LMG2652 GaN Half bridge issues?

[cupoftea]

Hi,
Just found the LMG2652 GaN Half Bridge Chip with 650V GaN FETs on it.

LMG2652

https://www.ti.com/lit/ds/symlink/lmg2652.pdf?ts=1740083407546

Was thinking here is the answer to the LLC and PSFB problems of reverse recovery.
Also, it would never suffer from spurious turn on of the bridge leg FET when the "other" FET is turned ON.......then i get to page 17 and it says....

QUOTE >>> The LMG2652 GaN power FETs are intended to be turned on in either zero-voltage switching (ZVS) or discontinuous-conduction mode (DCM) switching conditions <<<UNQUOTE

So why can it not handle turn-ON into eg CCM hard switched half bridge waveforms?
And why only stipulate about "turn-ON"?....in for example a PSFB, or LLC with low Lmag and operated way off the resonant frequency, the turn_off transient can be very hard switching....so why does it not mention that aswell?

 

[dick_freebird]

Because somebody sucks at their job, or was told that their job includes obfuscation?

 

[Easy peasy]

if you run it in a hard switched setup = the RFI will dominate the control - most peoples layout and decoupling is simply not up to the task.

 

[D.A.(Tony)Stewart]

Gallium Nitride Design Examples | GaN First Time Right | EPC

Get your gallium nitride design just right from the start with EPC's gallium nitride design examples.

epc-co.com

 

[mtwieg]

That is indeed a very bizarre admission, especially given that TI already makes similar devices which don't mention this restriction. Wouldn't be surprised if many designers never notice it buried in the datasheet like that.