30v MOSFET fails at 24 volts

[scott216]

Edit: Fixed MOSFET P/N and link.

I've got a n-ch MOSFET  P/N IRLML6344TRPBFCT-ND (Digikey link) hooked up for reverse voltage protection (see attached). 
I partially populated my PCB and applied power to the board.  I started low at 7 volts and worked my way up to 24 volts.  24 volts is the nominal operating voltage.  At 20 volts things were fine and the circuit was drawing less then 20mA.  But then I bumped the voltage to 24 volts and my MOSFET shorted out.  I had my power supply limit the current to 200 mA, so there was no magic smoke.  I want back to the data sheet and saw the the V_DS is 30 volts, but the V_GS +/- 12 volts.  So I guess it's the V_GS that got me.   Lesson learned.

[codeboy2k]

Yep.  Vgs did it.

I've got a n-ch MOSFET  P/N MC7812BDTRKG (Digikey link) hooked up for reverse voltage protection (see attached). 

I think you got confused.  MC7812 is a 12V regulator.  But later you linked to a IRML6344PBF mosfet, so I think you just got confused here.

When the mosfet is fully turned on, as in your case, then the Drain and Source are almost at the same potential. So Vgs is equal to your power supply voltage, minus the voltage drop Vds, which is Vds = ( Rds(on) * Iload ). This will be small since Rds(on) is small. So Vgs is always running really close to your supply voltage here.

The solution is to put a 10k resistor in the gate, then put a 12V zener diode across the Gate and Source pins. Vgs(max) is 12V on that device.

[scott216]

Yep.  Vgs did it.

I've got a n-ch MOSFET  P/N MC7812BDTRKG (Digikey link) hooked up for reverse voltage protection (see attached). 

I think you got confused.  MC7812 is a 12V regulator.  But later you linked to a IRML6344PBF mosfet, so I think you just got confused here.

When the mosfet is fully turned on, as in your case, then the Drain and Source are almost at the same potential. So Vgs is equal to your power supply voltage, minus the voltage drop Vds, which is Vds = ( Rds(on) * Iload ). This will be small since Rds(on) is small. So Vgs is always running really close to your supply voltage here.

The solution is to put a 10k resistor in the gate, then put a 12V zener diode across the Gate and Source pins. Vgs(max) is 12V on that device.

I fixed the P/N and link in the original post.  That voltage regulator is also in my project and I wasn't paying close enough attention to what I was cutting and pasting.

Thanks for the resistor + zener diode solution. 

[rx8pilot]

I would only add that going with a 10v zener would give you some safety margin. The gate should be fully saturated at 10v and you are a little bit further from blowing it up.

[Simon]

Very few mosfets go over 20V for the gate and that is an absolute maximum 12-15V is more than enough and the datasheet specs generally quote against 10V on the gate. Don't worry easy mistake to make, I saw a post grad engineer do this on a peice of kit that was meant to keep people alive, needless to say we changed subcontractor.

[scott216]

I like my method of limiting the current on my power supply for new circuits I just built.  It keeps things from blowing too quickly.