still be class d.
Got to keep a keen eye out for the snake oil anyway.
Cheers,
Chris
Come on guys, spell it right and there wont' be any confusion:
Class T "TM" for trademark!~
It's a marketing gimmick, nothing more.
How about class N?
Same thing.
I wish people on the forum would stop using the term.
The only switching frequency imposed on class d is caused by the components themselves, if you could make it happen at 10Ghz it would still be class d.
Got to keep a keen eye out for the snake oil anyway.
Cheers,
Chris
I won't argue that the appellation "Class-T" is a bit of marketing. How do you get to declare your own class?
But isn't what Tripath are calling class-t different enough from other switching amps to merit a new class?
As I said above - are class A, B and AB different enough to earn a unique name? I think so. Or should they all be lumped under "Class-L" for linear? Would that be useful?
Sure, it's a bit cheeky for Tripath to "invent" their own class, but the technique is different enough to deserve it.
I don't see the snake oil in it, just a bit of the usual audio hyperbole.
But isn't what Tripath are calling class-t different enough from other switching amps to merit a new class?
As I said above - are class A, B and AB different enough to earn a unique name? I think so. Or should they all be lumped under "Class-L" for linear? Would that be useful?
Sure, it's a bit cheeky for Tripath to "invent" their own class, but the technique is different enough to deserve it.
I don't see the snake oil in it, just a bit of the usual audio hyperbole.
Class-d amps switch between two rails (sometimes more) and the info is in the pulse density.
This is done within ordinary PWM and Tripath amps as well so no reason to call it an own class.
Regards
Charles
This is done within ordinary PWM and Tripath amps as well so no reason to call it an own class.
Regards
Charles
Re: Re: Re: What is the difference between d-class and t-class?
a much simpler way is for you to go through Tripath's claims one by one and compare it with other class D amps then you will be clear as to what is marketing BS.
Take the IRAudAmp1 for example: it modulates on square waves, at about 500khz, with a "resting" duty cycle of 50%.
it depends on how you define "digital".
a much simpler way is for you to go through Tripath's claims one by one and compare it with other class D amps then you will be clear as to what is marketing BS.
Take the IRAudAmp1 for example: it modulates on square waves, at about 500khz, with a "resting" duty cycle of 50%.
panomaniac said:
it depends on how you define "digital".
There were never any digital amps, there are no digital amps and there wil never be any digital amps out there.
Simple as that !
Regards
Charles
Simple as that !
Regards
Charles
phase_accurate said:
it depended on how you define digital amps; it depends on how you define gitial amps; and it will further depend on how you define digital amps.
simple as that!
Hi,
like all other amplifier classes, class D is defined by the state of the output devices. In class D, output devices are either totally on or totally off and act as switches. That is fundamental to class D. Type of modulation is totaly irelevant.
And before you start another fruitless debate about class D and digital, please reread Difference betweeen Class D and "Digital Amplifiers" .
If you can come up with some arguments that were not covered in that thread you're welcome to post more.
Best regards,
Jaka Racman
like all other amplifier classes, class D is defined by the state of the output devices. In class D, output devices are either totally on or totally off and act as switches. That is fundamental to class D. Type of modulation is totaly irelevant.
And before you start another fruitless debate about class D and digital, please reread Difference betweeen Class D and "Digital Amplifiers" .
If you can come up with some arguments that were not covered in that thread you're welcome to post more.
Best regards,
Jaka Racman
At first, one should consider what digital actually means: Processed/presented in numerical form.
You can store data, make calculations and logic decisions in the digital domain.
But an amplifier doesn't output numbers but voltage and current !
Regards
Charles
You can store data, make calculations and logic decisions in the digital domain.
But an amplifier doesn't output numbers but voltage and current !
Regards
Charles
So can we say that "Class-D" is a misnomer and that all these amps should be called "Class-S" for switching?
Would the different flavors of this class get thier own letters? D-T-J-whatever?
Would the different flavors of this class get thier own letters? D-T-J-whatever?
Hi.
I am under impression that you think that class D stands for digital. No more than class A stands for analog. It was simply the first letter that came along after class C (which is rarely mentioned because it is used in RF transmitters IIRC). Besides that, there is class E which is similar to class D, only switches operate in resonant mode.
After E, you have to invent new operational state of output devices, for the letter meaning anything else than the marketing BS.
Best regards,
Jaka Racman
I am under impression that you think that class D stands for digital. No more than class A stands for analog. It was simply the first letter that came along after class C (which is rarely mentioned because it is used in RF transmitters IIRC). Besides that, there is class E which is similar to class D, only switches operate in resonant mode.
After E, you have to invent new operational state of output devices, for the letter meaning anything else than the marketing BS.
Best regards,
Jaka Racman
I do not see why we should use S instead of D.
I think you are misunderstanding the meaning of the "D" (like marketing- and sales-people usually do). It has nothing to do with DIGITAL like class A has nothing to do with analog.
They are just operating classes that have been defined a long time ago. A to C are non switching classes defined by the angle of the current-flow through their output devices:
A:360 deg
B:180 deg
AB: >180 deg
C: < 180 deg
And D was just used for the next in line, i.e. the switching amp using pulse-density or pulse-width to represent the signal. There is also sonme switching amp class that is designated E.
The next ones interesting for audio are G and H and are used to name amps of classes A, AB or B with modulated or multiple PSU rails.
Regards
Charles
Edit: I now see that Jaka was quicker than me
I think you are misunderstanding the meaning of the "D" (like marketing- and sales-people usually do). It has nothing to do with DIGITAL like class A has nothing to do with analog.
They are just operating classes that have been defined a long time ago. A to C are non switching classes defined by the angle of the current-flow through their output devices:
A:360 deg
B:180 deg
AB: >180 deg
C: < 180 deg
And D was just used for the next in line, i.e. the switching amp using pulse-density or pulse-width to represent the signal. There is also sonme switching amp class that is designated E.
The next ones interesting for audio are G and H and are used to name amps of classes A, AB or B with modulated or multiple PSU rails.
Regards
Charles
Edit: I now see that Jaka was quicker than me
I know what a digital amp is! It is an amp to which when you input a "1" it ouputs a "1"!
Hehe, sorry couldn't resist!
Sébastien
Hehe, sorry couldn't resist!
Sébastien
- Status
- Not open for further replies.