advertisement


Anyone compared high end Class D power amps?

you don’t listen to the rectangular waveform as it is well past the human hearing range, you listen to a waveform reconstructed from the filtered pulse width modulated waveform which is in the audio band.
Hi,
The output at the amplifier terminals is a filtered rectangular wave.

There is no reconstruction, as the filter is a continuous function with a rectangular wave applied. The output of the filter is just predominantly the lower frequencies in the rectangular waveform. So you are listening to a rectangular waveform, it is just that the high frequency components that make up the rectangular wave are attenuated.

Here is an example from Audio Science Review :
https://www.audiosciencereview.com/...easurements-of-nord-one-nc500-amp.7704/page-3

The screen shot of the amplifier is from a Teac amplifier using the Hypex module.
index.php


I am not disputing a subjective review, but the filter output is still a noisy signal. What i think this shows us is that the ears are not as good as we believe them to be.

Regards,
Shadders.
 
Guys guys guys....no need for the macho text book waving...peace and love and all that, one is not better than the other..

You’re both equally massive dorks...:p
 
I've tried quite a few Class D amps now, Lyngdorf and Hypex sounding the best IMO.

I'm amp agnostic and i know this is anecdotal, but another member here was selling a pair of Audio Physics Calderas, that i went to have a listen to. We tried them with my Lyngdorf amp, a pair of Bel Canto Coherent modded monoblocks and a pair of Albarry 30W Monoblocks (the red perspex ones).

The Albarry's sounded much closer to the timbre of real instruments compared. Life in the old amps yet. :)
 
Did you test the Nords on the horns too HFN? I’m finding that for some reason class d works well with my horns.
A lot of it I’m putting down to the very defined soundstage they tend to have. An overblown soundstage is an inherent problem with my big LV horns. Using valves I’ve found (9/10 times) the soundstage is silly large. Singers sound like there’s a 30ft giant singing to you whilst kneeling down in your living room.

The Class D’s I’ve tried pin point everything. The singer is life size in front of you..
Ah not actually tried any nords yet , these asx50 are fit Ghent audio hobbies made by an aos member. Hope to try some soon
 
On the subject of class D amps and horns, I have a Nord One NC500DM powering a pair of JBL 4429s (which have high/mid-range horns and 12" direct radiating woofers).

Being a creature of habit, I have a second pair of 4429s in another system. Those are powered by a Devialet 220.

Even with identical speakers, comparison is a difficult thing, especially as the systems are (obviously) in different rooms with quite different acoustics. And the system with the Nord power amp isn't finished yet. It's waiting for a Lakewest DevDac, and in the meantime has a DSPeaker Antimode doing DAC and preamp duties, which probably isn't ideal.

Anyway, upshot is both amps do a great job, both systems sound very fine. Class D plus horns works for me.
 
I am not disputing a subjective review, but the filter output is still a noisy signal. What i think this shows us is that the ears are not as good as we believe them to be.

Interesting to see measurement evangelists over on ASR apologising that fuzzy mess away, the same folk who so often crusade against valve amps etc! Yes, the point the speaker isn’t fast enough to reproduce that 450kHz distortion is valid, but even so... give me a nice clean valve or class A amp!
 
It's not so much that it cant reproduce it, as so much as the crossover structure kills it stone dead.

Still, I dont like how they sound in the treble.
 
The Albarry's sounded much closer to the timbre of real instruments compared. Life in the old amps yet. :)

Not for everbody, not at a time when the original B&W Nautilus 801 was allegedly the state of the art, but the Albarrys could sound magnificent with the right speaker.

Brilliant combination with Heybrooks, even the wee HB2, the Keesonic KRF and also quite spectacular with the sadly short-lived Mordaunt Short Signifer.
 
Hi,
As per the image on page 12, the rectangular wave is the signal from the output stage of the class D amplifier, and it goes through a coiled up piece of wire before it gets to the speakers.

For a class A/B amplifier, the signal at the output stage is a near exact amplified replication (to within 0.001%, as an example) of the input signal, which goes through a coil (if there is one, 10x less inductance than class D) and a resistor in parallel, before it gets to the speaker.

We have people claiming to hear differences in cables, and resistors/capacitors too.

My point is, how can people describe class D and being clinical (accurate) when its signal is a rectangular wave passed through a coil, when in a qualitative way, it is no where near the same as a class A/B amplifier signal.

People claiming they can hear a difference between cables, which requires an extraordinary capability in a persons hearing, yet cannot hear the effect of a rectangular wave which is what they are listening to, for class D.

Of course, the math/engineering means that the "noise" is attenuated in class D, and what remains is an approximation to the input signal, but it does make me think at least, that human hearing is not as fantastic as some discussion lead people to believe it is.

[On another forum i was told i was deaf because i could not hear differences between cables, or different resistors]

Regards,
Shadders.
 
Hi,
As the oscilloscope plot shows, the signal is extremely noisy. A class A/B amplifier will be very much cleaner.

As i inferred, from a qualitative point of view, if people can claim hearing differences in cable where that cable adds absolutely no distortion to the signal, then how can those same people hear such a good sound from a class D amplifier, whose output is just an approximation (derived from a discontinuous rectangular wave).

Regards,
Shadders.
 
To answer the OPs question.

I tried these various amps after owning and trying the Devialet D220 Pro and D250 Pro; Mark Levinson 585, McIntosh MA9000, Vitus Ri100, Audionet DNA 2.0 and the BelCanto ACI-600, the ACI-600 was nearly twice the price of each the other but to me was the weakest sounding amp with a flat/dull and un-textured timbre. I kept the DNA 2.0, best dynamics, most realistic timbre and exceptional sound stage. IF I HAD to go Class D, the Devialet D250 Pro would hit the mark better than the ACI-600.
 
Last edited:
& another small aside, I'd really like to see some recent or well-known Class AB power amps (re)measured with that same AP 7-pole filter usually used by default for 'Class D'/ PWM-based amplifiers: 'But of course there's switching noise, so we use this filter for all measurements ...'

Just for an apples-with-apples numbers show, at least on the measurements front, since such a thing will have a profound effect on the %THD and SNR much above 10Khz on any amplifier..
 
& another small aside, I'd really like to see some recent or well-known Class AB power amps measured with that same AP 7-pole filter usually used by default for 'Class D'/ PWM-based amplifiers:

'But of course there's switching noise, so we use this filter for all measurements ...'

Just for an apples with apples numbers show, at least on the measurements front.
But you can't hear it, so it doesn't matter.....
 


advertisement


Back
Top