I've read all the MOSFET related posts here, but none seem to answer my question...

See image for clarification. When the Arduino UNO digital output 3 is set high, the motor does not turn on. With output 3 still high, I probe the motor terminals with a multi-meter and the motor begins to run and VERY slowly climbs in RPM. The measured voltage slowly climbs from a few volts. When the meter reads 10V, the MOSFET was so hot it melted the breadboard. The motor is: http://www.mabuchi-motor.co.jp/cgi-bin/catalog/e_catalog.cgi?CAT_ID=rs_365sh

Is this resulting from the MOSFET's inability to handle the inrush current from the motor? Poor MOSFET Choice? What am I missing here?


  • 2
    \$\begingroup\$ That image is not clarification. Draw a schematic instead. \$\endgroup\$
    – user16324
    Commented Mar 27, 2016 at 22:05
  • \$\begingroup\$ have you set pinMode to output? you need that for pin current handling. By "-18V" in the image did you mean GND (0V)? can you probe the Gate voltage and the FETs Rds(on)? \$\endgroup\$
    – Wesley Lee
    Commented Mar 27, 2016 at 22:06
  • 1
    \$\begingroup\$ Welcome to EE.SE, @B_Stryk. As this is an engineering site we like circuit schematics and don't like wiring diagrams (unless it's required). The schematic shows design intent and current flow can be worked out. This is difficult in a wiring diagram. There's a schematic button on the editor toolbar. \$\endgroup\$
    – Transistor
    Commented Mar 27, 2016 at 22:17
  • \$\begingroup\$ What about a start capacitor? \$\endgroup\$ Commented Mar 28, 2016 at 0:26
  • \$\begingroup\$ @B_Stryk did you fix your problem? \$\endgroup\$ Commented Mar 29, 2016 at 19:08

4 Answers 4


Two problems:

  1. The gate threshold voltage (Vth) is 3V-5V and the Arduino can only supply 5V. It is possible that you are only turning the part partially on. If you wish to drive the pin with an Arduino output, your max gate threshold voltage should probably be in the 4V range.

  2. The MOSFET that you have chosen has a nominal 2.2ohm resistance, which is pretty high for driving a motor. The motor link that you posted is actually for two models of motor, one that operates at 5.2A is the low voltage variant. If this is truly drawing 5.2A (I suspect that it isn't, but it is all of the info I have) AND the MOSFET were turned fully on, you would have

Pdissipated = 5.2A * 5.2A * 2.2Ω = 59.5Watts

If you have the other variant and your motor is drawing 1.32A, the power dissipated on the switch is still high for a plastic-tabbed FET:

Pdissipated = 1.32A * 1.32A * 2.2Ω = 3.8W

Your power dissipated on your switch is very high.

Do this:

  1. Hook up your motor to your power supply without the MOSFET and measure the current
  2. If your current is too high for this MOSFET, pick a MOSFET with a lower on-resistance and (preferably) lower threshold voltage

If you have to choose another MOSFET, choose a metal-tabbed MOSFET so that you can add a heatsink if necessary. Plastic tabs are great, but not for prototyping. Ideally, you want to dissipate as little power on the MOSFET as possible. Anything with an on resistance of 0.1ohms and a gate threshold voltage of less than 4V should work quite well.

To explain your symptoms, as the MOSFET heats up, the part actually becomes slightly more efficient. If you look at the operating curves vs. temperature, you generally get some small benefit... right up until failure. This is why the motor starts to speed up as your MOSFET gets warmer. Less voltage is dropped on the MOSFET compared to the motor.

  • \$\begingroup\$ What about a start capacitor? Just a suggestion. \$\endgroup\$ Commented Mar 28, 2016 at 0:25
  • \$\begingroup\$ A start capacitor will just make things worse. Plus, your motor is only rated for 6 to 20 volts, but you're trying to drive it with 36 volts. \$\endgroup\$ Commented Mar 28, 2016 at 0:49
  • \$\begingroup\$ @TimSpriggs You definitely need capacitance to protect the arduino from the motor, but the motor needs little such protection. The circuit should work without much modification as long as the MOSFET is appropriately selected. Although, how much current can the average breadboard withstand without damage...? \$\endgroup\$ Commented Mar 28, 2016 at 2:30

Thanks for all the feedback. slightlynybbled nailed the solution. I selected a new MOSFET (Digikey IRLB8721PBF-ND) and the problem was solved. This new model offers Rds of 8.7 mOhm for low power dissipation and Vgs of 2.35V. The Arduino 3.3V output level easily meets the requirements. This MOSFET model also offers a metal tab for heat control, but I'll unlikely need it as the circuit dissipation is 2.5A * 2.5A * .008 ohm = 40mWatts. Btw...my motor is running on 18V. I just did a poor job at indicating that correctly in the attached image.

I'll include capacitance in my circuit for Arduino protection as suggested. And for clarification, in the future I'll post a schematic as opposed to an image. Point taken.

  • \$\begingroup\$ Please, no signatures and closing 'Cheers' / 'Thanks'. Others will mercilessly downvote you. Say thanks by upvoting those who help you and recognize that sigs aren't welcome in this SE community. \$\endgroup\$ Commented Mar 30, 2016 at 17:19

I recommend using an opto-isolator between the Arduino output and the MOSFET. I also think the gate voltage is too low preventing full switch-on.

Here's an example with opto-isolator (note that the control will be inverted: when arduino pin is ON, motor will be OFF)


simulate this circuit – Schematic created using CircuitLab

These MOSFETS are nice for such tasks. They are rated for high voltage/current, have a very low on-resistance, and include built-in diode protection. Plus, you can get a few free samples if you ask nicely! :)


I am no expert, but I had a similar problem before. Most probably your MOSFET is not fully on at 5V at the gate. Even if the datasheet says something like threshold voltage is around 4V it looks like this is the point when the MOSFET begins to open, not when it is fully open. When in this "partially open" state MOSFETs tend to let less current flow through them and dissipate the rest as heat. Again, I am no expert, so you better don't take my word for it, but this was my observation and understanding about what is going on when I had practically the same problem as you.

  • 1
    \$\begingroup\$ "MOSFETs tend to let less current flow through them and dissipate the rest as heat", that is not what happens, a FET cant divert eletron flow to heat. Its resistance increases and current will cause more heat to be wasted at the device. You dont "lose part of the current" as heat.. \$\endgroup\$
    – Wesley Lee
    Commented Mar 27, 2016 at 22:09

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.