solving algebraic loop due to parasitic Gate Source Capacitance in 3-LVL-ANPC under different modulation strategies

Question

Hello,

I would like to investigate different modulation strategies for a 3-LVL-ANPC-SiC topology under different operating points in terms of overall efficiency and power dissipation (temperature) per switch.
 

To additionally investigate any overvoltages occurring at the switches, I would like to implement the parasitic drain source capacitances (C_DS = C_oss - C_rss). (Coss, Crss from the data sheet specifications)
I had already asked a question here regarding the occurrence and non-solving of algebraic loops at some operating points.

https://forum.plexim.com/8413/algebraic-variable-resistor-different-modulation-strategies

If I now install a drain source capacitance in parallel to the body diode as shown in the attached picture, I always get the following error message, regardless of the operating point under consideration:
Could not solve the algebraic loop comprising the following components:
1: 3_LVL_ANPC_SiC_VSI_Ron_Erec/S1a/MOSFET
2: 3_LVL_ANPC_SiC_VSI_Ron_Erec/S1a/2D-Table1
3: 3_LVL_ANPC_SiC_VSI_Ron_Erec/S1a/R/Product
4: 3_LVL_ANPC_SiC_VSI_Ron_Erec/S1a/R/Subtract

This leads to an immediate termination of the simulation, even at the operating points that previously worked without the additional capacity.

Hence my question: Is there a way to add an additional drain source capacitance for each switch without aborting the simulation due to not solving an algebraic loop?
I would really appreciate an answer. I have attached pictures to illustrate the problem.

Kind regards 
THK

Answer 1

I would not recommend running thermal models with a parallel switch capacitance modeling the Coss.  When the switch is closed the Coss will discharge into the MOSFET channel leading to a large current spike within the switch.  The drain-source voltage will also not change instantaneously.  This leads to an incorrect V & I entries into the thermal look-up tables, which are generally extracted under a different set of assumptions.

If one wants to model the Coss in the manner you're using, I would recommend using a Configurable Subsystem - one with a thermal model and the other with a more complicated electrical model.

The recommendations made in your other post about resolving algebraic loops apply here as well.