[clau90]

Hi,
I am trying to model the transport of non reactive particles in a blood vessel, which I have defined with the additional variable function. The diffusivity coefficient for these particles is dependent on the shear strain rate, with a linear relationship defined as:

Deff = alfa * sstrnr + D

Where alfa is a given coefficient. I have initially tried to assign this expression to the kinematic diffusivity, but the software wouldn't recognize the sstrnr variable. For this reason, I have defined another additional variable (unspecified) for the diffusivity equal to the expression, and then assigned this value to the diffusivity coefficient.

The software is not giving me any error. However, when I try to run my definition file, I get an error:

Error processing expression 'Kinematic Diffusivity'.
The expression is invalid because:
Diffusiv is not available for use in this term

Error processing expression: Kinematic Diffusivity = Diffusiv
+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| C:\Program Files\ANSYS Inc\v150\CFX\bin\winnt-amd64\ccl2flow.exe |
| was interrupted by signal QUIT (3) |
+------------------------------------------

I have tried to run a steady state simulation assigning a constant value to the diffusivity, so that the variables could be initialised and started my new one from that, but it's giving the same error. What am I doing wrong? How can I fix this error?

Thank you,

Claudia

[Opaque]

You can create the following CCL snippet, and use it at run time

RULES:
  PARAMETER: Kinematic Diffusivity
     Dependency List = ANY
  END
END

This will bypass the error check you have found in your previous attempts; however, there may be a very good reason why the Dependency List does not allow all possible combinations (possible recursions ?). Review your results carefully to be certain the software is executing what you intend it to do.

[ghorrocks]

The main reason you cannot define some variables as functions of some other variables is because it leads to additional cross terms in the fundamental equations which CFX does not account for. This is in addition to reasons like recursions and implementation issues like that.

In this case making the diffusivity a function of shear strain rate would introduce lots of horrible cross terms linking diffusivity to the momentum equation. If those terms are significant then there is no way of doing this model in CFX (if you are lucky you might be able to add some of the terms as source terms, but source terms are a little limited in scope). If you are really lucky then these terms are small and can be ignored - then you are OK for using CFX with the work-around Opaque describes.

[clau90]

Thank you so much for the reply! It worked fine, for the moment it seems like it is not affecting the flow (I have compared the flow domain with previous simulations and there is no difference) but I will be careful in checking the results.

Thank you again,

Claudia