Switching losses of a MOSFET in a Inverter-Bridge

Currently I am simulating a Inverter for an PMSM in Simulink/Simscape and my nex task is to determine the overall losses in the circuit. Of course the MOSFETs will have a major influence, so thats where I want to start. But before, I would like to make sure, that my considerations are correct.

Unfortunately I have to use a small step width, so my simulation is already a little bit slow. So instead of using the N-Channel MOSFET Simscape Block with datasheet values I expect to save some calculation effort with the MOSFET (Ideal, Switching) Block.

But how can I determine the switch-on and off losses for the block? My approach would be to pick 2 or 3 interesting operating points of the inverter, take a snapshot of the current voltages and currents at each MOSFET and rebuild it in a separate model but this time with a N-Channel MOSFET. Within that model I can once switch on and off the MOSFET and calculate the losses. Would that work?

I also noticed that since Version 2023b I can tick "Use last on/off-state current/voltage from previous cycle for turn-on/off losses" in the Ideal MOSFET Block. Is that a good substitution for the procedure above?

Once I got the switching losses I can use ee_getPowerLossSummary and for passive components in the circuit I'll find a way.

But I am also not sure if my understanding of Ideal Switching of the FET is correct. Does it just mean, that parasitic effects and resulting oscillations during operation are not considered?

Many thanks in advance and best regards

Kai

Answer 1

[Abhishek Kumar Singh]

Hello Kai,

I understand that you want to calculate switching losses for the Ideal N-channel MOSFET for switching applications.

The specified block introduces switching losses by incrementing the junction temperature with a value equivalent to the switching loss divided by the total thermal mass at the junction.

It is imperative to define the energy dissipated during a single switch-on and switch-off event. Additionally, you need to provide the corresponding values of the off-state voltage and on-state current at which the losses are determined. The customization of switching losses is feasible, depending on the available data.

Utilizing "Use last on/off-state current/voltage from the previous cycle for turn-on/off losses" may not be advantageous, as the documentation case suggests its specificity to modeling capacitance across the switching device or modeling switching device lead inductance, neither of which aligns with the current scenario.

For more information, you can refer to the documentation section explaining the thermal losses for the mentioned device here.

Parasitics do contribute to the switching losses. But there are other factors as well like frequency and conduction losses. For a comprehensive understanding you can refer to this third party journal (discretion required).

I hope it helps resolve your issue!