Arduino resets randomly when using a DC motor

Question

Following a question I posted here a few months ago Using P-Channel mosfet with arduino I'm using pchannel MOSFET for activating a CR02 valve like this one https://es.aliexpress.com/item/1938516987.html?spm=a219c.search0302.3.30.33c06d8djwoxkt&ws_ab_test=searchweb0_0,searchweb201602_0,searchweb201603_0,ppcSwitch_0&algo_pvid=42bb562b-db52-4060-8454-5a31a442a8ec&algo_expid=42bb562b-db52-4060-8454-5a31a442a8ec-4

enter image description here

This valve has 3 wires, one goes to +12V the other goes to GND and the remaning one is used to open/close the valve, if this wire is conected to +12V it opens the valve and if it's unconnected it closes the valve. I'm using the following circuit to operate this valve

enter image description here

VIN = +12V
QD = Arduino PIN
QDV = Valve control wire

It kind of works, I can open and close the valve but once every 6 or 7 times opening or closing it causes the arduino to reset and I don't know why. I'm using a 2A power supply and the specifications of the valve says it consumes 100mA so it should be enough, I'm also using a diode to protect from the spike of voltage when the motor is turned off, any idea what can be causing the problem?

• Bruce Abbott: Might be a 'ground loop'. Can you show us a photo of the wiring?
• Marko Buršič: 1n4001 could be a little slow.
• Marc: Which kind of diode should be better?
• Marko Buršič: A fast switching diode, but also it could be that your PSU has no input filter and very low capacitance. A differential choke or LC filter would also help.

Answer 1

As @Marko wrote, a slow diode can do quite a bit of trouble. Let's look at his model to see if we can find something. dióda1

Yes, it is here: dióda2

Since it was a load around 100mA, let's try a small switching diode: dióda3

It got significantly better, but this diode will still be small here. Let's look at a better one: dióda4

So the diode would be fine. If we’re already modeling, let’s do something about the huge latency caused by the too high value Gate resistor: dióda5

The 10k resistor was changed to 1k and a miracle happened: dióda6

The delay was almost completely gone. So replacing a diode and a resistor has improved the system a lot. Finally, here are all the interesting signs from this tiny system: dióda7

If nothing helps, replace the ground wire with a thicker one. :)

• Marc: I like that you have shown the troubleshooting process and not only the final answer, very instructional, hats off

Answer 2

Kind of an unsatisfactory-looking fix, but you may be able to improve the by slowing the turn-on of the MOSFET (turn-off is already slow).

For example, add a 10K resistor series with the MOSFET gate.

That will reduce the demands on layout, bypassing, and even the diode since it will further slow turn-off.