Some FETS (eg FREDFETS)(see superb free power semiconductor textbook
referred to a few days ago)  have fast internal reverse diodes. The "body
diode" in standard FETs tends to be slow and have a higher voltage drop than
the FET itself. Even if characterised it will generally be inferior to a
good diode.

IF you are able to turn the FET on (with "normal" FORWARD bias when you want
it to conduct in the REVERSE direction (the same direction as the reverse
diode) you get the best of all worlds. Essential;y a synchronous rectifier.

As an example an IRF250 is rated at 18 milliohm Rdson = on resistance, and
in excess of 20A. At 20A the drop would be about 0.35 volts. A Schottky
diode would be hard put to better this and you will find vanishingly few 200
volt Schottky diodes. A lower voltage FET of similar current rating will
have an even better Rdson.



        Russell McMahon



> *>Yes, it works well. On inspection of some data sheets it appears that
> *>most power MOSFETs have fully characterised D-S diodes with
> *>current/voltage ratings the same as the switching element. To me that
> *>implies that they are made for use as flyback diodes and the like.
>
> No, the diode results from the implementation geometry of a vertical
> mosfet transistor. PIC 'protection' diodes are 'made' the same way. Or so
> it is rumored. When they are characterised then they are safe to use ...
> but I don't think quick recovery because of Cgs/Cgd. They are certainly
> not shottky diodes.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.