|
[Sponsors] |
June 27, 2018, 07:27 |
"Surface Wrapper" vs "Extract Volume"
|
#1 |
Senior Member
Join Date: Mar 2016
Location: Bergamo
Posts: 157
Rep Power: 10 |
Hello everyone,
I would like to ask you which is the best technique to extract volume from a imported CAD geometry. Following the pipe tutorial, It is suggested to fill the holes with the Fill Holes operation command to determine the Inlet and Outlet surface. Then it's time for the Imprint command, but I don't understand well what it does and how works. Finally the volume is extracted with Extract Volume. In another case, where the geometry is an assembly and more complex, the tutorial suggests a different approach. It uses the "Surface Wrapper" to fill the holes (why a different command?), and then, after executing it, the volume is extracted. In a third case it uses the "Surface Repair" command What I don't understand is why there are two different approach to do the same thing. When have I to use the first or the second? What are the differences? Thanks for help |
|
June 29, 2018, 07:31 |
|
#2 | |
Senior Member
|
Quote:
|
||
July 12, 2018, 17:48 |
|
#3 | |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 24 |
Quote:
I disagree with this completely. The surface wrapper is a great tool, but in many cases a surface wrapped geometry will result in a longer mesh time with worse quality than taking the time to prep surfaces. That tradeoff needs to be analyzed on a case by case basis. One should not universally use the surface wrapper unless your geometry is so dirty and broken that taking the time to fix it is not worth it. The wrapper may also simplify geometry which has small features, which can be undesirable. Getting around this is possible, but it costs mesh time. To answer the question at hand: Pretend you have a geometry which is just a bunch of CAD parts. The parts are all clean and closed individually, but represent the solids (i.e. the fluid volume requiring meshing is not available). These parts are also unconnected, as in the triangles that make up their surfaces are not joined between parts. In this case, the first procedure is preferable. The order of operations are: 1. Create the faces that represent the inlet and outlet of the fluid volume via fill holes 2. Join the triangles of all of the individual parts using the imprint command, so they are now topologically connected together 3. Use the extract volume operation to give a closed surface which represents the fluid volume. 4. Ready to mesh. Now consider a case which is identical to the first, but some of the parts are not closed volumes. Some of them might be missing faces or have degenerate geometry, or they may not even be touching each other, making small gaps. If the geometry is complex enough, it's probably not worth your time to try and fix all of those errors. In that case, you can wrap the domain. You will lose some mesh quality where the wrapper fills gaps and whatnot, but you will have saved a lot of time. The choice is yours. IMO assess the geometry and make the decision individually. |
||
July 17, 2018, 09:30 |
|
#4 | |
Senior Member
|
Quote:
Yes. I was wrong. Sorry for being an amateur in this. Surface wrapper does changes the geometry so forming an interface between wrapped geometries is not recommended. Just faced the similar problem and volume extraction comes to the rescue there. I was unable to create interfaces between two cad parts but even after checking everything I was unable to do so. But then using volume extraction, I extracted volumes from my main geometry and interfaces are formed automatically. So, learn both. |
||
August 3, 2018, 10:10 |
|
#5 |
Senior Member
Join Date: Mar 2016
Location: Bergamo
Posts: 157
Rep Power: 10 |
I have another question about the behaviour of StarCCM+.
Yesterday I built a simple pipe by extruding a simple circle sketch. I didn't create a solid but a sheet, and the I filled the inlet and outlet surface by "fill surface". Then I use the automate mesh and, with surprise, the mesh was created, but I don't understand why because the pipe is empty inwardly, there is no inner volume. I thought that I should extract the volume before. Can anybody explain this? |
|
August 4, 2018, 12:16 |
|
#6 |
New Member
Owen
Join Date: Aug 2018
Posts: 20
Rep Power: 7 |
For posterity, a new question should probably be in a new post.
Your sheets formed a closed, two-manifold volume in which a mesh was created. Exctract Volume is usually used to Boolean two or more things together in order to create a volume. |
|
August 5, 2018, 05:10 |
|
#7 |
Senior Member
Join Date: Mar 2016
Location: Bergamo
Posts: 157
Rep Power: 10 |
So, if I understand well, for every sheet closed body is the internal fluid volume generated automatically? Because, in case of my pipe, I didn't perform any extract volume operation.
|
|
August 5, 2018, 08:01 |
|
#8 |
New Member
Owen
Join Date: Aug 2018
Posts: 20
Rep Power: 7 |
The output of an ExtractVolume is a close, two-manifold surface that can be used as input to a volume mesher. Your capped cylinder *is already* a closed, two-manifold surface. So when you put it in an AutoMeshOperation, the operation assumes that you want to mesh the volume you gave it.
I hope that clears it up. I encourage you to look at the documentation/tutorials on common uses of the ExtractVolume operation and what is expected of the input for AutoMeshOperation. |
|
|
|