200V/ns CMTI for bootstrap gate drive

[cupoftea]

Hi,
We now have the NV6128 offering record breaking 200V/ns in bootstrap hi side drive. With 650V GaNFET on chip.

https://www.navitassemi.com/wp-content/uploads/2020/07/NV6128_Datasheet_Final-12-29-2020.pdf

This looks like wiping out all gate drive transformer solutions the world over? (whether LLC, 2TF, HB, FB, PSFB)

 

[mtwieg]

Yes, new gate drivers are coming out which are suitable for hard switched GaN circuits.

If you want to use GaN FETs, then you should definitely consider using such new drivers. But most of us are still comfortable with MOSFETs, and these new drivers don't offer any advantages to MOSFETs.

 

[cupoftea]

Thanks, 200V/ns isnt seen in any bootstarp drive for mosfet though

 

[Easy peasy]

Wait till you have to pass EMC ...

 

[cupoftea]

Thanks, ayk, theres no intention to switch any faster than we would if we were using say a 150v/ns chip, just that it gives us more margin against those problems which can show up with these kind of bootstrap chips.

Though i suppose the main problem with bootstrap chips is the transient "going below zero volts" of the switching node, so maybe i am putting too much emphasis on the V/ns parameter.?

 

[dick_freebird]

Common mode transient dV/dt really bothers the low side to
high side level shifter which needs to be 110% bulletproof (leaving
10% on the table for bad PCB design and mishap, because that's
the kind of guy I am).

This was the main challenge in my GaN half-bridge design (once
the fastest going, now also the fastest-obsoleted). Nearly ten
years down the road and most people still haven't got an
assembly that can use the part (no longer manufactured). Only GaN
vendor eval boards.

"DC coupled" doesn't have a prayer at high voltages and dV/dt. At
some point the tolerable DC currents won't be able to defeat
Cstray*dV/dt and Bad Things Happen. I had to add a lot of "keep
it right, whilst slewing" stuff on the high side to ride out the
transitions. Couple of patents off that, in the end.