|
[Sponsors] |
How to set boundary layer of a moving body in GAMBIT to a mesh zone for dynamic mesh |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
March 22, 2009, 17:43 |
How to set boundary layer of a moving body in GAMBIT to a mesh zone for dynamic mesh
|
#1 |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
How to set boundary layer of a moving body in GAMBIT to a mesh zone for dynamic mesh?
I want to do dyanamic mesh in Fluent. Just a single paddle rotate. I want create boundary layer around the paddle, and let the boundary layer moving with the paddle without remeshing. The remeshing only happens outsite the boundary layer. So anyone know how to set this boundary layer in to a zone or something for achieve this? Thanks a lot Tom |
|
March 22, 2009, 23:55 |
|
#2 |
Senior Member
Join Date: Mar 2009
Posts: 225
Rep Power: 18 |
Well, I don't know what you really want to do, but I had some problems with boundary layer within dynamic meshes. My boundary was moving just in one direction lets say in x direction, so what I figured out my model had trouble with cell adjustment around the wall area, I was using layering method which results in cell collapse.
At the end I figured out area without boundary layer in this region did no make too much difference, since I was interested in result far away from what moving boundary was doing. Hope this helps. |
|
March 23, 2009, 06:13 |
|
#3 |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Thanks paka, but I still looking for the answer of the boundary layer. anybody can help? thx
|
|
March 23, 2009, 18:42 |
|
#4 |
Member
Join Date: Mar 2009
Posts: 49
Rep Power: 17 |
Hi there,
I'm calculating airfoils with boundary layers which shall not deform when i move the blade. Therefore I created a face around the blade which covers the boundary layer and which does not deform anyway so the deformation only takes place in the rest of the flow domain. this works quite good... |
|
March 23, 2009, 21:22 |
|
#5 |
Member
Join Date: Mar 2009
Posts: 35
Rep Power: 17 |
Using the sliding mesh approach may protect your boundary layers.
|
|
March 24, 2009, 04:59 |
|
#6 |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Thanks O.D.Y for ur answer, but I still not clear how can u create a face exactly has the same size at the boundary layer? Does it matter if the face is just a little bit bigger or smaller than the boundary layer? Hope for ur valueless answer!
|
|
March 24, 2009, 05:00 |
|
#7 | |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Quote:
|
||
March 24, 2009, 05:10 |
|
#8 |
Member
Join Date: Mar 2009
Posts: 49
Rep Power: 17 |
Well I'm of the opinion, that it doesn't matter if the face is a little bit bigger than the BL reaches. I mean your main intend is to keep the BL mesh constant close to the wall and if you extend the constant area a little bit further the only thing you have to take care for is that you have enough mesh in the rest of the domain left to cover the mesh deformation due to the body movement.... Well in my case I'm doing steady state calculations where I can estimate the boundary layer thickness and create another mesh afterwards with the proper face sorrounding the blade... Honestly this is an easy task for 2D calculations but the effort increases for 3D to make up a proper flow domain in gambit....
|
|
March 24, 2009, 05:15 |
|
#9 |
Senior Member
Join Date: Mar 2009
Posts: 225
Rep Power: 18 |
Well... I never did it, but it seems you can define BL faces as ones which move similarly to body edges. Don't you think this might work? So the BL part is so saying "rigid" with body deformation and only the part of the mesh outside BL should deform. I think this might also work in 3D.
|
|
March 24, 2009, 06:00 |
|
#10 |
Member
Join Date: Mar 2009
Posts: 49
Rep Power: 17 |
Yeahh that should work. Well I did this in 2D, meaning the "BL-face" moving as rigid body and deforming the mesh outside, I just mean its a bit more effort to make up the model in 3D concerning my cases ;-) but i guess this is far better than changing smoothing and remeshing parameters until it works "somehow good"....
|
|
March 24, 2009, 16:49 |
|
#11 | |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Quote:
|
||
March 25, 2009, 06:50 |
|
#12 |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
anybody can help, please
|
|
March 27, 2009, 13:18 |
|
#13 |
Member
Join Date: Mar 2009
Posts: 49
Rep Power: 17 |
Actually you should be able to find the boundary face in the boundary conditions menu.... and of course you should define the boundary face as fluid. And you need to make sure that the interfaces between the boundary face and the face for the rest of the domain have the same interface edges. That should work because I'm doing it that way without problems.... cheers
|
|
March 28, 2009, 10:23 |
|
#14 | |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Quote:
|
||
March 28, 2009, 15:21 |
|
#15 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
In 2d-->
I would create a curve surrounding your airfoil. The distance to the airfoil should be greater than your BL. Then split your domain with this curve. Mesh all your domain and add your BL. Then in BC, go to zone panel and create a surface zone as fluid (in will be your rigid body motion's surface). In fluent you should be able to set this one as rigid body in MDM panel. For 3d, replace curve with surface, and surface with volume. |
|
March 28, 2009, 16:23 |
|
#16 | |
New Member
ruiyang
Join Date: Mar 2009
Posts: 9
Rep Power: 17 |
Quote:
|
||
January 18, 2012, 05:47 |
Hi
|
#17 |
New Member
mojtaba
Join Date: Nov 2010
Posts: 25
Rep Power: 16 |
||
March 25, 2016, 05:36 |
|
#18 | |
Senior Member
|
Quote:
where in dynamic mesh do we specify not to move/disturb the boundary layer? |
||
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Moving mesh | Niklas Wikstrom (Wikstrom) | OpenFOAM Running, Solving & CFD | 122 | June 15, 2014 07:20 |
moving boundary / mesh | John | Main CFD Forum | 1 | May 8, 2008 11:28 |
Deal with wall boundary with moving mesh by FVM? | aiya | Main CFD Forum | 6 | May 10, 2007 12:33 |
can fluent capture BOUNDARY LAYER of POROUS zone? | Alex | FLUENT | 0 | April 22, 2006 08:14 |
Convective Heat Transfer - Heat Exchanger | Mark | CFX | 6 | November 15, 2004 16:55 |