My MOSFET gets so hot

Thread Starter

booboo

Joined Apr 25, 2015
168
Guys
I'm trying to control a heater using MCU and MOSFET. my MCU is an STM32 and is operating at 3v3 and my mosfet is F9530N. here is the circuit:



When I connect 24v to the 24v terminal, the MOSFET starts to get so hot as it's about to melt its solder while MCU pin is 0v and the gate voltage is 12v5. Why it's getting hot?

Edit: It apears that the MOSFET is conduncting because the heater starts to get hot.
 

OBW0549

Joined Mar 2, 2015
2,989
When I connect 24v to the 24v terminal, the MOSFET starts to get so hot as it's about to melt its solder while MCU pin is 0v and the gate voltage is 12v5. Why it's getting hot?
I don't see anything fundamentally wrong with your design, and I can think of only a few things that could be causing this overheating:
  • Wiring error;
  • Defective component;
  • Heater draws too much current for the particular MOSFET you are using; or
  • The MOSFET is functioning normally, but it needs a heat sink due to significant power dissipation.
What is the resistance of your heater, or how much current does it draw at 24 volts? We need to know that, otherwise we're just guessing what the problem might be.

I note that according to the IRF9530 data sheet, its Rds(on) is specified as 0.30Ω max, which suggests that with a drain current of more than a couple of amps it could be dissipating enough power to mandate use of a heat sink.
 

ebeowulf17

Joined Aug 12, 2014
2,942
Guys
I'm trying to control a heater using MCU and MOSFET. my MCU is an STM32 and is operating at 3v3 and my mosfet is F9530N. here is the circuit:



When I connect 24v to the 24v terminal, the MOSFET starts to get so hot as it's about to melt its solder while MCU pin is 0v and the gate voltage is 12v5. Why it's getting hot?

Edit: It apears that the MOSFET is conduncting because the heater starts to get hot.
Gate voltage of 12.5V implies to me that Q1 is conducting. You could try disconnecting Q1 from R9 (or completely remove either of those components) to see if Q2 still passes current. That would isolate the problem somewhat.

EDIT:
On second thought, I'm not so sure. If this were the problem, you wouldn't expect Q2 to heat up so fast. Seems likely that there's something else going wrong. Any chance Q2 is in backwards and conducting through its body diode?
 
Last edited:

Bordodynov

Joined May 20, 2015
2,402
The dissipated power of the transistor is approximately equal to P=I_Load^2*Rds(on)
If Rds(on)=0.3Ohm and I_Load=10A ==> P=30W , I_Load=5A ==> P=7.5W.
If the transistor is without a heat sink, then it will heat up very quickly!
 

Thread Starter

booboo

Joined Apr 25, 2015
168
What voltage do you see between Gate and Source? On / and Off?
For On: Gate:11.8 & Source:17 & Drain:16
For off: Gate:12.8 & Source:17 & Drain:13

...What is the resistance of your heater, or how much current does it draw at 24 volts? We need to know that, otherwise we're just guessing what the problem might be...
8Ohm
It's supposed to draw 3.5-4A

guys
I think the MOSFET doesn't turn off completely. How about to change R10?
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
2,942
For On: Gate:11.8 & Drain:17 & Source:16
For off: Gate:12.8 & Drain:17 & Source:13
Do I have my drain and source mixed up? Shouldn't the source be be higher than the drain in this configuration? More importantly, shouldn't one of them be 24V, unless the power supply isn't keeping up and there's some major voltage sag here under load?
 

OBW0549

Joined Mar 2, 2015
2,989
For On: Gate:11.8 & Drain:17 & Source:16
For off: Gate:12.8 & Drain:17 & Source:13
Those readings don't make ANY sense at all; are you sure you don't have a wiring error? Especially, are you sure you've got the MOSFET connected properly? Consult the data sheet for the pinout.

8Ohm
It's supposed to draw 3.5-4A
If that's the case, then at Id = 4A the MOSFET is dissipating 4 * 4 * 0.30 = 4.8 watts, and unless you have it mounted on a heatsink, it will get very, VERY hot since the case-to-ambient thermal resistance is around 60 °C/watt.

I think the MOSFET doesn't turn off completely. How about to change R10?
Changing R10 won't do a darned thing. It should work as-is, provided the MOSFET is adequately heatsinked. If you ground the MCU pin input, or remove jumper W5, does everything cool off? If not, that suggests a wiring error or short somewhere.
 

Thread Starter

booboo

Joined Apr 25, 2015
168
Do I have my drain and source mixed up? Shouldn't the source be be higher than the drain in this configuration? More importantly, shouldn't one of them be 24V, unless the power supply isn't keeping up and there's some major voltage sag here under load?
Oh, Sorry. Edited it. Thank you.
View attachment 130808
Check your connections to the BC547 too.
YES! that was the problem. I changed the direction and now my circuit is working correctly.

Thank you guys.
 

OBW0549

Joined Mar 2, 2015
2,989
YES! that was the problem. I changed the direction and now my circuit is working correctly.
That would certainly cause the symptoms you saw; if the emitter and collector of the BC547 were swapped (i.e., the transistor turned around), the transistor would appear to be "partially ON" all the time because the B-E junction would be breaking down and acting like a Zener diode with a Vz of 5-7 volts.

One other thing you might consider doing to reduce the power dissipation in the MOSFET is to use one with a lower Rds(on).
 

Marco-444

Joined Jun 14, 2018
2
Hello,

I am interrested for using this circuit, but I would like to know if you could give me the reference of the D4 zener diode you are using, or to give me the test current (sorry i am a beginner in analog electronics and i am not able to know how to calculate the current going through D4 when conducting...).
 

Marco-444

Joined Jun 14, 2018
2

ebp

Joined Feb 8, 2018
2,332
Marco, I can't see the schematic in the original post. It might be best to start a new thread with your questions.

however -
Zeners are specified at some specific test current. It doesn't mean you have to use them at that current. The only real requirement is that you don't exceed the power rating, and the power rating is based on temperature of the actual zener chip.

The zener you are considering is specified at 5 mA. The power rating 500 mW, if the temperature at the leads is no more than 75°C, At 15 volts, this works out to 33 mA. A zener is not a perfect voltage regulator, so the voltage varies somewhat with current. At 33 mA the voltage will be a bit higher than at 5 mA. If the current through the zener isn't continuous, it could handle even more than 33 mA, though the voltage would be higher still. For example, you could probably very safely allow it to conduct 100 mA if the duty cycle were no more than 25-30% AND the conduction time were short. 1 ms conduction and 3 ms off should be OK. 1 s on and 3 s off probably wouldn't because the temperature would get too high during the 1 second intermal.

Since I can't see the original circuit for some reason (odd, because apparently you can) I don't know if any of this is of any help at all.
 
Top