Class D "digital" amps. Opinions on performance

[313]

Thanks AgedHorse & Jerrold for addressing some of my questions about the PWM and switching losses in these audio applications. You folks have many different issues and challenges than I deal with in motor controls, but we share many of the same concerns. I can't imagine how you guys deal with PWM dead-time issues and switching losses at PWM frequencies >> 200 Khz with today's devices.

Are audio power electonics entities looking at the Silicon Carbide switches as possible replacement for MOSFETs yet? Much interest there in motor controls due to very low switching (and conduction) losses. These devices are hard to get right now (and expensive), which might preclude their use in consumer electronics for the immediate future.

 

[agedhorse]

Thanks AgedHorse & Jerrold for addressing some of my questions about the PWM and switching losses in these audio applications. You folks have many different issues and challenges than I deal with in motor controls, but we share many of the same concerns. I can't imagine how you guys deal with PWM dead-time issues and switching losses at PWM frequencies >> 200 Khz with today's devices.

How do we deal with them? Very, very carefully

Are audio power electonics entities looking at the Silicon Carbide switches as possible replacement for MOSFETs yet? Much interest there in motor controls due to very low switching (and conduction) losses. These devices are hard to get right now (and expensive), which might preclude their use in consumer electronics for the immediate future.

We (as an industry) are already using devices based on silicon carbide technology in LEDs, specifically blue and white. CREE is a leader in this material which offers LED efficiencies in the >125lm/watt range. Very cool stuff. I have been watching the silicon carbide based switching device technology for exactly the reasons you mention. Cost is an issue now, but as the technology matures, it is sure to come down.

For very large devices, any reduction in losses brings great benefits. For motor controls, especially those that operate 24/7, the payback can certainly justify the increased one time device cost difference IMO.

 

[eban3]

digital power supply amps will probably not withstand more than 5 or 6 years of steady use ,theyr'e throwaways I havent found 1 that lives up to the fidelity of an analog powered amp ,they are however very convenient ,,,,,,, I would never record with one,,, fidelity's just not there.....yet

 

[Bob Lee (QSC)]

Are there any digital power supply amps?

 

[Mr. Foxen]

I use my fingers to wind up my gramophone, does that count?

 

[BurningSkies]

I use my fingers to wind up my gramophone, does that count?

Your gramophone can't come close to sounding as good as my wax cylinders.

 

[f.c.geil]

Are there any digital power supply amps?

What's a digital power supply amp?

That means, "not that I know about."

 

[Mr. Foxen]

Your gramophone can't come close to sounding as good as my wax cylinders.

I checked youtube video of wax cylinder and it totally sounded digital, so I know they suck.

 

[Brad Johnson]

Impossible... Youtube* is analog.







*aka Youvalve in the UK

 

[Neil]

For the record (and yes this has been cleared up plenty of times) There are plenty of amps that are ultralite that aren't class d such as the little mark heads which use an A/B power amp... light doesnt always equal class d

Corey specifically which Markbass amps are class A/B? Thanks.

 

[BurningSkies]

Corey specifically which Markbass amps are class A/B? Thanks.

The littlemark ii & iii are class a/b.

 

[dukeorock]

digital power supply amps will probably not withstand more than 5 or 6 years of steady use ,theyr'e throwaways I havent found 1 that lives up to the fidelity of an analog powered amp ,they are however very convenient ,,,,,,, I would never record with one,,, fidelity's just not there.....yet

That's just silly talk! Sounds way more anecdotal than factual too. I just had a band in my studio...they were going for a raw, bluesy, old school sound...I hooked up 3 different rigs. Genz Benz ShuttleMax 12.0 into a Genz Uber 212, Fender TB1200 amp into a Genz Uber 410 and an Ashdown ABM500 into a Fender810 Pro cab...all pretty serious, 'for real' stuff...the bass player and the entire band picked the ShuttleMax in the blind test...oh yeah, did some cab swapping as well...didn't matter. They picked the ShuttleMax every time...that ain't exactly scientific, but that's what happened that day

 

[bryanspkr]

A device delivering power to a loudspeaker is by definition
analogue. The non-idealities of class D amplifiers
are analogue in nature as well, and must be
solved using analogue means. Error control (feedback)
is an important tool in the reduction of nonlinearities,
and is often the only available tool to address
linear errors such as response deviations. Since
the notion of “digital” is not applicable to amplifiers,
attempting to make amplifiers more digital by
abolishing error control is a futile and irrational endeavour.
All amplifiers are analogue, and the appreciation
of this fact can only lead to more and better
analogue amplifiers.
Invalid Link Removed

 

[grey area]

thanks fellas. you guys realy know your onions(i dont)from "what colour bass shall i buy"to the kind of complex stuff you guys are involved with it seems everything is covered.i raise a glass to you all . thanks.

ps quantum class amps anyone?

 

[Chef]

All I can say is this:
I've used the MBFusion500 for some years now, and just got the MBFusion800;
they both kick butt.

 

[grey area]

i second that chef

 

[Munjibunga]

If it sounds good, it is good. Each of my eight Class D amps sounds good. I'd put my Blue Soul, Shuttle 9.0, MB800 or MB Fusion 800 up against any non-Class D amps in sound and volume comparison tests. And I'd use all neo cabs to do it.

 

[seamonkey]

And "If it doesnt sound good, it isn't good"
One person's trash is another person's treasure.

But not class-d, It doesn't sound like anything - there's nothing to sound like.
And not because it's class-d or solid-state, and any of the hear-say
Transparency is the key.

Class-ab amp sections and class-d went beyond human hearing limits decades ago. They sound the same - like nothing.

The tone shaping is from the pre.

Extreme cases are very rare - and things have evened out.

 

[Vince Klortho]

Class-ab amp sections and class-d went beyond human hearing limits decades ago. They sound the same - like nothing.

The tone shaping is from the pre.

Extreme cases are very rare - and things have evened out.

There are a lot of guys in the world of large PA systems who would disagree entirely with these statements. They will claim that amplifiers do not all sound the same and there are many blind tests to prove it. It's not so much about the class of the amps either. [Invalid or Expired Link Removed] example of a guy's review on amplifiers that describes how he thinks most of them have different sound qualities.

 

[Vince Klortho]

IMPO, Class D and Class I are based on EXACTLY the same principles, class I being Crown's propriatary tradmark naming. It is class D but carries with it Crown's own name which is not an amplifier class as defined by the governing body of IEEE. I mentioned it as a "class" because that is how it is being discussed and how it's been misunderstood. Class T and Class TD are also not real classes but are really class D as well. I listed them based on this discussion so as to explain how they work and what specific class D mechanism each one uses.

I think you have completely misunderstood the difference between defined (real) amplifier classes and trademarked names like Class I, Class T, Class TD etc. which bear no direct relation to class other than what their technology and topology dictates. Incidently, all of these trademarked classes are in fact class D amplifiers.

Class TD seems to be a hybrid topology. Here is how one person describes it in [Invalid or Expired Link Removed] :

The actual output stages are, in fact, class AB. The class D section only supplies power to the output devices, a "class D tracking power supply" if you will. IOW, they're using a relatively inexpensive class D circuit design to track the input audio signal and feed only the amount of power to the output devices required to drive the load. The power supply is a SMPS design which powers the class D circuit, which then drives the output circuity. Pretty ingenious really. You can make a very high output amplifier that's very lightweight and not have to give up the sonic benefits of a class AB amp. And the complexity of the class D circuit filter doesn't need to be that great since it's only delivering varying rail power instead of actual audio.