Hey Guys,

I am doing a design revision on a quite powerful converter and here is my question.
There are these SiC MOSFETs that I need to drive. We are using IXDN630YI ICs for gate drive and in order to achieve fast turn off we are using anti parallel diodes to gate resistors. I need to make this circuit quite as compact and I could not think a good reason not to implement the same circuit as the second drive option, where I do not need the diodes for fast turn off. Am I missing something? Thank you for your time.

 

Do You actually need Gate-Resistors ?, or are You guessing ?

The Gate-Driver, and the Gate it's self,
both have a small amount of Resistance which "may" perform just fine,
( if the Gate-Driver is up to the task ),
depending, of course, on the reactions of the rest of the components in the Circuit.

Gate-Resistors are mostly needed for dealing with
the bad behavior of other Reactive-Components in the Circuit.
.
.
.

 

Hi,
The second option could cause some problems with the UVLO of the IXDN630 when the voltage drop on the resistors is to high.
I would use the gate-resistors because your gate driver has almost no ressistance and there will probably a lot or ringing at the gate because of the gate capacitance and the inductance of the TO247 package alone.

 

Do You actually need Gate-Resistors ?, or are You guessing ?

The Gate-Driver, and the Gate it's self,
both have a small amount of Resistance which "may" perform just fine,
( if the Gate-Driver is up to the task ),
depending, of course, on the reactions of the rest of the components in the Circuit.

Gate-Resistors are mostly needed for dealing with
the bad behavior of other Reactive-Components in the Circuit.
.
.
.

I need the gate resistors because as StefanZe already stated the internal gate resistors are too low and designed for damping the oscillation up to (approximately) 10 nH of inductance with combination in gate source capacitance in gate driver loop. Any higher gate drive loop inductance means I would have oscillation in my step response (Vgs).

Hi,
The second option could cause some problems with the UVLO of the IXDN630 when the voltage drop on the resistors is to high.
I would use the gate-resistors because your gate driver has almost no ressistance and there will probably a lot or ringing at the gate because of the gate capacitance and the inductance of the TO247 package alone.

That is what I have also found right after posting this. UVLO function of the driver will most likely trigger with the initial high turn-on current.

It seems a bit cheeky but what about if I put anti parallel zener diodes to resistors (in the second option) in order to prevent UVLO?

 

It seems a bit cheeky but what about if I put anti parallel zener diodes to resistors (in the second option) in order to prevent UVLO?

I don't think the zener diode will be fast enough. But you can test it.
Why do you need 2 resistors and 2 diodes in the first one? I think one resistor and one diode should be enough.