Hello,
I am trying to create a feedback loop between the lookup table of my MOSFET’s conduction losses and the electrical on-resistance Ron. We know that Ron = Von / Ion (linear approx.) and that both of these variables are determined by your circuit and the initial Ron you provide. However, when your MOSFET starts heating up and the junction temperature increases, Ron should change depending on the interpolation in the lookup table (if you provided a second linear correlation at an elevated temperature). This change in Ron will affect the current in the electrical circuit and consequently the thermal losses. This loop should continue untill the steady state situation has been acquired for this setpoint.
The same would apply to a diode’s voltage drop Vf. This parameter is also temperature dependent and will affect the current in the electrical circuit and thus the thermal losses. I feel that without this feedback loop, simulations with variable input profiles will be very inaccurate due to not implementing this temperature dependency of certain parameters.
Is there any way to implement this in PLECS or should I try different software?
Thanks in advance!
Hello,
Thank you for your great question! PLECS does not automatically feed back the thermal losses into the electrical domain at this time. It is possible to implement a temperature-dependent Ron for a MOSFET in PLECS, however, by using the Variable Resistor component in series. Note that a variable resistance inherently creates an algebraic loop, but as of a recent version of PLECS, it is possible to solve this situation.
Attached is an example model of a buck converter where a SiC MOSFET (E3M0280090D) from Wolfspeed is used. The datasheet and PLECS thermal description file are found here: https://www.wolfspeed.com/power/products/sic-mosfets/e3m0280090d. The datasheet has, for example, two plots (Figure 4. and Figure 5.) that can be used to develop a characteristic for the Junction Temperature vs. Ron behavior. Using this information, a basic dataset of Ron values at several operating temperatures and using a fixed Ids current of ~7A, has been captured using a lookup table in the included model.
Hopefully this example demonstrates the concept and can be used as a reference for you to create a higher resolution temperature dependency for your application. Please let us know if you have any questions.
Kind regards,
Kris
TemperatureDependentRdsOn.zip (8.82 KB)