|
[Sponsors] |
June 30, 2010, 16:32 |
Transient problem with changing geometry!
|
#1 |
New Member
Faisal
Join Date: Jun 2010
Posts: 11
Rep Power: 16 |
Hello all,
I am back once again. I feel it is wiser if I share the diagram for my problem that I posted earlier to help you contribute better. Here is the photo (or link to the photo): http://www.4shared.com/photo/2KI1nPIo/Detail.html As I mentioned before, Iam just new in using FLUENT and trying to acquint myself with the software, considering using it (FLUENT) to model a multiphase flow (Fresh concrete paste). The fresh concrete is characterized as a Bingham fluid with its rheological parameters (yield stress and plastic viscosity) determined experimentally. The geometry to be modeled is a truncated cone (both ends open). The end with larger diameter is placed on a solid base and filled with concrete from the top. When filled up, the top is leveled. Instantly, the cone is lifted and the concrete is allowed to flow radially over the metal base. At timed intervals, the diameter of the concrete is measured. How possible is it for me to achieve that using FLUENT? That is, setting the problem up and plotting the snap shots of the flow at intervals. In addition, the distribution of velocity values need to be plotted. Your help is a welcomed one. Last edited by Fascal; June 30, 2010 at 16:34. Reason: omission |
|
July 2, 2010, 21:45 |
Vof
|
#2 |
Member
Marco Evangelos Biancolini
Join Date: May 2009
Location: Rome - Italy
Posts: 80
Rep Power: 17 |
I think that you can do the job using VOF and moving mesh. Try first with VOF only (neglecting the movement of the bucket): you need a model large enough to contain your interface at the end of the simulation...
|
|
July 4, 2010, 12:06 |
Transient problem with changing geometry!
|
#3 |
New Member
Faisal
Join Date: Jun 2010
Posts: 11
Rep Power: 16 |
Dear meb,
I am highly grateful for your comment on my post. I tried using the VOF model as you suggested, but here is how I did it and the problems I encountered: But, as I told you that I am new in using FLUENT, I guess the problem is the geometry creation & boundary conditions. In gambit I created a big rectangle (that can contain the interface after simulation). Then, I created another shape to represent the bucket. Am I suppose to create the phases (concrete & air) this way? I specified the boundary condition on the two slanting sides of the bucket as "Pressure-outlet". Is this boundary condition correct? Finally I meshed the two faces independently. This brings confusion to me because I think the two media (air & concrete) should be continuous not the way I treated them here. After meshing, the geometry looks like the one in the following link: http://www.4shared.com/photo/4G7iXRK3/grid.html While solving in fluent, I found that I can only specify a rectangular shape in defining "cell register" for me to "patch" the concrete, where as my geometry (bucket) is not rectangular. So please, can you shed more light on what my mistakes are and how I can do it? Once again thank you for your concern in trying to help students likes me. Regards |
|
July 5, 2010, 06:26 |
Model set-up
|
#4 |
Member
Marco Evangelos Biancolini
Join Date: May 2009
Location: Rome - Italy
Posts: 80
Rep Power: 17 |
Using VOF you haven't to split the mesh in two region; the presence of concrete is given as an initial condition to species concentration.
Use a mesh with the bucket already separated (so you haven't to concern about deforming mesh). In the VOF model you have to prescribe the starting interface as an initial condition: pick a part of the cells in the upper part of the internal area of the bucket. |
|
July 5, 2010, 11:11 |
It worked!
|
#5 |
New Member
Faisal
Join Date: Jun 2010
Posts: 11
Rep Power: 16 |
Dear Marco, Thank you verymuch for your help. I did it the way you explained, and it worked excellently.
However, I had to use rectangular shape as my starting interface while prescribing the initial condition. Whereas the shape should be trapezoidal, and not rectangular. Therefore, the only problem still remains how the starting interface (shape of the concrete) can be prescribed as a bucket shape (trapezoidal). I was reading a tutorial about that. The tutorial explains that I have to define a cell register by going to Adapt -> Region where I should enter the(Xmin,Ymin) & (Xmax,Ymax) to specify my region of interest as shown in this link http://www.4shared.com/photo/PTMym3x..._register.html After that, I clicked Manage & clicked on hexahedron-r0 then Display. When I did that, I found that I can only specify a rectangular shape with that command. But that is how I continued to solve it by going to Solve->Initialize->Patch where I selected the register to patch as "hexahedron-r0". Please, tell me is there any way I can prescribe my region to appear as trapezoidal? Because it is my actual shape. Regards |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Star-CD transient simulation, problem of loss of mass | kit | STAR-CD | 3 | April 13, 2010 12:34 |
Problem in Initializing Blocking for 2d geometry | nitin Dewangan | CFX | 6 | October 24, 2008 02:03 |
Problem when I try to write the geometry file | massimo | Siemens | 0 | January 14, 2003 05:58 |
transient simulation: natural convection problem? | Basics | CFX | 3 | September 25, 2002 10:42 |
geometry file problem | a.melandinos | CFX | 1 | February 28, 2002 18:41 |