Learning Forum

[lgreen11]

Hello, my school group and I are attempting to use ANSYS Icepak AEDT to test a water cooling heat sink system inside a chamber. Attached is the photo of the system. Our boundary conditions are...

- Heat Source: 350W located directly on the heat plate.
- Recirculating Opening: Using the two wholes seen in the pictures at 1L/min
- Opening: Openings in the air system around the chamber.

We use the volume extract tool to get the interior geometry of the chamber and assign a water material to it. All the other geometry items in the chamber have been assigned a material.

The issue we keep running into is when we attempt to run the solution, we will constantly get a "Solver Failed because poor mesh quality, try increasing mesh quality". Our mesh is currently at 18 million elements. Whenever we try increasing the mesh quality HDM service crashes. We are stuck trying to figure out what to do. For context we aren't seeing any other populated errors in the Message Manager, and our school has purchased the full license for Icepack AEDT, so we are not on the student version.

Does anyone know if we might have something wrong in our model that is causing the solver not to run?

Thanks!

[Iceman]

Hi,

Icepak supports multile fluids, but all the fluids need to be separated. Please attach a small hollow cylinder (solve inside unchecked) to both the inlet and outlet of the water volume. This way, water can be separated from air. Thanks.

[lgreen11]

 

Hello, thank you for your response, I want to make sure i’m understanding this clearly. So I need to attach a cydlinder from each of the openings, then extend that cylinder out passed the boxed air regoin around the housing itself? This will allow them to be seperated?

 

[Iceman]

One end of the cylinder should overlap with the water opening. There is no need to extend the cylinder outside of the air region. The other end of the cylinder can be inside of the air region. The cylinder created should have solve inside unchecked. This way the interior of the cylinder won't be meshed and will be considered as the external of the computational domain. 

[lgreen11]

Hello, sorry for all the questions, here is what I have made on each side, is this the setup you meant?, I

[lgreen11]

We attempted running again with the same results with these cylinders in place, attached are more photos of the simulation.

[Iceman]

 

No worries. Do you have Solve Inside unchecked for these two cylinders? Also, can you change the recirculation openings to Velocity Inlet and Pressure Outlet? 

If the issue persists, please try to test a simplied version of your model. You can try to make Air region Non-model, only test the Water system. And then you can try to make the water volume Non-model to test the air side. 

The water side looks pretty complex, please make sure to use best facet quality if there are many curved surface. Please also check the mesh quality using Mesh viewer. 

 

Have a great holiday. 

 

[lgreen11]

I will give these options a shot, thank you very much!