CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [GAMBIT] meshing of wavy fins (http://www.cfd-online.com/Forums/ansys-meshing/84441-meshing-wavy-fins.html)

dldl January 29, 2011 20:59

meshing of wavy fins
 
2 Attachment(s)
Hi
I would model a heat exchanger with finned wavy fins. I drew the model in Gambit and wanted to mesh the model. I meshed the four pipes as the following:
-Mesh face scheme for the Fin-Tube: Quad – Map
-Mesh volume ccheme: Hex/Wedge – Cooper
I then tried to mesh the other volume with “Hex/Wedge – Cooper” But a problem always happens when I mesh Volume 3.
I tried later to mesh the volume 1 and 2 with “Hex/Wedge – Cooper” and the volume 3 with the scheme “T grid” but the problem has remained.
Can someone help me how I can mesh the volume 1, 2 and 3 without errors.
Thanks

-mAx- January 31, 2011 02:49

I don't understand the difference between volumes 1,2 and 3
As I understand, you have 4 pipes (annular pipe), and the other volume (the wavy one)

dldl January 31, 2011 04:57

Thanks Mr. Max
I apologize that my English is poor.

Iat the beginning drew 3 volumes. Volume 1 is the half thickness of fin1.
Volume 2 is the half thickness of fin2
Volume 3 is the gap between the two fins 1 and 2
These three volumes are connected and can be merged. I believe, however, that is not necessary!
the others volumes are annual pipes.
(if you can german, we can write in german)

-mAx- January 31, 2011 05:00

Can you post a picture where we can see volumes 1,2 and 3
Yes I can german, but since the thread could be usefull for anyone else, it is better to continue in english.
If I don't understand your english, I will send you a private message in german

dldl January 31, 2011 05:39

1 Attachment(s)
If the arc is not so long and the curve is not so great, I can mesh the model simply that the volume 1 and 2 with Scheme "Cooper" and Volume 3 with T- Grid.

-mAx- February 1, 2011 03:50

1 Attachment(s)
ok,
I saw your dbs file.
You sent me 2 versions, but you have to work with the second: the one with all the surfaces and edges merged.
It will avoid high skewness in your mesh, and it will make all volumes meshable with hex-cooper (automatically selected from Gambit).
This is exactly your second dbs file (arc3h11)
Now there is a problem while meshing. With my computer, you can mesh one volume, then the other makes the meshing prozess to be frozen.
Just for checking, if you merge all the volumes, then this problem disappear.
That mean the problem should come from the wavy and thin volume.
There is a workaround:
you need to split all your volumes for avoiding wavy extrusion in subvolume.
Doing this you can then mesh all your volumes directly with hexa.
Another advantage, is that you will have more control on your mesh and you will be able to refine the subvolumes with less effort from your hardware.
Since your dbs file is already made from virtual volumes, for splitting you need to:
*split edges >> vertex generated
*split surface from vertices >> edges liying on the surface is generated (edge is a split of your surface)
*creating wireframe surface from edges
*split your volume with the surface
Attachment 6294

dldl February 1, 2011 04:07

Thanks
I will follow these steps.

dldl February 1, 2011 06:18

I have a question about boundary conditions:
How to determine the BC for the lower and upper faces of Fins?
The following is known:
-Water temperature through pipes is 308 k
-Temperature of the air (inlet) is 293 k
I mean here the thermal condition!

ArunPrasanth February 18, 2015 14:10

So you are saying that each part should be meshed separately ??


All times are GMT -4. The time now is 15:08.