[ICEM] Help needed with blocking 3D lorry model in tunnel!

tak_li · November 19, 2018, 13:50

Hello,
I'm currently trying to block a 3D lorry model to be able to mesh it.
It is an imported CAD model in ICEM.
I know how to block, associate, mesh 2D models but am struggling with 3D.
Please let me know if you know how to thanks.

The model is in a very big region compared to its body size.
I've attached below a photo of the model and the model in the region (ignore the attempted block lines)

Thanks!

Gert-Jan · November 20, 2018, 04:52

Why not try unstructured?

tak_li · November 20, 2018, 07:26

Quote:

Originally Posted by Gert-Jan

Why not try unstructured?

Oh ok, I will try that. I still need to do the blocking right? and if so do you know the easiest way to split the region in reference to the curved surface?
Thank you

Gert-Jan · November 20, 2018, 07:39

No. With unstructured, you don't need blocking.
Only if you want hexa's (structured mesh), you need blocking. But with such a geometry, it might take you a week to create the bliocking structure, or a month if you are unexperienced. So I would recommend unstructured anyway.

tak_li · November 20, 2018, 07:47

Quote:

Originally Posted by Gert-Jan

No. With unstructured, you don't need blocking.
Only if you want hexa's (structured mesh), you need blocking. But with such a geometry, it might take you a week to create the bliocking structure, or a month if you are unexperienced. So I would recommend unstructured anyway.

Oh I see thank you, I will get on it!

bluebase · November 21, 2018, 06:23

Hi Tak,
i agree with Gert-Jan, fully structured blocking might take you a while.

Instead of meshing fully unstructured, i would use a hybrid approach.
Create a resolved unstructred mesh around the vehicle in a bounding box.
However, the windtunnel except the vehicle box could be meshed structuredly.

This approach has of course disadvantages as well as advantages.
The major disadvantage is, that it is not trivial to match the interface between unstructured and structured. However, there is a tutorial coming with icem explaining necessary steps. It might take some more work to also match element sizes.
A clear advantage, is that the structured part of the domain would need much fewer elements to fill it. The elements would also be aligned with the flow (better convergence, and higher accuracy). Generating the mesh after blocking is finished will also be much faster.

Also check out Simon's "Tips and Tricks" guide on page 45, searching "hexa tunnel".
You'll find the guide in this post: Before you start a thread, see if you can find answers here

Best regards,
Sebastian

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Use of k-epsilon and k-omega Models	Jade M	Main CFD Forum	40	January 27, 2023 08:18
Model Help Needed! (Design Modeler)	Evaansss	ANSYS	3	November 4, 2017 03:56
[ICEM] Blocking and Symmetry	BrolY	ANSYS Meshing & Geometry	32	August 24, 2012 04:13
turb model in a wind tunnel	fereshteh007	FLUENT	1	January 28, 2012 07:10
Basic advice on Importing Solid Edge model to CFX needed.	stephen__white	CFX	1	February 9, 2011 19:40

November 20, 2018, 04:52		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	Why not try unstructured?

November 20, 2018, 07:39		#4
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	No. With unstructured, you don't need blocking. Only if you want hexa's (structured mesh), you need blocking. But with such a geometry, it might take you a week to create the bliocking structure, or a month if you are unexperienced. So I would recommend unstructured anyway.

November 21, 2018, 06:23		#6
bluebase Senior Member Sebastian Engel Join Date: Jun 2011 Location: Germany Posts: 567 Rep Power: 21	Hi Tak, i agree with Gert-Jan, fully structured blocking might take you a while. Instead of meshing fully unstructured, i would use a hybrid approach. Create a resolved unstructred mesh around the vehicle in a bounding box. However, the windtunnel except the vehicle box could be meshed structuredly. This approach has of course disadvantages as well as advantages. The major disadvantage is, that it is not trivial to match the interface between unstructured and structured. However, there is a tutorial coming with icem explaining necessary steps. It might take some more work to also match element sizes. A clear advantage, is that the structured part of the domain would need much fewer elements to fill it. The elements would also be aligned with the flow (better convergence, and higher accuracy). Generating the mesh after blocking is finished will also be much faster. Also check out Simon's "Tips and Tricks" guide on page 45, searching "hexa tunnel". You'll find the guide in this post: Before you start a thread, see if you can find answers here Best regards, Sebastian