n-Channel Mosfet 3-D Lookup table, temperature dependent parameterization. Problems simulating different temperatures

Hello, I am trying to simulate the behavior of a MOSFET at different temperatures (25ºC and 17ºC), but I am having some issues. I have chosen the n-Channel MOSFET model from the Simulink library and below I attach the circuit and the parameters I have used based on the curves provided by the manufacturer.

The parameterization used is a 3-D Lookup table, temperature dependent, and in the following image, you can see the matrix used for IDs (datos_Ids) which depends on Vgs, Vds, and T.

When the simulation is run, even if the value in Temperature Source is changed, the same simulation always results; it seems to be independent of the temperature.

Any suggestions?

Thank you very much

Answer 1

[Shivam Gothi]

Hello @Ignacio,

As per my understanding, you are providing the junction temperature of MOSFET by using an external temperature source. By changing the junction temperature, there is no change seen in the voltage across drain-source terminals and the current flowing through it.

I tried to reproduce the issue by constructing the similar circuit in Simulink. (refer to the attached Simulink model). As the values of resistors were unknown, I selected the resistors value at my own. When I simulated the model, it was found to be working satisfactorily. That is, I was able to observe the effect of changing the junction temperature on the device characteristics. I took the reading of drain-source voltage (Vds) and drain current (Ids) at different temperatures, when the switch was “ON”. All the other circuit parameters were unchanged. The results are tabulated below:

Based on my understanding, the possible mistakes that you might be doing in simulation are:

• Based on the 3d look-up table created by you, it seems that the MOSFET is rated for higher current. If the current levels in your circuit are too low, then it become difficult to distinguish the effect of temperature change. Therefore, select a much lower value of resistor placed in series with the drain-source terminals of the device. I my case, I had chosen resistance of 5mΩ and “DC voltage source” value as 5v.

(NOTE: I have defined "datas_Ids" 3D-matrix as attached by you, in "InitFcn" section found in "model parameters" of the simulink model. Therefore, it is initiallised automatically when the simulink model is opened. You can change/edit it as per your requirnment).

I hope it helps !

[Ignacio]

Hi @Shivam Gothi.

Indeed, it is a power MOSFET, and I want to study the switching transients under different load conditions and temperatures. The bus voltage is 540V and a current of 200A (hence the value of R is 2.7 Ohms) due to the load, which can also be inductive (1uH or 10uH). These are the values in the schematic.

Looking at the output characteristics at 25 and 175ºC, it may be that for Ids = 200A, there is not much variation in Vds for Vgs between 18 and 20V.

The curious thing is that if I introduce the Lookup table (2D, temperature independent) parameterization into the Mosfet model to simulate first at 25ºC and then at 175ºC, different simulations are obtained

25ºC Parameters

175ºc Parameters

But if the parameterization used in the model is Lookup table (3-D, temperature dependent), where the parameters for the temperatures of 25 and 175ºC are introduced, the simulation results are the same for both temperatures (regardless of the value entered in the temperature source).

Additionally, the simulation result, comparing the simulation at 175ºC done with the Lookup table (2D, temperature independent) option, is the same.

Sorry for the lengthy explanation ;-)

Best regards,

Ignacio

Answer 2

[Ignacio]

Hi @Shivam Gothi.

Indeed, it is a power MOSFET, and I want to study the switching transients under different load conditions and temperatures. The bus voltage is 540V and a current of 200A (hence the value of R is 2.7 Ohms) due to the load, which can also be inductive (1uH or 10uH). These are the values in the schematic.

Answer 3

[Ignacio]

Looking at the output characteristics at 25 and 175ºC, it may be that for Ids = 200A, there is not much variation in Vds for Vgs between 18 and 20V.

The curious thing is that if I introduce the Lookup table (2D, temperature independent) parameterization into the Mosfet model to simulate first at 25ºC and then at 175ºC, different simulations are obtained

25ºC Parameters

175ºC Parameters

But if the parameterization used in the model is Lookup table (3-D, temperature dependent), where the parameters for the temperatures of 25 and 175ºC are introduced, the simulation results are the same for both temperatures (regardless of the value entered in the temperature source).

Additionally, the simulation result, comparing the simulation at 175ºC done with the Lookup table (2D, temperature independent) option, is the same.

Sorry for the lengthy explanation ;-)

Best regards,

Ignacio