0
\$\begingroup\$

I was using normal MOSFET to turn a 1.2 H+4.8 Ω electro magnet at 48 V and draw around 10 A. The MOSFETs are getting damaged after a few days. The MOSFET is just used as a switch (no continuous switching - once it's on it will remain on for a few minutes and be turned off.) During turn-off off, the electromagnet is discharged through 3.8 Ω via a free wheeling diode. What could be the reason for this strange behavior?

enter image description here

\$\endgroup\$
4
  • \$\begingroup\$ How often does it cycle? What is the physical length between M1, D2 and V3? And what type is M1, and R2 (e.g. wirewound)? \$\endgroup\$ Commented Dec 19, 2023 at 6:13
  • 1
    \$\begingroup\$ You are using 30V mosfet at 48V supply. Also 10A on 3.8ohm developes a spike across coil during disconnection. So together with 48V supply it is far away from spec. \$\endgroup\$ Commented Dec 19, 2023 at 6:26
  • 1
    \$\begingroup\$ What is the reason for the 3.8 ohm resistor? It presents a lot of additional stress on the FET... \$\endgroup\$
    – tobalt
    Commented Dec 19, 2023 at 7:19
  • \$\begingroup\$ The only benefit of having R2 is that it improves the run-down / switch-off time of the magnet, nearly halving it by increasing the run-down resistance from 4.8 ohms to 4.8+3.8 ohms. It throws an addition 38 V stress (10 A * 3.8 ohms) on M1 when it turns off. Is it worth it? \$\endgroup\$
    – Neil_UK
    Commented Dec 19, 2023 at 8:58

2 Answers 2

2
\$\begingroup\$

As mentioned in the comments, you are exceeding the ratings of the transistor.

The IRF1305 is rated for 30V from drain to source. Your circuit has the IRF1305 switching 48V. It isn't surprising that the MOSFET quits. It is surprising that it works at all.

You need a MOSFET with a higher voltage rating. It has to be rated for at least 48V to cover the power supply voltage. Because of the inductor and your discharge resistor, you should really get one rated for at least 100V. The MOSFET will be exposed to 48V + (10A x 3.8 ohms) = 48V + 38V = 86V.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ You need at least 48+10*3.8 V = 86 V. 100 V might be pushing it, unless it is avalanche rated. \$\endgroup\$
    – tobalt
    Commented Dec 19, 2023 at 7:18
  • \$\begingroup\$ I was using the resistor for discharging and MOSFET on the board planned is IRFB4127PBF--(200V) \$\endgroup\$
    – Jim
    Commented Dec 19, 2023 at 7:54
2
\$\begingroup\$

In addition to exceeding the ratings of the MOSFET as JRE mentions, you are also vastly exceeding the ratings of the gate driver. The UCC2751x is rated for a maximum Vdd supply voltage of 18 V, and you're giving it 48 V. I'm surprised it doesn't explode immediately.

This also means that, provided the gate driver somehow manages to work at 48 V, you're giving your MOSFET a gate voltage of 48 V, which I have never seen a MOSFET rated to withstand. Most silicon MOSFETs are designed to be driven at a gate voltage of 10 V, and will be destroyed by gate voltages greater than 20 V, although some can withstand 25 or 30 V.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.