|
[Sponsors] |
December 5, 2011, 21:07 |
Cylinder with slots meshing
|
#1 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
Hi...I am very new to the gambit software and wish to know how to specify the boundry zones for a cylinder with slots. I am trying the 2d meshing for the same. Kindly help me in understanding the concept because when I am getting an error saying the boundry condition specified is invalid for all the edges.
Thanks Kaushik |
|
December 6, 2011, 00:51 |
|
#2 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
what kind of BC are you trying to attach (and where)?
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 6, 2011, 03:01 |
|
#3 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
thanks for your reply max...Actually, i sorted out the boundry condition..I would be grateful if you could advise on the right meshing of the slotted cylinder which is my job. It is subjected to a liquid flow of 10 m/s. How should I divide the rectangular area so that I may get the best possible result regarding the pressure conditions around the cylinder at every slot position.
I am sorry but I am asking you the basics but pls help me as am a beginner in the area. I was also trying the same in 3d with a rectangular prism as my boundry area and a slotted cylinder. So how do I mesh the area to get the best result? I tried first Quad mapping the faces and also the cylindrical slots but I got an error. Thanks |
|
December 6, 2011, 03:07 |
|
#4 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
post a picture with your geometry and display how should flow your fluid
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 6, 2011, 03:26 |
|
#5 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
I am attaching the pic of the 3d drawing i made in gambit with a rectangular prism.
|
|
December 6, 2011, 03:34 |
|
#6 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
ok
For a quick but acceptable mesh, I would apply size function on cylinder-surfaces (choosen as source) >> http://my.fit.edu/itresources/manual...ommands_create Then go and mesh the volume (no element size is required since it would be driven because of your size function >> max element size) Choose tet-Hexcore algorithm You are also free to attach a boundary Layer on your cylinder-surfaces
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 6, 2011, 18:55 |
|
#7 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
thanks max..so should i select the source as 8 cylindrical slots and attachment as the entire cylindrical body volume using proximity type size function. My volume is a single entity consisting of rectangular prism minus the cylindrical part..could you please explain step wise..if i am wrong here..thanks for your time..
|
|
December 6, 2011, 21:40 |
|
#8 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
I got an error message " previously meshed edges on one of the faces is unacceptable" . I cant diagnoze the problem
Please help... |
|
December 7, 2011, 02:23 |
|
#9 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
how many volumes do you have?
Can you post a picture of your domain with shaded modus enabled?
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 7, 2011, 03:31 |
|
#10 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
I have only 1 volume as shown in the pic
|
|
December 7, 2011, 04:02 |
|
#11 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
Regarding size function: you need to select all surfaces from your cylinder as source, and your volume as attachment.
If you get one error message, check if Gambit gives you any information about edge or surface name.
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 8, 2011, 20:15 |
|
#12 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
Thanks max...it worked...now I would do the next step. Could you tell me how to rotate the slotted cylinder so that I can do a dynamic study?Do we have to change the meshing since mine is static right now?
|
|
December 12, 2011, 01:24 |
|
#13 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
this is now a complete different approach: you need to work with intefaces.
For that you create anoter cylinder (but bigger as your own), and you split your domain with new cylinder. Now you have 2 volumes: one will be stator, and the other rotor. To let the rotor rotate independantly from stator, you need to deconnect both together. Once it is done the only thing you need to define is interfaces. If everything is ok, you should have superposed surface where domains were connected. So at the "interface" just define surfaces from rotor as interface1 (INTERFACE - BC) and those from stator as interface2 ( INTERFACE - BC). Finally in Fluent you need to create the interfaces by picking both interfaces-bc. As fluid domains are now "reconnected"(data extrapolation through interfaces), you can rotate your rotor in the moving mesh pane.
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 12, 2011, 03:06 |
|
#14 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
Is there any tutorials available for rotating cylinder? ..it would help me get a better understanding...anyways Ill try your suggested methodology..
|
|
December 12, 2011, 03:11 |
|
#15 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
Is there a tutorial for 2d rotation of cylinder?Then I will try the 3d rotation..
|
|
December 12, 2011, 06:46 |
|
#16 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
Check here http://my.fit.edu/itresources/manual...tg/node162.htm, but watch out this is MRF, so no moving mesh.
If you want your rotor to rotate, you need to activate moving mesh
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 12, 2011, 21:30 |
|
#17 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
Hi..I was making a 2d cylinder rotation problem...I made a circular face of r=1 then a bigger circular face of radii 8. Then I split the 2 faces...i am not able to get the superimposed surfaces...please can u explain step by step...thanks for your time...
|
|
December 13, 2011, 01:05 |
|
#18 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
if you split the surface with the other surface, then your 2 domains are connected, and you don t get any superposed edges (2d).
To disconnect either you go to edge/disconnect. Or you can copy and translate your rotor domain, then delete the orginal rotor surface (connected), and finally move back the rotor-copy
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
December 14, 2011, 01:55 |
|
#19 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
I tried doing it by sliding mesh but I am not able to get the results that I expected.I feel my answer is wrong.I am confused on what should be the boundry conditions that I should set. I drew a cylinder over and above the rotor as interface. Is it correct? Should I give rotational velocity to the interface? I only want the slotted cylinder to rotate and gauge the effect of its rotation on incoming velocity from left as shown in the fig. Max help me to understand on how to specify boundry conditions?
I need to find the simulation for inlet water of 1 m/s and the slotted cylinder rotating at 5 m/s. Ineed to find the pressure and velocity distribution mainly at the slots of cylinder. This is my last request to you. Thanks for your patience. |
|
December 14, 2011, 01:58 |
|
#20 |
New Member
Kaushik
Join Date: Dec 2011
Posts: 16
Rep Power: 14 |
The result I obtained is in the pic..
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
meshing on cylinder fails ICEM | potac | ANSYS Meshing & Geometry | 0 | October 25, 2011 05:21 |
2D planar cylinder meshing problem | saisanthoshm88 | ANSYS Meshing & Geometry | 15 | December 27, 2010 11:14 |
[ICEM] (ICEM CFD) meshing over a cylinder | vmlxb6 | ANSYS Meshing & Geometry | 6 | August 11, 2010 14:37 |
[blockMesh] Meshing inside a cylinder so that the edges contain more cells than the center | Bertrand | OpenFOAM Meshing & Mesh Conversion | 1 | April 13, 2010 04:46 |
Best Meshing scheme for Cylinder | Nutrex | Main CFD Forum | 4 | July 29, 2008 11:03 |