There was a bit of heat in the Extreme Devices session this morning during which Y. Wu from Transphorm Inc. defended himself during the Q&A when asked why he presented a power stage that had four parallel GaN FETs. The person asking the question was basically implying that Y. Wu should have done an interleaved system instead of four parallel GaN FETs. Y Wu could simply have responded that there are not always multiphase controllers available for every application space or, simply, that this is the topic of the paper.
This morning’s best paper on ZCS fundamentals was from Dan Constinett from the University of Colorado Boulder. It was excellent because many people think it is easy to pick up the efficiency of ZVS (Zero Voltage Switching). But you have to worry about balancing the magnetics, so the currents do not run away and stress devices. Or, they could run with worse efficiency than simpler methods.
I am a bit surprised that there is a new 3/5 material vendor in the fray. Sarda Technologies of North Carolina announced a <20V GaAs (not GaN ), low voltage power FET with a good FOM (Figure of Merit) of 11. Their devices are a bit hard to drive because the gate leakage is 1 mA to 5 mA. But since their 14 mOhms device has a gate charge of 0.76 nC, it can run at 2 MHz.
To read more of my blog, “Jeff does APEC”, visit TI’s Power House blog