Need help about voltage spikes

how to get rid from voltage spikes

Hi everyone,

I'm trying to make a dc motor driver circuit using 2 relays for direction control and one MOSFET for switching. The schematic of my circuit is given below. The motor is connected to the connector J3 on the schematic. The supply voltage is 24V and the motor draws 6A of current at this voltage.



The problem i faced here is voltage spikes and oscillations across the MOSFET and the DC motor. Also the motor and its parallel diode gets hot even they're rated for a much higher working current and voltage. I tried to connect capacitors paralel to the MOSFET and the spikes are gone but this time the turn-on time of the MOSFET increased and both the motor and its parallel diode got hot. I also checked the ground rail and there wasn't any oscillations.

Here is a .doc file which contains the waveforms i measured with the oscilloscope:
**broken link removed**
I took these measurements from the circuit when there wasn't any capacitors connected across the MOSFET.

Especially the voltage across the motor is very bad. What can be reason of this and how can i get rid of these spikes and oscillations?

I also tried to put a TVS across the MOSFET and it clamped the oscillations but this time the TVS heat up and burned.

voltage motor spikes

First off if your components are getting hot you have too much current flowing through them.

As for the spikes and oscillations you should ditch the caps and use a snubber circuit.

I would also go back and look at all my parts current rating and check to see if they are at least 2x the amount of current you are seeing. If they are not they should be.

Also the MOSFET should be automotive rated just to be safe.

dc motor transzorb snubber

The current rating of the MOSFET is nearly 10x higher than the current flowing through it in this circuit. Voltage rating is also more than 2x. Same thing applies to all of the diodes in the circuit. So the amplitude of the peak current and voltage shouldn't heat the MOSFETs since they're much lower than the rated values. Couldn't be there any other reason that can cause the MOSFETs or diodes to get hot?

I know that spikes and oscillations should be handled by a snubber and therefore also tried a RC snubber across the MOSFET but it didn't change anything. I calculated the values of the snubber elements by looking to an application note (an280 from ST). I also tried different values for R and C and they didn't cause much change. But i'm not %100 sure that i used the right values anyway. Maybe i miscalculated the values of the snubber elements. If so, how can i find out the right way to calculate it?

The circuit is on a breadboard. Could it be the inductance of the wires, which cause these oscillations? I don't think so because the amplitude of the oscillations are nearly twice the supply voltage.

You also may want to read this the section on catch diodes.

Regarding the heating have you checked for commutation over lap?(slow turn off mosfet still conducting and drain still blocking)

If your RC snubber had no effect then yes you probably calculated wrong.You could try a zener (transzorb) clamp,or a RCD rate of rise clamp.
Heres another AN on snubber's
They usually take some experimentation at least for me not to often your first attempt will give the desired results.

And yes from personal experience a breadboard switching fast high peak currents can create problems that wouldn't exist on a properly laid out board.

Good luck