High Efficiency Speaker Asylum

Posted by tomservo (M) on April 22, 2024 at 12:28:28

In Reply to: RE: Anyone ran horn speakers with Class D amps? posted by Cougar on April 20, 2024 at 07:59:27

There has been a change in the way amp's work or rather how they behave.

In the old days (before class D evolved from servomotors in robotics to audio) an amplifier rated at say 400Watts into 4 Ohms could put out a 40 Volt sine wave into 4 Ohms = 10 Amps times 40Volts = 400 Watts RMS.
At this point if you asked for say 43 Volts, the top of the sine wave would be flat topped (on a good day).

Back when Ice power amps came out, they used a somewhat different way to rate the power. Initially i didn't like it but actually this makes more sense.

The reason can be seen looking at the power rating of loudspeakers where it isn't a sine wave but pink noise or shaped pink noise that is the signal.
For example the standard noise signal used now has a 10dB crest factor, meaning that the average power (how much heat is produced) is 1/10th the largest peak values.
So if one wanted to deliver 400 Watts with this noise signal to a loudspeaker, you actually need a 4000 Watt amplifier to produce the entire signal.

So the class D amp's like the ice power amps can deliver a much larger peak signal than the continuous value.

For example i used on of those amps that could deliver a peak of 1000Watts...for some seconds and then scaled down ward. If driving a continuous signal after the amplifier heated up, the output scaled back to around 60 Watts, not 1000.

Ok why the change and does it make sense?

Well yes but it makes it much harder to tell what's going on.

The most compressed FM radio signal you can find, still has a crest factor of around 10dB, a decent recording can be around 20dB which means the average level is 1/100 the power of the large peaks. Some exotic recordings can be more, a crest factor of 30dB means the peaks are 1000 times larger than the average level and 40dB is 10,000:1 .

So it may come as a surprise to know that when one reads the SPL with a meter, even on "fast", what you see is an average level, same looking at Voltage with a simpson260, a fast average.

Funny thing, if you look at the output of an amplifier and you play music loudly what you see is well before you hear "amplifier clipping" from the speakers, you will see it on the oscilloscope. You can't hear it as "clipping" until it lasts long enough.

But you can detect it if you can compare back and forth to an un-clipped signal. The unclipped signal (music in our test) sounded a little more dynamic and it actually was.

Bottom line, if your going to try a class D amp, pick one with a good rep and maybe try a larger one than you would normally get. The dynamics present in some modern recordings can literally give one goose bumps.

hope that helps
Tom