MOSFET as a low-drop power-diode

Thread Starter

Robin66

Joined Jan 5, 2016
275
I have a low voltage, high peak power device that I want to protect from reverse polarity. I was going to bias a MOSFET in the usual direction (as in pic) but discussions online suggest biasing so that the body diode (not shown) is forward biased. Can anyone give any indication of the merits of the alternative? To keep my pic simple I tidied the Gate to +Vcc, but in reality I'll have a ~10V source available to ensure the FET is hard on and can handle 30A+ with ~0.3V drop across D-S.
 

Attachments

crutschow

Joined Mar 14, 2008
34,464
If you look at the direction of the body diode you will see why the circuit won't work as you drew.
Connecting the N-MOSFET with source to ground will allow a reverse voltage to forward bias the body diode and thus appear across the power device.
That's why you need to connect the transistor with the drain to ground.
A MOSFET conducts equally well in either direction when ON, but with the drain to ground, the body diode is now reversed biased with the application of a reverse voltage thus blocking the reverse voltage to the device.
 

crutschow

Joined Mar 14, 2008
34,464
Of course, how embarrassing. Is that true? Is the current/voltage-drop capability really as good in either direction? ie. does this graph equally apply when the body-diode is forward biased? (source: https://www.fairchildsemi.com/datasheets/FD/FDD16AN08A0.pdf)
View attachment 99017
It basically follows the same Ids versus Vgs curve below the point where the substrate diode starts to significantly turn on (below a Vds of about 0.6V).
Above that point the transistor can't block the voltage, of course, since the diode starts conducting.
 
Last edited:
Top