Do ClassDAudio Amps Have Proper Decay Of Notes?

BTW, someone said that Class D is not "digital" which is partly true, but PWM is a digital signal processing technique, so it's understandable for those amps to be called "digital" especially from a marketing angle.

Just had to put that in there.... that 'someone' being me...

Class D was first demonstrated during the vacuum tube era. The reason it is called 'class D' is that class C came before it and so was already taken. PWM (Pulse Width Modulation) is an analog technique. Just ask any keyboard player with an analog synthesizer (by varying the pulse width, a string sound can be obtained from a square wave).

From the Wikipedia page (https://en.wikipedia.org/wiki/Class-D_amplifier):

The term "class D" is sometimes misunderstood as meaning a "digital" amplifier. While some class-D amps may indeed be controlled by digital circuits or include digital signal processing devices, the power stage deals with voltage and current as a function of non-quantized time. The smallest amount of noise, timing uncertainty, voltage ripple or any other non-ideality immediately results in an irreversible change of the output signal. The same errors in a digital system will only lead to incorrect results when they become so large that a signal representing a digit is distorted beyond recognition. Up to that point, non-idealities have no impact on the transmitted signal. Generally, digital signals are quantized in both amplitude and wavelength, while analog signals are quantized in one (e.g. PWM) or (usually) neither quantity.

NAD is apparently using the hybrid in the D 3020 but the Direct Digital approach in the D 7050.

Despite the advertising on their website, their 'Direct Digital' approach has a considerable amount of analog circuitry- a lot has to do with definitions of terms used in their advertising. I can certainly see how that could invoke confusion in the marketplace!

^^ Yes, that was certainly not obvious since buying a class D audio kit is pretty easy to do. It seems that a company name like that would cause a lot of confusion in the marketplace. It probably can't be trademarked.

@auxinput   If that's really the case, the problem is that the switching frequency is too low because its only at the frequencies of the digital codices. And I can tell you from having designed Class D amps that no DAC out there can drive the outputs directly- so there is likely a level shifter and driver circuit as well. Since the codices are such low frequency, a converter of some sort would have to be employed, along with some sort of computer control (like @ostensible mentioned earlier) to switch from one algorithm to the other. Otherwise the distortion of the amp would be quite high; funny that distortion can be low with digital but if you switch at that frequency distortion is unacceptable.


That's why I don't think there is a true digital amplifier out there.

hence the amp itself gets the moniker "digital" amp and "class D" is conflated with "digital" when in fact the amplification itself an analog process controlled by digital microcontrollers.

Our class D amp does not use any digital circuits or microcontrollers. You don't need a computer for a class D circuit. 

However, would you agree that what you just excerpted about PWM describes a process whereby the waveform (square wave being the trigger for lack of proper terminology at the moment) is a combination of ON or OFF states in an array of switches (transistors or vacuum tubes) and that ON = 1, OFF = 0 which would easily lead people to assume that the process is "digital" even when it was a plain square wave and not a complex microcontroller dictating the ’sampling’ process?

Yes- I've been aware of the confusion around this topic for some time :)

Finally, would you agree or disagree then, that today's class D amps, with built in DACs and which use complex microcontrollers to sample the output and alter the switching algorithms can indeed be called "digital amps"?

'Built in DAC' suggests that the amp is amplifying an analog signal, since the acronym 'DAC' stands for Digital to Analog Converter'. So as the sentence is written, my answer is no, I don't agree. If we let go of that problem though, another arises, which is that the amplifier portion which has the digital stream as its input still has a lot of analog processes in it to do its job. One example is the dead time circuits that are present in class D output sections. They are essential to prevent shoot through currents which can damage the output devices. Digital circuits don't need anything like that; dead time is an analog issue brought on by the fact that transistors take a finite time to turn on and even longer to turn off. Keep in mind the conversion implications involved because Class D amps usually switch at much higher frequencies than a digital bit stream.

From a marketing perspective, the term 'digital amplifier' (which isn't an engineering term) is lazy writing at best and serves to ironically confuse the market, which readily associates the term with class D (but not in the way that it should; hence the irony). It might apply to any amplifier (Class D or otherwise) that had a DAC built-in.
An amp that I would consider to be a true 'digital amp' would be one where the process of conversion from digital to analog happened only at the output, IOW that the entire circuit is one big DAC.

Every component & every connection is important in audio if you want realistic playback. Trivializing the power amp-speaker connection will lead to only trouble for you. Neglect it at your peril.....

+1!

Low level detail (which includes decay) is a common problem!

My recommendation is to avoid difficult loads for the amp- with any amplifier including class D, you can see the distortion climb in the specs when you drive lower impedances and you can hear it as a loss of detail and increased harshness. Put another way, if you can reduce distortion in the amp, it will sound smoother and more detailed. Good Luck!

The ClassDAudio goes out to 35khz, do you think this is far enough to avoid the ringing issues that switching amps have?

Its generally agreed that you need more bandwidth than that. Even the AES which tends to be overly conservative recommends 2 octaves above 20KHz (80KHz).

I can't comment about ringing- that seems to have more to do with the individual design rather than an overall characteristic.

While I've seen graphs that dramatize a digital waveform I honestly can't hear anything that would suggest a stair stepped waveform or "digital noise" from any of the switching amplifiers I've listened to

Just so we are clear here, class D is not digital. Its an analog process.