Mesh in decreasing region between two surfaces

arnbeb89 · February 25, 2015, 11:03

Does anybody have suggestions how to set up suitable dynamic mesh options for a region between two surfaces which moves closer and closer together, for example the region between a closing valve and its seat. It would be desirable to eventually have no space between the surfaces, however, this would result in infinite small cells.

This is for a project involvning simulation of a closing swing check valve.
All suggestions are welcome!

/arnbeb89

Ramnik · December 9, 2015, 12:20

arnbeb89,

I am trying to simulate a similar problem where my valve is closing and thus the cell volumes become very small. Just wanted to see if you found a solution?

Thanks,
Ramnik

Jonas Holdeman · December 10, 2015, 10:27

My first thoughts on this problem would be to reverse the process used in mesh refinement, or maybe consider the restriction phase of multigrid.

mprinkey · December 10, 2015, 12:04

From my grad school days, I remember KIVA used a combination of stretching and layer removal/addition to handle piston and valve motion. I am fuzzy on the details, but, say, a ten-cell deep mesh (each layer of unit height) would be scaled down until it reached 9 units high...9/10s of its original height, then a layer was removed and the layers were stretched back to all unit height. Then repeat. Eventually you end up with only a layer that one cell thick that you can scale down to zero and then delete it entirely.

EDIT. Looks like they set the last layer to SOLID rather than deleting it entirely. See Section F on Page 14 of this:

http://www.lanl.gov/projects/feynman...cs/KIVA-3V.pdf

LVDH · December 10, 2015, 18:02

Just go for the next generation cfd:
https://www.youtube.com/watch?v=qi7rhgZ5A2o

February 25, 2015, 11:03	Mesh in decreasing region between two surfaces	#1
arnbeb89 New Member Christoffer Join Date: Feb 2015 Posts: 2 Rep Power: 0	Does anybody have suggestions how to set up suitable dynamic mesh options for a region between two surfaces which moves closer and closer together, for example the region between a closing valve and its seat. It would be desirable to eventually have no space between the surfaces, however, this would result in infinite small cells. This is for a project involvning simulation of a closing swing check valve. All suggestions are welcome! /arnbeb89

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Gambit problems	Althea	FLUENT	22	January 4, 2017 04:19
[Other] Material properties in multi region foam and refining mesh for each region	mukut	OpenFOAM Meshing & Mesh Conversion	2	October 7, 2013 03:08
[ICEM] Generating Mesh for STL Car in Windtunnel Simulation	tommymoose	ANSYS Meshing & Geometry	48	April 15, 2013 05:24
[Commercial meshers] Using starToFoam	clo	OpenFOAM Meshing & Mesh Conversion	33	September 26, 2012 05:04
[Other] StarToFoam error	Kart	OpenFOAM Meshing & Mesh Conversion	1	February 4, 2010 05:38

December 9, 2015, 12:20		#2
Ramnik New Member Ramnik Singh Join Date: Jun 2009 Location: Stuttgart, Germany Posts: 18 Rep Power: 17	arnbeb89, I am trying to simulate a similar problem where my valve is closing and thus the cell volumes become very small. Just wanted to see if you found a solution? Thanks, Ramnik

December 10, 2015, 10:27		#3
Jonas Holdeman Senior Member Jonas T. Holdeman, Jr. Join Date: Mar 2009 Location: Knoxville, Tennessee Posts: 128 Rep Power: 18	My first thoughts on this problem would be to reverse the process used in mesh refinement, or maybe consider the restriction phase of multigrid.

December 10, 2015, 12:04		#4
mprinkey Senior Member Michael Prinkey Join Date: Mar 2009 Location: Pittsburgh PA Posts: 363 Rep Power: 25	From my grad school days, I remember KIVA used a combination of stretching and layer removal/addition to handle piston and valve motion. I am fuzzy on the details, but, say, a ten-cell deep mesh (each layer of unit height) would be scaled down until it reached 9 units high...9/10s of its original height, then a layer was removed and the layers were stretched back to all unit height. Then repeat. Eventually you end up with only a layer that one cell thick that you can scale down to zero and then delete it entirely. EDIT. Looks like they set the last layer to SOLID rather than deleting it entirely. See Section F on Page 14 of this: http://www.lanl.gov/projects/feynman...cs/KIVA-3V.pdf

December 10, 2015, 18:02		#5
LVDH Member Andre Z Join Date: Dec 2009 Posts: 75 Rep Power: 16	Just go for the next generation cfd: https://www.youtube.com/watch?v=qi7rhgZ5A2o