CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Outlet Problem (http://www.cfd-online.com/Forums/cfx/63453-outlet-problem.html)

cfxuser1 April 8, 2009 16:04

Outlet Problem
 
Hi everybody. I am simulating a flow around a body and am getting the error message below. I do not really understand why I am getting this error. Can anybody please help me understand that.
Thanks.
Faraaz

-------------------------------------------------------------------+
| ****** Notice ****** |
| A wall has been placed at portion(s) of an OUTLET |
| boundary condition (at 52.0% of the faces, 51.6% of the area) |
| to prevent fluid from flowing into the domain. |
| The boundary condition name is: Outlet. |
| The fluid name is: Air at 25 C. |
| If this situation persists, consider switching |
| to an Opening type boundary condition instead. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Floating point exception: Overflow |
| |
| |
| |
| |
| |
+--------------------------------------------------------------------+

ghorrocks April 8, 2009 18:56

Hi,

The first message says you have reverse flow at that outlet. It is only a warning, this is not always a problem.

The second is an overflow error, the simulation has diverged.

Glenn Horrocks

bharath April 9, 2009 09:25

outlet problem
 
hi,
try this.
first change the boundary condition to opening.then try.
if it is not successful then check ur mesh quality.it may have negative elements. or use some coarse mesh.check ur boundary conditions too

cfxuser1 April 9, 2009 15:51

Hi,

Thanks guys for helping. Yes I am trying some stuff right now. If there's any problem, I will let you guys know.

Chilli83 April 10, 2009 12:02

Hi all

I have the same problem but also two other questions.

1)
How can I be sure that there will be no inflow in a opening as bc for outlet?

2)
Also what would be the most "correct" way to define/assume the temperature at the outlet?

I have used a subroutine to calculate the average temperature for my outlet region since i thougt this was smart to do.

kind regards
Michael

bharath April 11, 2009 08:47

hi,
1) check your areaave,massave variables at opening.if the deviation between variables are high then there may be inflow.

2) could u explain ur problem further more detail?

Chilli83 April 11, 2009 08:59

Quote:

Originally Posted by bharath (Post 212585)
hi,
1) check your areaave,massave variables at opening.if the deviation between variables are high then there may be inflow.

2) could u explain ur problem further more detail?

Hi bbarath

The problem is that when I define a opening a temperature at the opening is nessecary. So by the subroutine the program calculate a average temperature for the region I have define as the outlet. I am in doubt if this is the right way to set the temperature or is I just could type in a given temperature?

kind regards
Michael

bharath April 11, 2009 11:15

you just set the opening temperature as an approximate guessed one.when solving the problem, solver will fix the temperature.

Tassi April 15, 2009 06:50

Hi Michael

Your method for defining the temperature at Opening is correct. Wall generation at Outlet can signify many things. See the CFX documentation on setting boundary conditions.

BOL

Tassi

Chilli83 April 15, 2009 07:22

Quote:

Originally Posted by Tassi (Post 212983)
Hi Michael

Your method for defining the temperature at Opening is correct. Wall generation at Outlet can signify many things. See the CFX documentation on setting boundary conditions.

BOL

Tassi

Hi Tassi.

I think the reason why the solver creates artificial walls is caused by a recirculation zone. So i thougth that it would be appropriate to model the outlet as an opening to avoid numerical error caused be a wrong pressure field.
I have added i screendump of my model where the inlets and outlet region is illustrated. The 3D blocks are persons. It is the region below the 3D blocks the recirculation zone is expected and here the solver creates artificial walls.

http://img245.imageshack.us/img245/724/modelmixing.jpg

Kind regards
Michael

bharath April 15, 2009 21:04

hi Michael,
i have a doubt on the simulation. Even if you give the outlet as opening, whether the artificial wall created?

Tassi April 16, 2009 00:27

Hi Michael
Yes the solver creates artificial wall cause of recerculation.
You can deactivate solver from creating wall at boundaries by deactivating the function in expert parameters in Pre. Hasn't been of much help though.
Normally I solve such problem by extending the outlet a little away from such recirculations. Sometimes adjusting mesh size and decreasing timesteps have helped.

BOL

Tassi


All times are GMT -4. The time now is 15:59.