Controller builders: a gem about shoot through protection.

[h0tr0d]

Excellent topic right here, really GREAT info!

Teh Stork, thanks for sharing. :wink:

Almost bought this fet's today...
IXFK-FX230N20T
http://ixapps.ixys.com/DataSheet/DS100133A(IXFK-FX230N20T).pdf

After reading this topic, I'm not sure it's a good idea, trise=35ns tfall=29ns, trr=200ns max.


My idea was a XieChang (Infineon) controller and a daughterboard with 12 of those, phase current limited to 150A with Allegro Sensors, good array of low ESR/ESI caps, 40S Lipo Battery and a custom wound c80-100.

 

[h0tr0d]

Speaking of reducing dead time, what do you guys thinks about this:
http://endless-sphere.com/forums/viewtopic.php?f=30&t=35387&start=75#p528810
"Put a diode in parallel with the gate resistor to get a fast turn off time and still have the same turn on time."

 

[bearing]

Later in that thread:

Word of warning about diode return path. When I added this, switching off definitely was faster - but under load the mosfetdrivers popped like popcorn. Adding 4ohm in series with diode fixed the problem (10ohm normal resistor). Negative transient on HS (high side fet turn off) seems to be the thing that blew the drivers.

http://endless-sphere.com/forums/viewtopic.php?f=30&t=35387&start=285#p662228

 

[Arlo1]

I used a diode parallel to 30 ohm resistor then followed by a 3.4 ohm resistor in series. SO I have 33.4 ohms on resistance and 3.4 ohms off resistance and it works I tested it to 170v!
With my ixfk230n20t fets. I am just redesigning the driver boards ATM to get them close to the fet legs. You should always have some resistance on and off.