OT: Class D Amplifier Design

I think it'll need a few tweaks, switching loses will be high the way DSM tend to switch as much as possible

-Lasse

It jolly well is at my age, mate! ;-)

Sure, if you equate "better" with "faster". But I'd have expected a complimentary pair in that configuration.

It is a classic engineering compromise: Switching losses or output filter complexity. Many simple applications skip the filter.

There lives a question begging to be asked, since we are in SED after all.

What would be the best way to implement current limiting in a class D amp ? You have no emitter resistors and foldback in the power supply could not react fast enough so what do you do ?

An output device such as a loudspeaker acts as a low pass filter to an extent by itself anyway.

You can implement current limiting in a 400V-bus IGBT motor controller such that it can withstand a dead short. If they're not doing it in an audio amplifier it's because they don't want to.

--sp

Yes.

As long as there is no wire between the amplifier and the speaker and you don't care about little things like the FCC.

Why would fold-back not react fast enough? Some power supplies use cycle-by-cycle current control for their own use. It shouldn't be a big deal to do the same on the class-D output, or just use the power supply's current limiting.

With class-D one wouldn't worry too much about matching transistors anyway? They're driven into saturation so there isn't that much difference, given the appropriate overdrive.

Sure. There is no reason they couldn't use a sense resistor or just the Rds(on) of the FET. It'll take a little work to get if fast (cycle by cycle) but it's quite do-able, particularly if integrated. A class-D amplifier isn't much different than a couple of synchronous bucks (no storage inductor to complicate things, either).

el. These are easily replaced but Behringer will >not sell them to third pa rties. "

Then Behringer needs to be out of business.

Worldwide. They, like most electronics manufacturers want to see their prod uct in a landfill the day after the warranty expires. All those assholes wi ll sell are like knobs or power cords. They are a detriment to both economy and ecology. I will never ever buy a Behringer product even at gunpoint if it was free.

I got an old Pioneer SX-850 out in the garage. Broken glass in front, no to p, no bottom and the power regulator board has been Nrigged because it was broken in three places. And there is low bias in one channel because it sho rted and burned up the bias pot so I jumped it out. Know what ? It is not f or sale. And I would not trade it for a modern unit no matter what it is su pposedly worth.

A refreshingly off-topic post.

The problem I have with audiophoolery or audiophilary is that most members of this interest group can't agree on any measurement of quality.

In reality, most aspects of audio can be measured, from missing codes in AD Conversion, to harmonic distortion etc. Yet despite of this Class A will always rule over Class D or any other class even when measurements show it's inferior.

ll.

mp ? You have no emitter resistors and foldback in the power supply could n ot react fast enough so what do you do ?

or even two regular bucks,

Access denied.

Nonsense. The units of quality are the buck (or kilobuck, depending on how serious the audiophool).

That's simply not true, in many ways and in many markets. Class-D isn't the end-all.

Yeah, and what they don't find they *plant* - if your face doesn't fit or you display a "bad attitude."

ll.

mp ? You have no emitter resistors and foldback in the power supply could n ot react fast enough so what do you do ?

My thesis project, final year, was a Class D amplifier (back in 1999)

At that time, the Icepower was discussed a lot, and the discussion was abou t modelling the power stage, which in that case meant that the RDSon was a model input parameter, or doing it by some sort of feedback regulation, lik e Icepower does

I talked to the designer of the Tact Millenium:

Their first design was a fully digital one, SMPS followed by Class D. THen they tried to sell it, and the audiophiles lifted it, and it weighed to lit tle (you can tell class A by weight). So they had to replace the SMPS with a big transformer, even though it was stupid to do so.

Cheers

Klaus