I’m trying to make switching loss model for partial ZVS or ZCS converter. I believe achieving ZVS means they don’t have turn-on loss but how about partial ZVS and ZCS converter?
I already put it each legend, we have gate signal, Vds voltage/current waveform and last one is switching loss.
as we can see there, there is NO turn on loss. even there are still voltage on the Vds, also there is no current on the MOSFET. It means basically operated like partial ZVS or ZCS.
So I was trying to understand of this behavior, because there is no energy model on the Turn-on loss table especially 0A condition(as we can see below loss model). I think that’s why we can’t see the turn-on loss on the scope. This is because there is no current on the MOSFET at the turn on moment, because it operate like ZCS or partial ZVS.
So do we already have solution for this one? Because so far, I’ve check each energy model, which made by MOSFET company also does not have energy loss @0A on their loss plot. Please give me any idea of this, It would be really helpful to make loss model.
The important point here is that the turn-on and turn-off loss look-up tables in a PLECS thermal description are primarily intended to represent hard-switching losses. The data provided by semiconductor manufacturers is also characterized under defined hard-switching test conditions.
PLECS does not model the actual electrical switching transition when using an ideal semiconductor together with a thermal description. Instead, it evaluates the switching event based on the voltage and current directly around the switching instant and obtains the corresponding switching energy from the loss model.
For this reason, ZVS and ZCS should normally be represented with zero switching loss in such a thermal description. In particular, the loss model should contain 0 J at 0 V for ZVS and at 0 A for ZCS to avoid unwanted interpolation or extrapolation. We have discussed a similar case here:
This also explains the result you are seeing. At the turn-on event, the MOSFET current is 0 A. Therefore, from the perspective of the thermal loss model, this is a ZCS event and the turn-on loss is evaluated as zero, even though Vds has not yet reached zero.
For partial ZVS or partial ZCS, the situation is more complicated. The remaining losses depend on the actual switching transient, including effects such as the device output capacitance, parasitic elements and the detailed commutation process. These effects cannot generally be derived accurately from the manufacturer’s hard-switching Eon look-up tables.
If such a detailed investigation of the switching transition is required, the suitable approach at the moment would be a PLECS Spice simulation with an appropriate semiconductor SPICE model. This allows the electrical switching transient itself to be resolved rather than representing the complete event by a single switching-energy value.
