# Help identifying this power MOSFET

This MOSFET is from a good board for an RC plane. On a bad board, the MOSFET is completely fried and the number is not visible. The MOSFET drives a motor, and the motor stalled during a crash repair, which caused the MOSFET to pop, as well as damaging a diode (probably the catch/flywheel diode as it is across the motor.) In future, I will add a fast blow ~10A fuse to the motor line to protect the MOSFET.

I can replace the diode with any fast Schottky but I need to know the ratings of the MOSFET and I cannot find a datasheet anywhere. I suspect the pin out is gate, drain, source from left to right, but I cannot be certain. I'm not even sure if it is actually a MOSFET; for all I know, it could be a power NPN transistor. It has no heatsink on it (likely what caused it to fry when the motor stalled) and looks like a TO263 package. The numbers on it are "U438" and "BA8112".

AOU438L from Alpha Omega semiconductor.

http://www.alldatasheet.com/datasheet-pdf/pdf/135752/AOSMD/AOU438L.html

VDS (V) = 30V
ID = 85A (VGS = 10V)
RDS(ON) < 4.5mΩ (VGS = 10V)
RDS(ON) < 6.5mΩ (VGS = 4.5V)


Datasheet mentions it is designed for low-side switching in CPU power conversion. So you could probably scavenge a suitable replacement from an old motherboard.

• 85A, ouch... So how likely is it was fried by a stalled motor? I'm thinking another failure now. – Thomas O May 12 '11 at 20:53
• Maybe the catch diode failed first and overvolted the FET? Don't read too much into ID numbers, they are fantasy. Design with RDS(on) instead. – markrages May 12 '11 at 21:01
• Yeah I noticed that diode went pop at first, powered it up and the fet started blowing smoke, but the motor still runs (it's just uncontrollable, the fet is shorted.) – Thomas O May 12 '11 at 21:14
• I agree with @markrages. These sorts of FETs are hard to kill by current, and much easier to kill by avalanche voltage. You may want to look into a more robust clamp diode for the motor to prevent this sort of damage from happening again. Is that 1k gate resistor in the picture still good? If not, the MOSFET driver may also be damaged. – Adam Lawrence May 13 '11 at 16:51