Sim of custom waterblock

[bikerbub]

I've run into a good few issues getting my sim to the point that it's at now, and I think I've gotten past most of the Win8.1/installation related issues. I've moved all of the project files over to my C:/ drive; they were originally on an internal storage volume. I'm new to CFM, but I've got enough experience with Inventor to be comfortable with the interface. I've spent a few days' free time reading through the wikihelp trying to get this sim to work, and I keep getting the "The Analysis has stopped because the Solver has exited unexpectedly" error. 

The goal of the design is to keep two power-mosfets cool. The Mosfets are represented in the sim by the copper pieces that are flush-constrained to the face opposite the inlet/outlet ports. They are through-hole mounted in the circuit design, so the Compact Thermal Model approach wouldn't quite work, but I don't know that my current approach is that much better. 

If I supress all parts of the imported model except the water, I can solve for flow and heat transfer, though the heat transfer information is useless at that point. If I add any of the parts back in to the simulation, it solves as far as "Meshing..." then quits with the aforementioned error message.

Thanks for your help
Josh

[Jon.Wilde · Accepted solution]

I made a few changes:

1. Remove the internal radiation boundary conditions (this gets it running)
2. Add a temp to the inlet 
3. Use Gap Refinement with 3 gap elements
4. 10 layer of mesh enhancement at 1.2 
5. Use SST turbulence model (hence the mesh enhancement change)

The mesh on the solids is pretty coarse, often only 1 element through the thickness, consider refinement.

Thanks,

Jon

[bikerbub]

Thanks Jon,

At its current rate, the sim will be done around 2am; hopefully it's worth it.

I am curious as to the purpose of the radiation boundary constraint though. Is it just an easier way of redefining emittivity as opposed to editing the material?

Josh