|
[Sponsors] |
May 29, 2016, 17:51 |
Computational domain, lid tool
|
#1 |
New Member
Join Date: May 2016
Posts: 20
Rep Power: 10 |
Hello good people,
I am running FloEFD on student's license and I am facing some difficulties regarding setting up an internal analysis. I have simplified the geometry, but left some openings which now I want to turn into environment pressure boundary conditions and following difficulties occur. - Each lid has its width and I can apply pressure b.c. only on interface of solid and computational domain, therefore in order to place the wall in computational domain, the lid should have stick outside the box (see picture). The lid is created between inner and outer wall of enclosure so I cannot use it to apply b.c. 2. It is difficult to snap the computational domain to the geometry. I literally have to scroll (zoom in) as much as it is possible an pull the computational domain onto desired wall. Sometimes it does not snap perfectly with the material, but the software does not show any error( I guess it's a floating point issue). Am I doing something wrong or does this happen to you too? 3. What happens if in internal analysis the computational domain exceeds the solid boundary (enclousure/rectangular box with pcbs and fans inside)? |
|
May 29, 2016, 19:42 |
|
#2 |
New Member
Join Date: May 2016
Posts: 20
Rep Power: 10 |
Ok, I got it running.
Similar issue was brought up before -> FloWorks - "Face xx is not laying on the boundary between the solid and the fluid reg but maybe I will tell what helped me: 1. Fluid volume was zero (or really really close to zero) in internal simulation, because my model was not watertight (or some little little part was, due to sloppy geometry), therefore I got the message that b.c. (fan, pressure or inlet) was not placed on the interface of solid and fluid (there was no fluid at all). 2. I runned check the geometry, which resolved all edge to edge and edge to surface connections, and fluid volume was recognised well. 3. Attached desired b.c.s with no error (both fans and pressure outlets). I'm still not sure how to treat internal simulation - shouldnt the computational domain snap to the outer walls of enclosure, or can I zoom in and put it right next to it and that wont hurt the project? |
|
August 16, 2016, 13:14 |
|
#3 |
Disabled
Join Date: Jul 2009
Posts: 616
Rep Power: 24 |
Hi Pawel,
FloEFD snaps the computational domain to the fluid volume only if you have a closed (water tight) inner fluid volume and no heat conduction. As soon as you have heat conduction all solids will be included in the computational domain. If you have a tiny gap somewhere it will report it with an error in case you set internal and a boundary condition is applied and FloEFD cannot find the internal space. There is a way to find these tiny gaps with the help of the leak tracking tool. Here you select an internal and an external face and it will create a band from the inside to the outside through the first hole it finds. You can zoom in and follow the band to the point where it suddenly slips through the walls for some reason. This is where the gap or hole is located. Once you fixed it, check again as there might be another hole somewhere else. If you do that until you found the last hole then it will tell you it couldn't find any connection between the two faces. It is that easy :-) Boris |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
sliding mesh problem in CFX | Saima | CFX | 46 | September 11, 2021 08:38 |
[GAMBIT] decreasing computational domain | aja1345 | ANSYS Meshing & Geometry | 2 | April 20, 2015 02:15 |
Computational domain size | stevie | Main CFD Forum | 2 | March 10, 2015 11:16 |
Accumulation of particles inside the computational domain | liliana | FLUENT | 0 | June 2, 2014 20:10 |
Computational domain for a blade cascade | aledexe | CFX | 2 | June 3, 2012 07:43 |