CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   A problem about density in liquid air definition (

alloveyou May 12, 2011 21:29

A problem about density in liquid air definition
Hello Forum,

I have a problem.
I want to simulate an liquid air turbine.Here i have to define a new material: liquid air at 95-105K.
There is a problem:when the density is set to a constant value or values dependent on p(pressure),the solver can simulate it.But if it is set to a function related to p and T, the solver make errors.

the density definition is:
(837.336029[kg m^-3]+(4.927578[kg m^-3 K^-1])*T-(23.125020[kg m^-3 Pa^-1])*(p/1000000)-(0.056504[kg m^-3 K^-2])*T^2+(0.287167[kg m^-3 K^-1 Pa^-1])*(T*p/1000000)-(0.221818[kg m^-3 Pa^-2])*(p/1000000)^2)
the error shows:
Fatal bounds error detected
Variable: Density Derivative wrt Pressure at Constant Temperature
Locale : R1

I have read the help and the help shows below:

Density and Specific Heat Dependencies

For a pure substance, the flow solver only allows density and specific heat capacity to be functions of temperature or pressure.
When you select the Value option for density or specific heat capacity, you can use an expression to specify Density and/or Specific Heat Capacity, but the expression is only allowed to depend on pressure and temperature for liquids and gases. For a solid, they can depend on temperature or spatial location (x, y, z), but not both at the same time.
For ideal gases, the density is evaluated using the Ideal Gas law and the specific heat may be a function of temperature only.
Both density and specific heat capacity can also be a function of an algebraic additional variable, but the algebraic additional variable must only be a function of temperature and pressure. ANSYS CFX-Pre will not allow you to set this and you must therefore edit the CCL file to implement such a case.
If density, specific heat, or other properties are set as expression which depend on pressure, the solver automatically uses the absolute pressure when evaluating the expression.
I think the definition is right but I can't find what is wrong...

I am using turbulence: SST.... buoyancy : no buoyant

Any help or advise would be appreciated.....
Thank you very much!

viking June 14, 2012 04:31

Hello alloveyou,

i have the same problem simulating a cavitative model. did you check your table generation in the material definition? well, as i imported your expression, i notice that your density is negative for some high pressure values about p = 8e+07 Pa. this could be one reason for the error, because the solver checks the density value for a high range of pressure. so check your density equation in the range of your table generation in material definition.

however, did you solve this problem? please let me know how you did fix it.

kind regards

viking June 14, 2012 14:20

hi there, its me again,

i think it is useful to have a CEL function for the density that can be derivated for gradients. so i tried a tanh as a function (derivated 1/cosh^2 and still continuous) of p to implement my blending situation between two densities in one single fluid material. it works fine and the error message did not appear up to now. but have a look on your table generation as well.

All times are GMT -4. The time now is 08:03.