Blowing up MOSFETS

Thread Starter

Tobias

Joined May 19, 2008
158
Using 24v system to power an electric motor in one direction. I have a low side gate driver to control the gate on the N Type MOSFET. My problem is that the MOSFETS short out when I try to PWM the gate using 24 volts. To circumvent the kickback current I have an ES2D rectifier. The circuit works great using 12v. On 24v it works great until I start to pulse the Gate, then the gate to drain shorts out quickly. What am I doing wrong? Do I need a cap between the drain and 24v?
Attached is the schematic, MOSFET spec and motor spec
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
What is your Vdd? If it's not 10v to 12v, that is why you are having problems.

I don't see a resistor between your gate driver and the gate. You should have a small resistance value there, close to the gate as possible, or you may have "ringing" on the gate. The 'ringing' will turn the MOSFET on and off very rapidly, causing heat dissipation.
 

jpanhalt

Joined Jan 18, 2008
11,087
The circuit works great using 12v. On 24v it works great until I start to pulse the Gate, then the gate to drain shorts out quickly. What am I doing wrong?
Neither your driver nor your mosfet is rated for 24 V. See Fig. 41 in the Fan3121 datasheet. Its output voltage is roughly the same as its supply voltage.

John
 

Thread Starter

Tobias

Joined May 19, 2008
158
VDD is 5v.

The MOSFET is rated for 60V and 45A.

I am feeding the driver with 5v.

If I turn on the motor and leave it running at 100%, all is good. I check resistance between source and drain and its good. When I try to run the motors at say 50% then the MOSFETS go into a permanently closed-state.

Do you think the ES2D is adequate? If I put two in parallel, will I get twice the 'protection' on the kickback current when the MOSFET turns off during PWM?
 

eblc1388

Joined Nov 28, 2008
1,542
VDD is 5v. The MOSFET is rated for 60V and 45A.
I am feeding the driver with 5v.
Your MOSFET is not of the logic level type. You might be able to get away with 100% duty using 5V gate voltage but certainly not PWM.

Either drives the existing MOSFET gate with 10V PWM or use a logic level MOSFET.
 

Thread Starter

Tobias

Joined May 19, 2008
158
From the spec sheet Gate to Source Threshold Voltage 2-4v. Doesn't that mean that 2-4 volts will turn it on?
 

Audioguru

Joined Dec 20, 2007
11,248
The threshold voltage for a Mosfet is the gate voltage where it is barely turned on (0.25mA) or is barely turned off. Yours is probably partially turned on which makes it very hot.
Your Mosfet is guaranteed to turn on fully when its gate to source voltage is 10V.
 

SgtWookie

Joined Jul 17, 2007
22,230
No. You'll fry the MOSFET.

Look at the charts in the datasheet; Figure 7 (transfer characteristics) and Figure 8 (saturation characteristics) for Vds vs Vgs vs Id.

When Vgs=5v, the MOSFET is barely conducting.
 

eblc1388

Joined Nov 28, 2008
1,542
Tobias, I have lost count of how many times people have taken Vgs(th) to mean the voltage that the MOSFET should be driven normally.

The same question had popped up at least once a month in forum if not more frequent.

Tips for looking at MOSFET datasheet. If one see any specification item which stated Vgs=10V as part of the measurement criteria, then the MOSFET should be driven with 10V.

 

Attachments

Top