Hello,I want to model thermal coupling between all IGBTs and diodes in a 3-phase inverter, similarly to what is answered to this question (PLECS User Forum). So, if I’m right, a unique heat sink cannot include additional thermal impedances between components: a heat sink for each component should be used instead.
Differently from the previous provided answer, I would want to know if heat sinks’ temperatures are simulated as voltages, so they may influence each other and extra parallel thermal circuits with controlled thermal sources are not needed. In the attached screenshot, that implies power losses on the IGBT will rise the temperature on the diode’s heat sink and also affect the junction temperature at the diode (maybe these are two separate issues?).
EDIT reading user manual: “Heat conduction from one heat sink to another or to ambient temperature is modeled with lumped thermal resistances and capacitances…”. Heat sinks are indeed affected by each other: do they affect junction temperatures through Zth j-c?.
Additionally, thermal impedance between junction and case, modelled as Foster networks, have no physical meaning when it comes to new thermal impedances to be added like Zth heat sink - ambient (ref. Intro to thermal domain tutorial’s reference (M. Maerz, P. Nance, Thermal Modeling of Power Electronic Systems, p8)). But as PLECS converts all Foster networks to Cauer at the beginning, is it right to add Zth case - heat sink and Zth heat sink - ambient separately to Zth j-c while maintaing manufacturer’s Zth j-c?
Thanks in advance.