Learning Forum

[Sidharth Mahesh]

Hi all,

I have modelled a conduction problem in Fluent of a Stator along with its conductors. I have split the conductor into 2, the active part along the stack length as well as the end winding. I have given cyclic orthotropic thermal conductivity property to it.

Unfortunately when I run the simulation, I'm getting erroneous results. There is a sudden jump in the steady state temperature result over iterations making it fluctuate from -ve to a high positive value with the following errors,

"temperature limited to 1 in fluent"

"temperature limited to 1.0000e+00001 in cells"

But when I model the end winding as constant thermal conductivity, the solution is converging and I'm getting a proper result.

The mesh is with the best quality: Min Orthogonality =0.11 & Max Skewness = 0.85

Can you guys help in finding a solution?

[Rob]

I think you set the ortho part in the cell zone, are the axis settings correct? 

[Sidharth Mahesh]

I had created a new material with thermal conductivity as cyclic orthotropic. The axis is created wrt global axis of the assembly.

 

 

[Rob]

You may need to review CCS, https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v232/en/flu_ug/flu_ug_sec_hxfer.html%23section_srh_1wv_hsb  Each end winding would be a separate material definition as you set it at present. I checked with one of the testers. 

[Sidharth Mahesh]

When I increase the value of thermal conductivity from 0.8 (like 5-10 W/mk), Iam able to get a solution without fluctuations. I'm not understanding why ?!

I rechecked the assigned CCS. I found it correct.

[Rob]

Maybe the coil is conducting enough to avoid any hot spots forming? The hot spots then stress the solver/result and then cause problems. 

[Sidharth Mahesh]

I tried reducing the heat load, still the result appears to be the same. The end winding temperature plot is with spikes which is later disturbing the active winding plot.

Also, i observed that the energy residual plots are increasing over iterations

Kindly help me reach a solution.

I also tried the option orthotropic, with conductivities along 00,11 and 22. Its results are also the same.

[Rob]

It wasn't clear, did CCS solve the problem?  How well resolved are the flow gaps in the system? Low conductivity will give hot spots, which will make the solver struggle if they're nonphysical or not well resolved which may explain the convergence.