Doubts about MOSFET modelisation, body diode and Wolfspeed's thermal descriptions

Question

Hi everyone.

I have three separate questions but linked each other.

1 - A Mosfet is a bidirectional device and from what I know, this happens because of the presence of an intrinsic diode (body diode) that exists due to the way the device is built. 
Now, as correctly modelled in Plecs, the MOSFET (as a single element of the library) is closed as long as a non-zero gate signal is applied.
That said, when the current is negative (from source to drain) it flows through the body diode (in reality) and this is modelled through the bidirectionality of the MOSFET in Plecs.  Is this correct so far?

2 - I am using Wolfspeed MOSFETs (C2M0025120D) for a DAB application and the same company provides the thermal description of the component. If I analyse it, I find that in the "conduction losses" tab, Von values are defined even for negative currents. Does this mean that those Von values (for negative currents) are those relating to the body diode? From the point of view of the thermal description, explained better, are they describing both the losses for positive currents (through the MOSFET) and the negative ones (through the body diode)?
If the answer is yes, why do they also provide a separate thermal description for the body diode with a final string "_bodydiode"?
If the answer is no: what are they thermally describing with the negative current values?
I checked the values between the two thermal descriptions (MOSFET (C2M0025120D) and Diode (C2M0025120D_bodydiode)) and at this point, for the negative currents of the MOSFET, I expected to find the same values for the positive currents of the diode thermal description but this was not the case.

3 - Going to check the DualActiveBridge demo model, the body diode is simulated with a separate diode but no current ever flows in this one and so there are never any conduction losses related to the body diode thermal description. However, the MOSFET current is actually negative at certain times and that loss is calculated using the MOSFET's thermal characteristic for negative currents. Isn't that strange? Or am I missing something?

Thank you in advance for your reply and sorry for the long thread.
Most probably, these questions are also related to:

https://forum.plexim.com/2291/diode-on-resistance and https://forum.plexim.com/459/distinguish-between-different-components-component-library

 

Answer 1

Hello,

1) Actually, during third quadrant conduction, current can flow through the channel and/or the body diode in a MOSFET (including SiC), depending on the technology and magnitude of the current (typically only when the current is high does Vd exceed Vf such that the diode begins conducting).

2) The Von values for negative currents in the MOSFET thermal description .xml file are for third quadrant operation of the channel only. The body diode forward conduction losses are in the separate diode file.

3) The comparable electrical impedance of the MOSFET and diode will determine the conduction behavior of the two paths. So for the DAB, you should check what the on-resistance and forward voltage values are that are provided and how manipulating these will change how the current flows (just for understanding what is happening).

I hope this helps, but let me know if you have further questions!

Kris

Comments

Thank you for your answer! Now it's more clear!