CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Error when activating Bouyancy (http://www.cfd-online.com/Forums/cfx/97239-error-when-activating-bouyancy.html)

monkey1 February 13, 2012 05:10

Error when activating Bouyancy
 
Hello!
When using Bouyancy in CFX I had to correct the pressure at the opening bc with a height dependent pressure formula (hydrostatic and barometric pressure contribution).

To correct the pressure I am using the following CEL expressions:
barometric = 1[atm]*(exp((-g*Molar Mass *z)/(8.314472[J/mol/K]*T)) )
hydrostatic = 1[atm]-Density *g*z
pstatic = barometric-hydrostatic

and I set pstatic as the pressure "value" in the opening BC.

For several cases it worked all fine. But now suddenly I keep on getting errors before any calculation would be runnig. The solver stops with the following message:

+--------------------------------------------------------------------+
| Buoyancy Reference Information |
+--------------------------------------------------------------------+

Domain Group: Tunnel

Buoyancy has been activated. The absolute pressure will include
hydrostatic pressure contribution, using the following reference
coordinates: ( 5.00000E+00, 1.50000E+00, 0.00000E+00).
Slave: 3 ----------------------------------
Slave: 3 Error in subroutine GET_SPECVAR :
Slave: 3 A recursion problem was encountered when evaluating the expression for
Slave: 3 "Pressure" at the following location: "Abzug". Please carefully check that the
Slave: 3 result of your expression does not depend on itself.
Slave: 3 GETVAR originally called by subroutine CAL_PRES_BCS

Parallel run: Received message from slave
-----------------------------------------
Slave partition : 3
Slave routine : ErrAction
Master location : RCVBUF,MSGTAG=1020
Message label : 001100279
Message follows below - :

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Stopped in routine GV_ERROR |
| |
| |
| |
| |
| |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| The ANSYS CFX solver exited with return code 1. No results file |
| has been created. |
+--------------------------------------------------------------------+

End of solution stage.

I am using CFX Version 13.0

Has anyone an idea what my problem could be?

Thanks in adavance for your answers!

ghorrocks February 13, 2012 07:24

The error says the problem with a function for "Pressure", not the functions you have shown. This is a really bad name for a variable as it clashes with the existing variable. You should give it another name to not clash.

monkey1 February 13, 2012 07:57

Thx Glenn!
Gonna try to solve the problem.

monkey1 February 13, 2012 08:35

Just checked the file again. There no function nor variable with the name of Pressure that I defined...so this wasn't the problem....hmmmm
any other guess?

ghorrocks February 13, 2012 17:34

Did you refer to a variable named "Pressure"? The correct variable name is "p".

monkey1 February 14, 2012 05:08

Yes, i referred to a variable Pressure, now i changed it to p. The first error message disappeared but now an analog error occurs as shown below.



Buoyancy has been activated. The absolute pressure will include
hydrostatic pressure contribution, using the following reference
coordinates: ( 5.00000E+00, 1.50000E+00, 0.00000E+00).
Slave: 3 ----------------------------------
Slave: 3 Error in subroutine GET_SPECVAR :
Slave: 3 A recursion problem was encountered when evaluating the expression for "Total
Slave: 3 Pressure" at the following location: "Abzug". Please carefully check that the
Slave: 3 result of your expression does not depend on itself.
Slave: 3 GETVAR originally called by subroutine CAL_PRES_BCS

Parallel run: Received message from slave
-----------------------------------------
Slave partition : 3
Slave routine : ErrAction
Master location : RCVBUF,MSGTAG=1020
Message label : 001100279
Message follows below - :


I canīt find the reference to "Total Pressure" in my case except for the list of selected output variables for the .trn files.


I refer to a Variable called "Reference Pressure", maybe that's the Problem? Is there also another appelation I should use instead?


And how comes that using the names the CFX pre proposes (e.g. Pressure) makes a difference to using the variable (p)? Is this relevant for all other variables like e.g. Substancename Mass Fraction to substancename.mf??


Thanks a lot!

ghorrocks February 14, 2012 17:41

You should look at the CF Reference guide. It has a list of the correct variable names of all variables.

monkey1 February 15, 2012 03:22

Thx for the info!

happy April 19, 2012 05:04

for your help
 
Quote:

Originally Posted by monkey1 (Post 344505)
Thx for the info!

Hi monkey,
your experssion of hydrostatic pressure contains (T)!!so your equation contains 3 variables, your pre., Z, and T. T should be ambient temp. in your case (298.15 k for example). your equation actually takes into account change the air pre. w. r. t. height if you used large large domain so your equation control inflow air at the opening from outside your domain which has constant temp. ( you assume). So whay you put T instead 25 ( value).
another thing, is look at your your refernce pressure and density, did you defined them propeply?
what is your model for?:D
Regards

happy April 19, 2012 05:36

for your help
 
Hi again..
your expression includes Density!!!How you want the cfx calculate this one. it is varaible come from calculations and B. Cs should be more clear values and did not depend on calculation within domani ( mean depend on itself/stand alon). when comapre between Y and density, y is distance provided by mesh did not need to calculate but density needs. So my advise is replace Density by ambient density of your fluid.

Regards

happy April 19, 2012 05:49

could you explain your problem more to have good help
 
what sort of flow you have? hot air or fire? is your domain too big?
what sort of B.C.s U used???
You told us U used opening so which option you used? opening or static?
opening& direction or entrainment??
I know before now that CFX cannot model buoyancy correctly so that user should apply some workaround.
if U interst to share your knowledge with me, You are very welcome:rolleyes:

Regards

happy May 9, 2012 21:52

question
 
hi again
so what you mean about the hydrostatic = 1[atm]-Density *g*z. where it takes place.
Regards


All times are GMT -4. The time now is 03:15.