|
[Sponsors] |
January 30, 2010, 03:27 |
meshing a long pipe
|
#1 |
Senior Member
|
hi i'm going to mesh a long pipe. it's 180 meter with 1.5 meter diameter. this pipe has 90 degree curve after 30 m from inlet whit curve radius of 15 meter. i want to generate a prism mesh with triangular base that every cell has 1 meter length along the pipe. how can i do that.
look forward to your answer. kind regards. |
|
January 30, 2010, 08:50 |
|
#2 |
Senior Member
|
nobody knows?
|
|
January 30, 2010, 13:32 |
|
#3 |
New Member
Join Date: Dec 2009
Posts: 6
Rep Power: 16 |
Did you follow tuturial No1 of ICEM
|
|
January 30, 2010, 15:29 |
|
#4 |
Senior Member
|
That tutorial is 2D. My project is 3D and in that tutorial mesh is tetrahedral. That means its pyramid but I want to generate perism mesh.
|
|
February 2, 2010, 05:28 |
|
#5 |
New Member
Join Date: Oct 2009
Posts: 20
Rep Power: 16 |
post a picture of your long pipe
|
|
February 3, 2010, 11:58 |
options...
|
#6 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
Are you looking for a swept mesh or a tetra prism mesh?
If it is tetra/prism, I think that a 1 meter long cell with a 1.5 meter diameter pipe is too coarse. You won't get enough curvature capture (assuming the prism is equilateral). I would suggest a sweep method. Mesh the end of the pipe (quad dominant, but with inflation (2D-BLayer)) and then use the option to extrude the mesh along the curve of the pipe. You will be able to have a decent cross section mesh, but have high aspect ratio elements so you can easily achieve your 1 meter length elements, but with good fit in your 1.5 meter diameter pipe. ICEM CFD Hexa would also be a very good choice for this geometry. you could block it quickly and use an OGrid to very precisely control the boundary layer. |
|
February 4, 2010, 15:54 |
|
#7 |
Senior Member
|
i have use Hexa mesh and blocking but i think if i use tetra mesh i would obtain better result. i want to use a mesh in shape of triangular that extruded normaly.
|
|
February 5, 2010, 01:23 |
|
#8 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
Since your pipe bends, a simple normal sweep won't work, but you could run a curve along the center of the pipe (or perhaps a seam along the side or inner surface of the bend would work...)
Then Extrude the mesh along the curve. Simon |
|
February 5, 2010, 10:15 |
|
#9 |
New Member
Join Date: Apr 2009
Posts: 14
Rep Power: 16 |
Why do you think tetra mesh is better in your situation? On the contrary, Hexa mesh is well-known to produce better results with less number of nodes/elements in wall-bounded flow.
|
|
February 6, 2010, 06:43 |
|
#10 |
Senior Member
|
thanks a lot
|
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Missing math.h header | Travis | FLUENT | 4 | January 15, 2009 12:48 |
Problem with meshing long, thin faces in CFX | Martin | CFX | 3 | January 8, 2009 21:51 |
Problem meshing very thin pipe | B. Hemmen | FLUENT | 2 | May 16, 2006 09:29 |
Meshing divergent nozzle entry of a long pipe | Aly | FLUENT | 1 | September 25, 2005 18:07 |
Terrible Mistake In Fluid Dynamics History | Abhi | Main CFD Forum | 12 | July 8, 2002 10:11 |