So I have made a H-bridge with two relay and four IRF520N mosfets. I control them with my microcontroller secured with an ULN2803. However when I connect +5V to the H-bridge board and +12V for the motors only mosfet T21 gets really hot and T22 als gets a bit warm. What could cause it ?

 

IRF520N needs 10V between the gate and source to turn on fully and be really low resistance. You are applying only 5V which means the FET will be higher resistance and hence dissipating lots of power.

 

What could cause it ?

Not reading the data sheet.

The MOSFET's Vgs threshold voltage can vary significantly from unit to unit (see below), so ones with a higher threshold, will have a higher on-resistance with only a 5V Vgs, and thus dissipate more power for a given drain current.



Note the Vgs voltage of 10V for testing the on-resistance of the IRF520N (below).
You need a logic-level type MOSFET where the on-resistance is rated at a Vgs of 5V or less.

 

Depending on your FET drive signals and the pick-up/drop-out times of the relays it seems to me that the relay switches K1 and K2 each provide a possible path for short-circuiting the 12V supply?

 

Depending on your FET drive signals and the pick-up/drop-out times of the relays it seems to me that the relay switches K1 and K2 each provide a possible path for short-circuiting the 12V supply?

Read data sheet
Understand data sheet
Follow data sheet

 

Read data sheet
Understand data sheet
Follow data sheet

Needs to be tattooed on every newbies forehead.

 

Depending on your FET drive signals and the pick-up/drop-out times of the relays it seems to me that the relay switches K1 and K2 each provide a possible path for short-circuiting the 12V supply?

I only put a signal on the first relay and two mosfets with a delay to prevent shortcircuit . However these two didn't get hot.

 

However these two didn't get hot.

We explained that.

 

Not reading the data sheet.

The MOSFET's Vgs threshold voltage can vary significantly from unit to unit (see below), so ones with a higher threshold, will have a higher on-resistance with only a 5V Vgs, and thus dissipate more power for a given drain current.

View attachment 179187

Note the Vgs voltage of 10V for testing the on-resistance of the IRF520N (below).
You need a logic-level type MOSFET where the on-resistance is rated at a Vgs of 5V or less.

View attachment 179188

Vgs wasn't the problem why only two Mosfets got hot. I only controlled the first two Mosfets causing a floating signal on the other two, that's probably why they got hot. Now I control four of them and not a single one gets hot.

 

Vgs wasn't the problem why only two Mosfets got hot. I only controlled the first two Mosfets causing a floating signal on the other two

Oh, really? How do you know what Vgs was on the floating inputs?

Bob

 

Vgs wasn't the problem why only two Mosfets got hot. I only controlled the first two Mosfets causing a floating signal on the other two, that's probably why they got hot. Now I control four of them and not a single one gets hot.

Okay, you lucked out with some MOSFETs that have a lower than maximum Vgs threshold voltage.
But for future good design practice, always use logic-level type MOSFETs if the Vgs control voltage is <10V.