Matching Interfaces

Fiedde1887 · April 14, 2015, 09:38

Hello,
i want to simulate a ship-propeller in open-water conditions. I try to realize this with RBM and a tetrahedral/polyhedral mesh. After discretizing the moving region (fine mesh with the propeller) and a coarse mesh which is fixed, i want to realize an interface between these reagions with a matchin mesh.
I don´t want to unse the region-based mashing...so i want to ask you, if anyone know a solution for my problem with the parts-based mesher.
I´m using Star-CCM+ 10.02.
Thank you very much in advance for your answer!

Fiedde1887 · April 14, 2015, 11:27

I´ve already created an interface between this two parts and reagions. The interface contains an type: internal interface and the topology: in-place.
The geometry option contains an interface/interface contact which is created by the outer surface of the rotaiting part and the inner surface of the static part of my solution.
I don´t get it. The mesh of the static-region and the mesh of the rotationg region don´t match at the interface.

fluid23 · April 14, 2015, 11:55

Are you married to this approach? There are better ways to do this.

I would recommend using poly cells with overset mesh.

Fiedde1887 · April 15, 2015, 08:42

I´m new with Star. I use it for writing my Master-Thesis. Is there no possibility to realize matching interfaces with this set-up? I want to compare this set-up with an other set-up with polyhedral mesh. After that i want to try the overset-grid-method.
Thank you for the quick answer.
Greetings
Nils

fluid23 · April 15, 2015, 09:20

I am really not sure what to tell you. I never use tetmesh or prescribe rotation such as you have done.

It may be that for this setup star-ccm+ doesn't need a confromal inteface. It would have to have a different mesh for each position in the rotation. There could be a non-conformal interface map being used to translate between regions.

Have you tried to run this yet?

Fiedde1887 · April 15, 2015, 09:49

I tried this set-up and it works. I hoped to increase the accuracy with a conformal mesh. The matching interface during the rotation can be realized with a defined number and width of the layers. If i have the number i can set the delta t. That was my intention to get a better mesh and result.
But i will try the overset-mesh tomorow.
Thank you

ggulgulia · April 16, 2015, 06:51

Hi

As MBond suggested, I too will recommend you to use polyhedral mesh. It certainly is far better than tet-mesh. (google polyhedral mesh if you do not believe its advantages)

As for interface creation, I think it is a matter of having similar boundaries in two regions, select them both and create inplace interface.

ggulgulia · April 16, 2015, 06:54

also if this is a simple rotation problem, I would recommend simple rigid body rotation since it will have significantly lesser cell count as compared to overset mesh. Overset mesh are better where there's more than one degree of freedom.

April 14, 2015, 09:38	Matching Interfaces	#1
Fiedde1887 Member Nils Hennig Join Date: Apr 2015 Posts: 44 Rep Power: 11	Hello, i want to simulate a ship-propeller in open-water conditions. I try to realize this with RBM and a tetrahedral/polyhedral mesh. After discretizing the moving region (fine mesh with the propeller) and a coarse mesh which is fixed, i want to realize an interface between these reagions with a matchin mesh. I don´t want to unse the region-based mashing...so i want to ask you, if anyone know a solution for my problem with the parts-based mesher. I´m using Star-CCM+ 10.02. Thank you very much in advance for your answer!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Discontinuity across interfaces	Nurzhan	CFX	2	March 11, 2015 01:59
What type is defined as the interfaces between stator and stator on OF-2.1.1?	renyun0511	OpenFOAM Running, Solving & CFD	3	May 18, 2013 08:11
[snappyHexMesh] jagged, ragged edges...	ziemowitzima	OpenFOAM Meshing & Mesh Conversion	138	July 23, 2012 23:41
Subdomain or Interfaces (CHT)	sandeep_tu	CFX	8	July 14, 2009 11:06
grid interfaces	kiko	FLUENT	0	February 13, 2007 10:28

April 14, 2015, 11:55		#3
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 17	Are you married to this approach? There are better ways to do this. I would recommend using poly cells with overset mesh.

April 15, 2015, 08:42		#4
Fiedde1887 Member Nils Hennig Join Date: Apr 2015 Posts: 44 Rep Power: 11	I´m new with Star. I use it for writing my Master-Thesis. Is there no possibility to realize matching interfaces with this set-up? I want to compare this set-up with an other set-up with polyhedral mesh. After that i want to try the overset-grid-method. Thank you for the quick answer. Greetings Nils

April 15, 2015, 09:20		#5
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 17	I am really not sure what to tell you. I never use tetmesh or prescribe rotation such as you have done. It may be that for this setup star-ccm+ doesn't need a confromal inteface. It would have to have a different mesh for each position in the rotation. There could be a non-conformal interface map being used to translate between regions. Have you tried to run this yet?

April 15, 2015, 09:49		#6
Fiedde1887 Member Nils Hennig Join Date: Apr 2015 Posts: 44 Rep Power: 11	I tried this set-up and it works. I hoped to increase the accuracy with a conformal mesh. The matching interface during the rotation can be realized with a defined number and width of the layers. If i have the number i can set the delta t. That was my intention to get a better mesh and result. But i will try the overset-mesh tomorow. Thank you

April 16, 2015, 06:51		#7
ggulgulia Senior Member Gajendra Gulgulia Join Date: Apr 2013 Location: Munich Posts: 144 Rep Power: 13	Hi As MBond suggested, I too will recommend you to use polyhedral mesh. It certainly is far better than tet-mesh. (google polyhedral mesh if you do not believe its advantages) As for interface creation, I think it is a matter of having similar boundaries in two regions, select them both and create inplace interface.

April 16, 2015, 06:54		#8
ggulgulia Senior Member Gajendra Gulgulia Join Date: Apr 2013 Location: Munich Posts: 144 Rep Power: 13	also if this is a simple rotation problem, I would recommend simple rigid body rotation since it will have significantly lesser cell count as compared to overset mesh. Overset mesh are better where there's more than one degree of freedom.