Learning Forum

[d.jemoronm]

Hello 

I am simulating an electric machine with all the thermal properties of the materials such as density, Cp, and thermal conductivity. I am also using a heat source term of 2000000 W/m^3 in the coils. However, I'm experiencing some issues as the temperature distribution reaches 5000 K. I am also receiving messages such as "temperature limited to 5.000000e+03 in 1575181 cells on zone 228 in domain 1."

Can someone help me identify what the issue might be, whether it's related to boundary conditions, initial configuration, or any limitations in the Fluent model?

 

[Rob]

What wall boundary conditions did you set? Heat is added by the source term, but you also need a path for it to leave. The 5000K limiter is for stability reasons, other than some specialist combustion/metal processing cases it's very unlikely you'd reach that value as most materials melt before then. 

[d.jemoronm]

 

 

Thank you very much for your response.

I have now coupled the electromagnetic model with Fluent to export the Maxwell losses and obtain temperature distributions in Fluent. In the thermal example of the Prius engine from your tutorials, a convective boundary condition is applied to the outer frame wall with a heat transfer coefficient of 10 W/m^2 K. In my case, since I have not yet installed the frame, I assigned it to the stator, which improved the model’s results. However, I expected to observe higher temperature values in the coils, but that is not the case. Although the temperature variations in the coils are very small compared to the other elements of the PMSG.

My other question is whether I should add a similar term at the boundary for the shaft, rotor, magnets, and coils.

Additionally, my residual is decreasing but has not yet stabilized.

I have shared the thermal material properties.

                  Density (kg/m^3)             Cp (J/kg*K)            Thermal Conductivity  (W/m*K)

Air-Solid:            1.1                              1006                                       0.242

Cooper:             8890                            0.385                                      386

NdFeB-N35:     7449.8                          460.548                                  6.7409

M22-26G:          7420                            460                                         25

Mesh Quality:

Minimum Orthogonal Quality = 3.26275e-01 cell 578462 on zone 228 (ID: 8276606 on partition: 10) at location ( 2.70118e-02, 4.51158e-02, 1.24323e-18)

Maximum Aspect Ratio = 1.08217e+01 cell 445919 on zone 226 (ID: 4004725 on partition: 5) at location ( 1.57994e-02, 1.02075e-02, -6.55000e-02)

 

 

 

 

 

[Rob]

You'll need a lot more than 10 iterations: even the tutorials don't converge that quickly! 

I can only give guidance on the various boundaries. Basically, you have a heat source into the coils due to the electric bit, possibly other parts due to parasitic losses. Heat must then leave the various components, and a convective boundary is sensible for the solids: note the external HTC is something you need to calculate, it'll generally be higher for moving parts than stationary.