class D - anyone?

[vitalstates]

Hi all, A. Heretic here.

Has anybody got any experience of any of the class D offerings from :

Magnatec, Hypex or LCAudio.

I'm just curious as to how they might hold out against electrostats. All the blurb I've managed to find says they are good into nasty loads, and one even says they will drive a short.....mmmmm

Regards

Ed

[Toppsy]

Ed,
I suppose the main question to ask about these amps is... are they capable of delivering high current at 2-ohms load to drive your ER Audio ESLIII's? If so they should be OK. As you are aware these speakers were developed around Class 'A' SS amps.

[vitalstates]

Hi Colin

thanks for the reply...yes unfortunately I am well aware of the SS connection as I still havnt found anything that improves on the Meridians capabilities....I think the trend in Oz at the moment is for Nelson Pass designs but I still havnt got round to building the Aleph, I know I do keep banging on about it but its a big build and therefore a big risk for an experiment....and a lot of heat.

Having said that, Class D is not SS class A....and having read all the blurb it seems most of the boards are in the 250-300 watt area when run on max rail voltage, although all of the ones I have read about will run at much reduced voltages......so that does imply there is current capability......

all the reviews mention SE type detail, even at low SPL, which is exactly what I'm looking for, but I havnt read about anybody driving ESLs.

ho hum.....

Regards

Ed

[Toppsy]

Hi Ed,

If memory serves I think Rob in OZ had a dabble with some Class D amps last year. It might be worth an email to him?

At the moment I'm driving the Acorns with the Yarland 845SE amp I bought from Paul after Eggfest3 meet. This has around 16W per channel and drives the ESLs far far better than my 6550 amp with loads of emotion, detail and bass aplenty. I'm going to swap over to the ESLIII's this weekend and drive then full range without active subs with the Yarland to compare. I think this amp will address the bass lightness of the 6500/ESLIII combo.

I do think the ESLIII are more amp dependant than the Acorns to get the best out of them. The Acorns with their larger panel area are more efficient and are a lighter amplifier load over the ESLIIIs, in my experience.

[petercom]

Class D? Why ever are you interested in Class D? In fact why would anyone think that converting a perfectly good signal, whether analogue or digital, through a D to A or D to D converter, then applying huge levels of digital feedback and EQ to turn what is essentially a highly non-linear, heavily load dependant high frequency amplifier into something that resembles an audio amp. Then you pass the resultant dynamically compressed hf signal through an output filter, with inductors that run into saturation on peaks, to remove the hf carrier and call the remaining mush hi-fi? I don't think so!

And you want to do what? Drive the lowest coloration, lowest distortion, lowest impedance, highest capacitance speakers with it? And not expect it to sound like garbage? You've got to be joking.

[pre65]

Hi-have you ever tried using the Quad II amps with the ESL ?

they are about 15wpc and were designed to go with the original ESL speakers.

Just a thought.

Philip

[vitalstates]

Thanks Peter

Bring it on.......at least thats an angle I can relate to....But like the tree in the forest, if it falls and theres nobody there to hear it, did it make a noise?

I instinctively(but blindly) agree with you. I just want to hear it to satisfy myself.

Hi Phil

I've had lots of quad gear over the years and listened to lots I havnt had, and no offence but quad II is definately not to my taste. Thanks for the idea though.

Regards

Ed

[petercom]

I felt the same way. I spent some time dreaming of an all digital system. The holy grail was to keep everything in the digital domain right up to the speaker drive unit terminals. Not having to 'convert' back and forwards would be a good concept.

Then I spent months trying out all the circuits, and worked with an American firm that had progressed to the point where it could all be manipulated in software.

When it didn't perform as expected I started to look into why. That's when I found out the truth about what was going on inside these low heat, super efficient, black boxes. I'm not saying the problems can't be overcome, but not while everyone thinks that you can do anything to a signal as long as it remains in the digital domain just because it can be apparently reconstituted back to its original form. I would like to think this was true, but it doesn't seem to work out in practice. And until we know why I think it is safer to keep clear of it.

Look at feedback, for example. Because of the success of the Williamson amplifier, which took great pains to solve the instability problems due to feedback, the audio world went mad trying to apply more and more feedback, using more and more gain, to produce less and less distortion. We now know why high levels of loop feedback are a bad thing, but it has taken thirty years or more to find out.

We are in the same stage with digital. Digital is yet young. We know very little about why it sometimes sounds good and why it can easily go bad. The audio world also has a blinkered approach to all this. It's digital, so it must be pure. Digital products are praised to the skies for no apparent reason. The general public are buying digital radios that sound worse than FM, digital TV that has compression artefacts that are visually disturbing, and now digital power amps. Look at all the fuss made about those Flying Mole amps, amongst others, in the press. I've listened to them, on the end of some not so demanding speakers, and I wouldn't give them house room. In the pro field they fulfill a requirement for a reliable, low heat, efficient unit for PA and so forth but hi-fi they are not.

I have a theory that digital demands resolution and over sampling way, way above what we think is necessary for the bandwidth we want to store. It is all down to filtering and conversion problems. You can see it in photography as well as in audio as well as in video. In photography a 3 MegaPixel image should be equivalent to 35mm - it isn't, you need 11 MegaPixels to produce as good a print as you can from film. In video the current digital standards fall so far behind analogue you can see it in every piece of sports relay by the BBC or in CGI film quality on movies. Only with HDTV will we find something better than analogue TV, and then only if they decide not to compress the hell out of the signal.

Audio is much more fragile. You can't get away with the losses you see in video without it being noticeable for its effect on the music. And every conversion carries a loss, every filter introduces distortion - and it's not inoffensive, musically related, harmonic distortion like a valve amp, it is horrible, crunchy, rasping ear offending distortion that, even at a low level, you don't want to hear.

Tonally digital amps can be made to sound nice. They can be made to sound very 'clean' and precise, especially if carefully bandwidth limited. But they don't convey music like a good valve amp does, any more than an SS amp will. But I'll leave your ears to find that out, if you ever manage to hear one.

[alnewall]

In fact why would anyone think that converting a perfectly good signal, whether analogue or digital, through a D to A or D to D converter, then applying huge levels of digital feedback and EQ to turn what is essentially a highly non-linear, heavily load dependant high frequency amplifier into something that resembles an audio amp. Then you pass the resultant dynamically compressed hf signal through an output filter, with inductors that run into saturation on peaks, to remove the hf carrier and call the remaining mush hi-fi? I don't think so!

Blimey, its a miracle mine works at all.

[steve s]

Originally Posted by alnewall

Blimey, its a miracle mine works at all.

Al, Ed asked me the other night about d class, your's was all i had heard, it sound quite good to me....?

cheers steve