Questions about junction temperature

[chocolate3018]

hello, I have question about junction temperature.

I thermal simulated the chip (CPM2-1700-0045B, CREE) in LTspice.

However, I do not understand because it is very different from the chip (C3M0045065D, CREE) result.

The result of the chip (C3M0045065D) is that the heat gradually increases and then enters a steady-state state.

However, in the case of chip (CPM2-1700-0045B), heat is generated only when switching is on/off, and heat disappears again in conduction state.

In my opinion, it is correct to gradually increase and then enter a steady state, like the result of chip (C3M0045065D).

Finally, I have two questions.

First of all, did the chip (CPM2-1700-0045B) result come out properly? If yes, what is the junction temperature of chip (CPM2-1700-0045B)..?

Second, if the result of chip (CPM2-1700-0045B) is not correct, where should I fix it?

Please let me know. Please....

thanks!!

 

[Diver300]

Both types of MOSFET have the same nominal Rds(on) of 45 mOhms. With 100 V supply, and a 10 Ohm load, there will be 10 A, so there will be about 4.5 W of heat generated in the MOSFETs. If the duty cycle is 50%, the average heat generated due to the Rds(on) will be 2.25 W.

That is the easy part. The heat generated during switching depends on how fast the MOSFET is switching. With at 10 Ohm load and 100 V supply, the peak power will be 250 W, and how long that continues for makes a huge difference to the average heat generated by the MOSFETs.

If the MOSFET is not turned fully off, or is not turned fully on, then the power generated when off or on will be much more as well.

The temperature of the MOSFET depends on the heat generated by it, the ambient temperature, and the thermal resistance of whatever heatsink is used. On the chip MOSFET (CPM2-1700-0045B), the thermal resistance isn't quoted, because the chip could be bonded to anything. For the TO-247-3 MOSFET (C3M0045065D) the thermal resistance to the case is 0.85 °C/W but with no heat sink the thermal resistance is around 50 times larger.

This means that the heat sink characteristics and how the MOSFETs are switched are generally more important than the RDS(on) as far as the temperature of the MOSFET is concerned.

I don't quite understand how the temperatures are being modeled. It seems that R2 and R4 in your diagrams represent the thermal resistances, but I don't know if the modelling software would also model electric current as well as heat through those.

 

[chocolate3018]

Thank you for the reply