An amplifier without any heating

[AP2]

This is not (practically) possible. Per your example of 300W output, let's be generous and say that is across two channels instead of one, so 2 x 8 ohm (speakers) load = 16 ohm. That means 18.75V rail to rail output through a couple of mosfets each channel, 4 total so let's add 0.8V for those (an impossibly low 0.2V each just to show how impossible the whole idea is), . 300W / (18.75V + 0.8V) = 15A current, and this ignores all the losses in the amp anywhere else including the PSU.

15A * 0.8V = 12W loss (heat) in the last pairs of output transistors alone.

Hi,
Sorry, I wrong dissipation. (not remember, but voltage loss is right)
Mosfet use in SDA-300 have 50mR static rdson.
power stage not use shunt.
@4R (300w) rail=50V; output clip @ 49,2vp
then dissipation is 9,2w.
for audio use, not require a heatSink. (only small profile in photo)

Regards

 

[Pafi]

This is not (practically) possible. Per your example of 300W output, let's be generous and say that is across two channels instead of one, so 2 x 8 ohm (speakers) load = 16 ohm. That means 18.75V rail to rail output through a couple of mosfets each channel, 4 total so let's add 0.8V for those (an impossibly low 0.2V each just to show how impossible the whole idea is), . 300W / (18.75V + 0.8V) = 15A current, and this ignores all the losses in the amp anywhere else including the PSU.

15A * 0.8V = 12W loss (heat) in the last pairs of output transistors alone.

What?????

Where the hell did you get those numbers from?

And how could you ignore Ohm's law? (16ohm, 18.75V, 15A ???)

Theoretical maximum efficiency is 1-(Rdson/Rload). For example with FDP3652 (Rdson=16 mohm, Rload=4 ohm) 99.6 %. But in reality switching loss is dominant in most cases.

And as it's commonly known, nobody listen for sine wave, but music, which has a much lower average power than peak.