RE: RF Mosfets

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Mccauley, Daniel H" <daniel.h.mccauley-at-lmco-dot-com> 


I'm curious to how you are going to be running these devices?

If you are going to switch these like standard SSTCs (ON/OFF) square
wave, then these RF FETs won't
work that well.  These FETs are primarily designed to be switched by a
sinusoidal type waveform and
act linearly.

If you are looking for a simple high power SSTC FET which can dissipate
lots of power, then the low-cost
IXYS IXFN type SO-227 ISOPAK FETs are just fine and a much better bet.

The particular FET you are looking at has a Rds of 0.11 ohm.  Typically,
the Hi-Per FETs (ISOPAK) FETs have
Rds on the order of 0.01 ohms.  Thats an order of magnitude lower.

So assuming everything else is the same, if your system switches that
IXYS FET with a loss of 10W, your DE475 FET
will be dissipating 100W.  Sure, the package can dissipate more heat
(only if cooled properly), but you will be
dissipating almost an order of magnitude more heat the a different FET.

You can get standard IXYS Hi-Per FETs for $10.00 each in quantities of
10 from www.future-active-dot-com

Dan



 > Some posted earlier in the list of some massive mosfets
 > (1800W Pd, .11 ohm
 > Rds-on, 500v, 48A, 30 MHz). Part # DE475-501N44A. These
 > aren't as bad price
 > wise as I expected. I got them for $71.05 each plus shipping
 > (100+ gets
 > discount). Could these be run past their current rating with slower
 > switching (never heard of a 30 MHz tesla coil, 5 MHz rarely)
 > as long as the
 > Pd isn't exceeded? Anyone got any links to calculate power
 > dissipation with
 > certain voltage ,current, waveform, frequency, and ambient temp? I'm
 > waiting on a quote from another company on a slower version
 > of these (high
 > power types, 1000W+). Thanks all.
 >
 >
 >