|
[Sponsors] |
possibility of converting 3D->2D for saving simulation time |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
April 11, 2013, 20:04 |
possibility of converting 3D->2D for saving simulation time
|
#1 |
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Hello Foamers,
I am just wondering if I can reduce a 3D symmetrical simulation to a 2D simulation (maybe it is a dull question)? What I am doing is to find the oil spreading time on cylindrical water tank. Since I need to make refined mesh of the interface of oil and water, but I can not afford the huge computational time for a 3D modeling. I am not sure whether the simulation result is same or not. Please give me some suggestion if possible. |
|
April 14, 2013, 19:31 |
|
#2 | |
New Member
Eric
Join Date: Mar 2013
Posts: 22
Rep Power: 13 |
Quote:
Hello, you can try using a 2D axisymmetric wedge. However, I cannot comment the accuracy. Would it be possible for you to coarsen the 3D mesh? |
||
April 14, 2013, 19:45 |
|
#3 | ||
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Thanks for the reply, do you have any idea about how to set up the boundary condition?
Here is my U file for 3D case Quote:
Quote:
Thanks very much for that. |
|||
April 15, 2013, 00:46 |
|
#4 |
Member
Akshay Kumar
Join Date: Aug 2010
Location: India
Posts: 84
Rep Power: 15 |
Hi Guifan
The 2d axis-symmetric wedge bc should work fine as Eric mentioned. What you have written down is the way to go. Put up a few pictures of your case and let's see how it's supposed to be done. A small wedge angle needs to be there from the start. You cannot bring in an actual 2D model. Cheers! Akshay |
|
April 15, 2013, 10:56 |
|
#5 | |||||||
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Hi Akshay,
Firstly, thanks very much for you comment. Here is the case I use a 2D to simulate the 3D( equivalent to: I cut a piece from the 3D water tank centre) http://www.cfd-online.com/Forums/ope...tml#post420524 According to your suggestion, I also tried the 2D-axisymmetrical wedge and here is the details. I following the following steps 1) build a rectangular mesh and convert it to a wedge using makeAxialMesh -axis axi -wedge frontAndBackPlanes ultility then I get the following file in the newly created folder, and I replace the original Ploymesh folder under contant/ Quote:
I get Quote:
Since the "axi" and "frontAndBackPlanes" are with nFaces 0, so I deleted them in the boundary file and updated the alpha.org, U and p_rgh file accordingly. 5) I run checkMesh, I got the errors Quote:
When I run multiphaseInterFoam I got the erros Quote:
Quote:
Quote:
Quote:
|
||||||||
April 15, 2013, 11:46 |
|
#6 | |
Senior Member
Niels Gjoel Jacobsen
Join Date: Mar 2009
Location: Copenhagen, Denmark
Posts: 1,900
Rep Power: 37 |
Hi Guifan,
The reported problem in your simulation is with your constant/transportProperties file. Please post this as well. Where I see this? Quote:
Niels |
||
April 15, 2013, 12:01 |
|
#7 | |
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Thanks for the reply!
Sorry for that I forgot to put it. Quote:
|
||
April 15, 2013, 12:06 |
|
#8 |
Senior Member
Niels Gjoel Jacobsen
Join Date: Mar 2009
Location: Copenhagen, Denmark
Posts: 1,900
Rep Power: 37 |
The problem is that you are using a transportProperties file for interFoam, however, you have executed multiphaseInterFoam.
Take a look in its tutorial to see how you should define the fluid properties. Kind regards Niels |
|
April 15, 2013, 12:15 |
|
#9 | |
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Hi Niels,
Thanks for pointing out this for me and I will give it another try to see what happen later this afternoon. Beside this problem, do you think my boundary for "axi" which is the axis correct. Since in tutorial of NozzelFlow2D, its BC for axis is empty. However in my case after I execuated makeAxialMesh command. The BC for me axis is Quote:
Kind regard, Guifan |
||
April 15, 2013, 12:17 |
|
#10 |
Senior Member
Niels Gjoel Jacobsen
Join Date: Mar 2009
Location: Copenhagen, Denmark
Posts: 1,900
Rep Power: 37 |
Hi Guifan,
I can unfortunately help you, since I do not have any experience with wedge-type meshes. Kind regards Niels |
|
April 15, 2013, 12:18 |
|
#11 |
Member
Guifan Li
Join Date: Apr 2011
Location: New York City, U.S.
Posts: 96
Rep Power: 15 |
Thanks anyway for the help! I will try to test more case to see the difference.
Cheers! Kind regards, Guifan |
|
April 16, 2013, 00:43 |
|
#12 |
Member
Akshay Kumar
Join Date: Aug 2010
Location: India
Posts: 84
Rep Power: 15 |
Hi Guifan!
The 'axi' boundary has 0 faces anyway, so it shouldn't really matter what BC you're declaring it as. As Niels rightly pointed out.. "--> FOAM FATAL IO ERROR: Attempt to return dictionary entry as a primitive" The way you have declared the phases are wrong. The materials should fall inside a phases block. Let me know if this gets you through. Post your fvSolution file as well. Cheers! |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Sudden jump in Courant number | NJG | OpenFOAM Running, Solving & CFD | 7 | May 15, 2014 13:52 |
Simulation time vs. Simulation Cell Count | chrisoturner | Main CFD Forum | 7 | September 8, 2012 17:46 |
plot over time | fferroni | OpenFOAM Post-Processing | 7 | June 8, 2012 07:56 |
Total time for simulation | Meerkat | Main CFD Forum | 1 | September 5, 2011 07:17 |
Time step in transient simulation | shib | FLUENT | 0 | June 17, 2010 13:07 |