# Oscillating sense current in overcurrent protection

I'm stuck at overcurrent protection for SCT3160KL using UCC21732 (or UCC21750, trying to see differences in Desat and Overcurrent protection). This is my circuit

Waveforms

I see that there is a lot of oscillation in the sense current. I put an RC filter with minimum cutoff frequency = 10 times switching frequency (500 kHz), and keeping R53 within 100 ohms, It does not seem to work. What can I do further??

(I chose R54 (sense resistor) as 120 mOhms since the drain current maximum is 17 A).

EDIT 1:- I changed V32 to two voltage sources in series so that it resembles a double pulse test. After reading DKNguyen's answer I changed my circuit a little. Here it is:-

Now I tried to change my RC filter values so that the phase lag between sense voltage and OC voltage is reduced. Here is the waveform:-

Originally I had kept L1 as 1 mH, but I understood that it was too big and changed it to 1 or 2 uH.

My problem is that the OC voltage is going above 800mV (the tripping voltage) when detecting overcurrent. How can I rectify this?

Also, supposing here in this waveform we assume OC trip voltage is correct, and so the FLT_N pin is going low. Then I wouldn't be able to see the other pulse right? I had given a gap of 100 microseconds between the 2 pulses, and I've simulated it for 120 microseconds.

• What is the threshold level for the OC input? Oct 11, 2020 at 7:25
• @Andyaka 0.7 V typically.
– SM32
Oct 11, 2020 at 7:33
• Is L5 an actual component or a parasitic? Because if it is an actual component, why isn't R54 ground referenced? The IC is comparing the current sense voltage to a ground-referenced voltage reference, but yours is floating on top of L5. Oct 11, 2020 at 7:34
• It looks like it works just fine to me then. Oct 11, 2020 at 7:34
• @DKNguyen I added L5 to account for parasitic inductance. Even if I remove L5 it gives similar waveforms, just minor changes in the delay time between /FLT and OC.
– SM32
Oct 11, 2020 at 7:38