[fluidmechanics]

Dear All

I am going to consider variable fluid properties as a function of temperature and pressure as a real gas model. These properties are defined as a polynomials at certain temperature and pressure and must be interpolated at any temperature and pressure of flow domain. I am really confused how to do this in Ansys CFX.

Any help or advice is appreciated.

Sincerely

[ghorrocks]

Please describe the polynomials and interpolation you intend to use.

Also - make sure you have done the CFX tutorial examples so you know how to use CEL expressions, as it is probably going to use CEL expressions.

[fluidmechanics]

Hi Glenn

Thanks for your reply. Following your suggestion, I studied for CEL expressions and tutorials. By the way, I am going to evaluate CFX capability for a high velocity flow over a sphere in order to get true wall heat flux. The considered Mach is 10. I have imported compressibility factor and specific heat flux and transport properties as a table which are a function of temperature and pressure. So in order to import the table as 3D, the third independent variable is set to an optional constant (e.g. 0). A pure material is user defined with a density defined as expression including compressibility factor. Cp and transport properties are as CEL expressions, too. I have defined a velocity expression in order to increase farfield slowly and step by step. I have some questions:

- Is it necessary to set reference state parameters?
- What about table generation? I have set min and max values of T & Pabs such that flow field corresponding values are within this range.
- What about error tolerance? Default 1e-2 value causes problem. Increasing it to 0.5 yields in better convergence up to Mach=7, but higher Mach number fails the simulation. How high can this error tolerance be?

In out file the following message are printed:

User Defined Monitor Information |
+--------------------------------------------------------------------+

Monitor Point: Monitor Point 1

+--------------------------------------------------------------------+
| ****** Notice ****** |
| |
| Adaptive table generation failed for the material property: |
| |
| Static Enthalpy |
| |
| This may have occurred because the equation of state or specific |
| heat capacity function contains regions with large derivatives |
| or because the error tolerance is too small. Execution will |
| continue with a uniform distribution using 100 points in |
| each coordinate direction. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| ****** Notice ****** |
| |
| Adaptive table generation failed for the material property: |
| |
| Static Entropy |
| |
| This may have occurred because the equation of state or specific |
| heat capacity function contains regions with large derivatives |
| or because the error tolerance is too small. Execution will |
| continue with a uniform distribution using 100 points in |
| each coordinate direction.


What should I do?
I am a fan of CFX. I want to implement such flow conditions in CFX. That's your kind if you can help me.

Sincerely,

Mike

[Opaque]

Would you mind describing which material properties you are providing?

In particular, thermodynamic properties, say Density, Specific Heat, else?

What kind of variation are you accounting for? T only? T and P?

T and P variations are tricky and not arbitrary (see the documentation for details)

[fluidmechanics]

Hi my friend.
Density, specific heat, momentum and energy transport coefficients which have been considered as a function of both T and P.

[Gert-Jan]

I doubt if CFX is suitable up to Mach =10. Did you check support on this?

[ghorrocks]

Most gasses have significant dissociation at Mach 5. You are double that speed. CFX does not have a plasma model so I don't think CFX is a suitable code for this. Are you sure CFX is going to be able to handle your application?

[fluidmechanics]

Originally Posted by ghorrocks

Most gasses have significant dissociation at Mach 5. You are double that speed. CFX does not have a plasma model so I don't think CFX is a suitable code for this. Are you sure CFX is going to be able to handle your application?

Hi Glenn.
Yes, I know about dissociation and all about real gas behavior of air in high temperature. I have progressed a bit and now my question is how to have both specific heat and static enthalpy as functions of both T and P in Cfx. I have studied some about RGP format in CFX but as I see there we should define critical pressure and temperature. Is there any way to define both cp and h variable?
Sincerely