Devices we may be using in a few years for higher switching speed Class D

[bwaslo]

http://www.irf.com/product-info/ganp...roganpowir.pdf

Is the limit for Class D switching speed/rate the FETS, or other devices?
Or is switching faster even a good thing to go after?

[bwaslo]

for those without powerpoint (or wanting more detail):
www.powersystemsdesign.com/2009/na/JanFeb/power_conversion.pdf

[iand]

Originally posted by bwaslo
http://www.irf.com/product-info/ganp...roganpowir.pdf

Is the limit for Class D switching speed/rate the FETS, or other devices?
Or is switching faster even a good thing to go after?

Faster switching reduces loss at higher clock frequencies, which opens up a lot of possibilities for efficient power D/A conversion (like Zetex DDFA which uses 850kHz) -- for "conventional" class D amps there's little advantage in using more than 400kHz.

It's not clear from the article but it looks as if there's no body diode, which means a much faster separate diode could be used to reduce switching loss -- the best 200V silicon MOSFETs have trr around 100ns, discrete diodes are up to 4x faster.

So we could end up with half the power losses even if the switching frequency doubles.

The 200V devices aren't available yet, though :-)

Ian

[iand]

According to this (p.16) there is a body diode but it's also several times faster than silicon with about 5x lower Qrr:

http://www.irf.com/product-info/ganp...roganpowir.pdf

Would make a damn good class-D output stage, then...

Ian

P.S. 200V devices are planned for 2010