# Meshing Sudden Contraction and Expansion

 Register Blogs Members List Search Today's Posts Mark Forums Read

January 19, 2017, 12:30
#21
New Member

parham
Join Date: Oct 2015
Posts: 24
Rep Power: 10
Thank you Max for your help. I was able to mesh my geometry. Just for curiosity, Is there any way to mesh Face 2 by Quad map instead of Quad pave?, because it can significantly improve my calculation downstream pipe (volume 2). I was only able to mesh Face 1 with Quad map (since it is annulus face).
Attached Images
 schematic.png (27.5 KB, 12 views)

 January 20, 2017, 01:33 #22 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 yes with o-grid __________________ In memory of my friend Hervé: CFD engineer & freerider

January 20, 2017, 12:30
#23
New Member

parham
Join Date: Oct 2015
Posts: 24
Rep Power: 10
Dear Max,

I tried to use O-ring method based on another Thread that you suggested, here is the procedure which, unfortunately, failed:

First, I created a cube inside a circle, I chose four vertexes on the circle's edges, connect corner of cube to four vertexes on circle. I separated circle edges into four edges.now, I have five faces, including cube. Idid the same on the other face wich has a larger diameter at contraction point. When I reapet the procedure, as you suggested before, to create a surface inside upstream volume by edges, an error occurred. I attached to you geometry. What should I do in this case.
Attached Images
 o-ring.jpg (76.9 KB, 14 views)

 January 23, 2017, 01:24 #24 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 alright I understand you cannot create the surfaces parallel to the cylinder axis, right? Did you split the cylinder surface (smallest diameter of the annulus volume)? __________________ In memory of my friend Hervé: CFD engineer & freerider Last edited by -mAx-; January 23, 2017 at 07:48.

 January 23, 2017, 07:42 #25 New Member   parham Join Date: Oct 2015 Posts: 24 Rep Power: 10 Exactly, I cannot create surface from edges parallel to the pipe axis. I also splited small circle into four arc edges. Now in the small circle, I have 5 faces, including one cube.

 January 23, 2017, 07:54 #26 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 As I mentionned, you need to split the cylinder surface with circle's vertices Capture.JPG __________________ In memory of my friend Hervé: CFD engineer & freerider

 January 23, 2017, 08:49 #27 New Member   parham Join Date: Oct 2015 Posts: 24 Rep Power: 10 As I told you, I have separated the circle from vertexes. Still not working.

 January 23, 2017, 08:52 #28 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 how did you create the lines parallel to the axis? __________________ In memory of my friend Hervé: CFD engineer & freerider

 January 23, 2017, 10:09 #29 New Member   parham Join Date: Oct 2015 Posts: 24 Rep Power: 10 I created two cubes inside circles. I connected vertexes on first and second circles. Also, each vertex on cube to similar one in the second cube. I simply used create edges from vertexes. But, the problem is that when I want to create surface from edges.

 January 23, 2017, 10:49 #30 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 If you "simple" create Lines like that; then you will not be able to Split the Volume. That s Why I suggested to Split the surface with vertex Sent from my iPhone using CFD Online Forum mobile app __________________ In memory of my friend Hervé: CFD engineer & freerider

January 25, 2017, 16:30
#31
New Member

parham
Join Date: Oct 2015
Posts: 24
Rep Power: 10
Here is the procedure that I did:

As you told, I created four vertexes on circle, then, I have splited face (circle) from four vertexes. So, the result would be a cube inside circle. Eventually, I have 5 faces, see pic 1. I did the same for the second circle (see pic 2).
I have linked each vertex on the first circle to the similar one on the second circle (see pic 3). I tried to create face from edges, but still unsuccessful (see pic 4). Something else that I should know?
Attached Images
 pic 1.png (21.2 KB, 10 views) pic 2.jpg (34.2 KB, 10 views) pic 3.jpg (40.9 KB, 9 views) pic 4.jpg (60.0 KB, 7 views)

 January 26, 2017, 01:28 #32 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 No, your splits are wrong. What you proposed at #23 was good. You only didn't get my point regarding the splits on the cylinder surface. For that I post a picture: I split the blue surface with both vertices (red) Capture1.JPG Capture.JPG __________________ In memory of my friend Hervé: CFD engineer & freerider

January 26, 2017, 16:54
#33
New Member

parham
Join Date: Oct 2015
Posts: 24
Rep Power: 10
Excuse me Max, but I do not think you get my problem. I understand that I should split the blue surface from red vertexes in your shared photo. My problem is not to separate two volumes. The big problem is :
When I created four vertexes on small circles in both sides (circles 1 and 2), then created edges 1,2,3,4.
I created surfaces from edge 1, edge 3, edge 5, and edge 9. I created also surface from edges 3, 4, edge 6, and 10 and so on (see picture 1). I separated volume from created surfaces to create two volumes. So, the step you suggested is not needed. Question: How can I make a Quad map in picture 3?
Attached Images
 picture 1.jpg (58.9 KB, 3 views) picture 2.jpg (44.5 KB, 1 views) picture 3.png (73.5 KB, 3 views)

 January 27, 2017, 00:56 #34 Super Moderator     Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41 you cannot because this surface is not meshable with map. You didn't split it as it should be. The 4 corners of your square don't have to lie on the circle. I previously mentionned you that this split was not correct. See your post #23, where you displayed your previous split (and it was good) __________________ In memory of my friend Hervé: CFD engineer & freerider