D class amps with Soundlab speakers

My understanding is the efficiency of the high speed switching design of
the class d amp is what enables its compact size and weight including that
of the power supply compared to similar power rated class a or class a/b.

It is really a major technical step forward these days in amp size, weight
power and performance compared to older design approaches. Very
innovative!

In case it would be of any significance in the present discussion, I have a Spectron with all the upgrades driving a pair of Acoustat Spectra 44. It does it with a lot of ease with firmer bass performance and overall definition than my Coda Model 11 100 watts full class A. And it remains quite cool to the touch. The preamp is a CJ Premier 16 LS2 and I find the mix quite listenable too.

I could add a class D amp, at least a higher quality one, is also a plus in terms of silent running, which is detrimental in retrieving all the details of a recording for which electrostats are mostly famous.

"I find that class D works a lot better if they aren't being pushed hard."

For most class D amp its true. However, Mark Levinson has huge power supplies. Jeff Rowland has very large, Daniel Weiss amps are very powerful too and my Spectron has 3500 watt peak power (and stable to 0.1 Ohm load)

It has almost nothing to do with output stage (be it class D or other calss) - only one thing is important, the temperature should not raise above safe operatating levels.

The weight of the power supplies is "important" only in linear PSU but in switching like that of Jeff Rowland (not ICE Power, sorry) the weight is low.

Among class D I mentioned above, Mark levinson cost $50k, Weiss $30k, Jeff Rowland $15k and Spectron with all upgardes about $5.5k or in monoblock mode - $11k. You pay for quality...

Class D amps get away with "lighter" power supplies because they are filtering at many KHz instead of the 60 -120 Hz of linear supplies. Thus, they can use much smaller capacitors for a given current or power rating.

Look at PC power supplies, for example, which use a similar switching technology. They currently build 1000W-1500W power supplies for computers that fit in a box 6"x6"x8". And these run at 100% continuous power, as opposed to audio systems where the average power is about 1/10 of the max at full output.

That said, I have always been suspicious of the audio performance for such amplifiers, except in subwoofer applications. Maybe I'm old fashioned, but if Class A/B is inferior to Class A, the class D has got to be inferior to Class A/B, simply because it operates completely out of the linear range of the output devices (hard on to hard off).

I'm not sure why the PC type of switching supply is not used more for audio (actually it is in car audio, but only there), but I suspect the audible side effects of such supplies are intolerable when compared to old fashioned linear supplies with thousands of microfarads of capcitance.

They are the cheapest way to make really big power, but I am not sure if they are the best.

You may find the class D will strip some harmonic texture compared to a tube, & there can be a "cheapness" in the treble which can be addressed somewhat. But if you have some tone control on you xovers like the big Soundlabs, you may be able to dial in some warmth to help this slight starkness in the highs.

I also found if you use the "MW" setting on the PS audio regenerators it helps sweeten things too. But in return you will get an enormous, solid soundstage with effortless dynamics & very low distortion & absolutely no noise.

They can be great.