EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: step_s on November 24, 2016, 11:54:25 pm

Title: High side N-MOS fried. Voltage too spiky?
Post by: step_s on November 24, 2016, 11:54:25 pm

Hi EEV'ers.
So I just built a buck setup with an FDS9926A https://www.fairchildsemi.com/datasheets/FD/FDS9926A.pdf (https://www.fairchildsemi.com/datasheets/FD/FDS9926A.pdf)
The dual N-MOS is driven by a NCP5901 http://www.onsemi.com/pub_link/Collateral/NCP5901-D.PDF (http://www.onsemi.com/pub_link/Collateral/NCP5901-D.PDF)
I tested it all, and it worked perfectly with 5V on the controller, and 5V on the buck setup, but alas, this was not the case with 12V on the buck setup. The high side N-MOS's gate is toasted, and I'm now wondering why.

The controller runs on 5V and the Drain of the N-MOS was at 12V (input of the buck converter). The boost on the controller was supplied with 12V through a diode, in order to make it run at 12V plus the 5V from the controller. Since the MOSFET has a maximum rating of 10V from gate to source my guess is that the SW (connected to Source) was still climbing when the high side drive instantly went to 17V, frying the gate.

So how do you deal with this? Other than keeping the gate-source rating above input + boost? Resistor to gate to introduce some sort of slew-rate?

Title: Re: High side N-MOS fried. Voltage too spiky?
Post by: EPTech on November 25, 2016, 03:06:40 pm

Hi there,

A low capacitance zener diode of about 15V across gate-source will do just fine.

Happy converting. ;)