Learning Forum

[RezaKdr]

Hello,

I am modelling force convection heat transfer in a channel using energy and momentum source terms (qi and fi, respectively). I defined the source term using the Expressions option in Fluent 2020R2. The flow is turbulent and the turbulent model is realisable k-epsilon with enhanced wall function. I have no problem with the momentum source term but when I activate the energy source term the following error will appear:

Fluent Error: chip-exec: function "qi" not found

the energy source term is the momentum source term* x-velocity, as

Velocity.x*fi

I tried different scenarios, such as testing other turbulent models, considering other sorts of energy source terms such as a constant number; however, I got the same error. I also tried the series and parallel simulations, and also different solution methods (SIMPLE and Coupled), but still getting the same error.

I would appreciate it if you help me with this issue.

[Rob]

It usually means a value that is being used is either not defined or not present in the model. With UDFs it's occasionally caused by model changes removing a required field (eg a turbulent setting from a UDF is removed by switching to laminar) and the UDF then being unloaded: the solver still looks for the UDF but doesn't actually need the value.
How are "qi" and "fi" defined?

[RezaKdr]

I used expressions to define them, I didn't use UDF.
"fi" is defined as:
-0.25*Density*exp(- (y^2)/2)*Velocity.x*abs(Velocity.x)
and "qi" is defined as:
Velocity.x*fi
The issue is that even if I define a constant number, say 2 W/m^3, in expression and use it as a heat source, I will get the same error.

[Rob]

Use a longer label to check "fi" and "qi" aren't a defined variable in Fluent.

[RezaKdr]

I have changed the labels but still, I am getting the same error.
Error: chip-exec: function "Energy_source" not found.

[Rob]

Does the model run?

[RezaKdr]

Yes, it is working okay with the momentum source term. However, when I activate the energy source term, the model runs but with error. It runs without taking the source term into account.

[Rob]

Set "fi" as a fixed value and see what happens.

[RezaKdr]

The same error repeated

[Rob]

Change "qi" to be "qsomething" and see what happens. There are a few hard coded macros but no list.

[RezaKdr]

I have changed it to "qChanged" and the same error was repeated.

[Rob]

Still an error for qi or for qChanged? If it's the former there must be a setting that's got stuck in the solver.

[RezaKdr]

The same error is for qChanged.

[Rob]

To check:
fi = momentum source
qi = energy source
Did you check the units? Expressions need units, unlike UDFs which assume the values going into and leaving the UDF are SI and correspond to whatever you're using them for.

[RezaKdr]

Yes, that's correct.
I checked the units. Everything is fine.

[RezaKdr]

I finally found the reason for that error. This error appears when a heat source defined in the "named expressions" is used with a periodic boundary condition. The error disappeared when I switched from periodic boundary conditions to velocity inlet and pressure outlet.  No issue was found when the heat source is plugged in using UDF.