Improving Mesh Quality in ICEM CFD
 

Hello everyone! I am completely new to ICEM CFD and ANSYS in general and this has been the most confusing task for me to accomplish. I managed to get a pre-mesh for my PV/T-PCM collector, however, there are some problematic areas within the mesh that I need to correct to improve quality.

I have no idea what needs to be done to fix these edges. Could anyone please guide me on this? I would really appreciate it. Below I have attached an image of the collector, an image of the problem area, and lastly a close-up of how the mesh looks like in the problem area.

Attachment 81913

Attachment 81914

Attachment 81915

If possible, please explain in a simple way since I really depended on youtube tutorials for the majority of this task and still don't have a good grasp on how this software works. :(

Hello Athena,

How did you mesh around the cylinder, did you use an O-Grid?

It might help to include a photo of your blocking.

Hello! Thank you so much for the reply, yes I did use o-grid. I followed a youtube tutorial to do it for the cylinder which I'll link below.

https://www.youtube.com/watch?v=SuuB...tmikoErlambang

Also, the blocking pictures are attached here.

Attachment 81923

Attachment 81924

I tried increasing the number of elements in the meshing but that didn't seem to make a difference with this problem either. :(

Hello Athena,

That is an interesting blocking strategy. I myself have followed the following tutorial; it uses a different blocking strategy, one that I had no problems with negative volume elements for: (for 3D): https://www.youtube.com/watch?v=l-Tz6MHlT1M
And this is a similar one for 2D: https://www.youtube.com/watch?v=T2Dxa0OZf58

Also, looking at the first set of pictures in your original post... The third picture of the problem area with the yellow box around it looks like an association is off (i.e. it looks green where it should be purple). Is the surface there correctly associated with the purple (looks like PCM_INLET)? It appears to be associated with the PCM_BOTTOM_WALL. However, in any case, ICEM really doesn't do well with sharp corners as in those elements, so it is not surprising you have issues there. It should look more like what you have done for the bottom of the cylinder.

I think I'll probably redo my blocking in that case. Just scared that this wont work out either because I tried using ANSYS meshing as well for this model and I really couldnt get a good orthogonal quality for it and I'm losing hope. :")

But thank you so much for your response!

About the green area that you mentioned should be purple, the thing is the pipe is actually attached to the plate above it (the absorber plate) so that area cant be purple. Do you think theres a problem in the mesh because of the cylinder being attached to the upper surface?

Also, the video that you sent is basically a cylinder going through a wall, but mine's a pipe, so theres another layer for the circular external walls. Not just a hole. Will I have to do something different for it during the blocking?

Hello,

Ah, I understand what that section is for now. Yes, ICEM, and I am sure ANSYS Meshing, would have problems with that green section. Check the skew (Angle) in ICEM of those elements. I am sure it is under the minimum acceptable skew angle of 18 deg. in those sections.

Sorry, yes, there will be a small difference between the video and your geometry, as your pipe has a thickness. It was just to illustrate a different blocking strategy you could use to lower the skew angle in that area. You would likely need to do a double O-Grid with the interior block.