Gate driver latching

[Miroslavus]

Hi Everyone 

I'm trying to drive a coil with an arduino. I'm using IRF3205  MOSFET to control the coil and a gate driver MIC4422YN. Schematic is also attached. Whenever I turn on the coil it looks like the gate latches. The coil stays on and I still have 12V on the gate of the MOSFET and around 3V on the input pin of gate driver despite it being pulled down. Oddly enough I have actually made six of these circuits on PCB and only 3 of them have the problem. I have triple checked and I could find any difference.

I have tried some troubleshooting myself and I found out that without any load the driver won't latch  . I'm using bench lab power supply(3A max) and I've noticed that when I remove the flyback diode the latching is fixed but I need the diode as I want to be powering it with batteries. I have replaced the MOSFET and gate driver multiple times but I don't think they are the problem.

I have also tried putting 1 resistor in series with flyback diode and it help a bit. It would latch just occasionally. The flyback diode I'm using is  BYP60A6. If anyone would be able to help I'd really appreciate that 

[fourtytwo42]

Well there is nothing wrong with your schematic EXCEPT you have no gate resistor on the mosfet and one hell of a hefty driver so it's gonna turn on VERY fast. For some reason you are using a HUGE flyback diode who's reverse recovery is probably awfull so basically you have a big current spike at turn-on. That combined with potential layout problems could explain your problem. Please post pcb layout and why the presfit diode ? also solonoid charecteristics could be helpful

[Miroslavus]

Thank you for the quick response 

I have tried adding 68ohm on the gate and it actually fixed the problem. Thank you for that  . I had couple more questions. The value of 68ohms is something I have used before. What it the proper way to calculate this?

Also the diode is so massive because I'll driving it with battery and the current will be around 55A. When I was doing research I found out that the flyback diode should be able to handle as much current as will be flowing trough the inductor. Correct me if I'm wrong on that or if there is better way of doing this . It's press fit simply because that's what they had in local store 

[fourtytwo42]

Thank you for the quick response 

I have tried adding 68ohm on the gate and it actually fixed the problem. Thank you for that  . I had couple more questions. The value of 68ohms is something I have used before. What it the proper way to calculate this?

It is rather tedious to calculate so most do it by observing waveforms and rule of thumb! It would be helpful to know the charecteristics of your solonoid, but I guess the current is maybe 30A. If the mosfet turns on to slowly (gate resistor to high) the peak power during turn-on will be to high. So maybe I would sugest trying 22R. Another problem is the higher the resistor the slower the turn-off and thats very undesirable, a way of overcoming that is to shunt the resistor with a small schottky diode (BAT85) with cathode towards driver.

Also the diode is so massive because I'll driving it with battery and the current will be around 55A. When I was doing research I found out that the flyback diode should be able to handle as much current as will be flowing trough the inductor. Correct me if I'm wrong on that or if there is better way of doing this . It's press fit simply because that's what they had in local store 

The diode only handles a peak current while the magnetic field collapses at turn-off not continious but depends on switching frequency that imagine is very low. Anyway with a fast switch (mosfet) it is important the diode is fast to avoid the problem you have seen. A 5amp schottky would easily do this but depends on what you have localy available. As long as you keep the switch turn on moderatly slow just keep the diode you have but if you are repeating this design for others you need a faster diode.