[YASHA]

Hello everyone
I am engaged in numerical modeling of heat exchangers and it is critical for me to calculate the heat flow through heat exchanger inlets and outlets to find its heat balance. My way is to calculate enthalpy and integrate it on the inlet and outlet slices of each exchanger stream.
For the evaluation purposes, I firstly examine the Tecplot integration capability with a Fluent converged data for a sample, unfortunately results of Tecplot was thoroughly wrong. They has have different values from those of Fluent.
The worse is that Integration for total heat exchanged by one of fluids is not matched that of the other one. I check fluid properties, field variables and all other settings affect the results in Tecplot. Also I import the velocity-inlet and pressure-outlet boundaries directly from the Fluent instead of my own slices, but the results are the same. It is vital for me to do a so-defined integration and if someone has any such experience and come up with a solution to prevent wrong integration it will be appreciated.
Regards,
Y. Shafiei
K.N. Toosi University of Technology

[wsfowler]

Yasha,

This will be difficult to diagnose in a forum. I'd advise you to contact our support staff and we'll help you out. Please contact support@tecplot.com. Once we find a resolution we'll post it here for public record.

Thanks,
Scott Fowler
Tecplot 360 Product Manager

[YASHA]

Dear Mr. Fowler, Thanks for your attention.
Fortunately after a lot of effort I find the reason why the inconsistency arose, and it was originated from two separate distinct problems, may be interesting for you as a potential for the improvement of your software:
1-When the Case and Data files of a fluent solution are read in the Tecplot in the Nodal form, none of the averaging methods presented take the boundary condition into account, and boundary conditions are reflected only through separate boundary zones. In other words, a unite node can be a node of an internal volume element or a node of boundary surface element, and only in the former one the boundary condition can directly be sensed. When the Tecplot substitutes a quantity for the node, the boundary condition is forgot, so required value is calculated from an interpolation between internal volume elements centers. Therefore it may cause a great difference between interpolated velocities, temperatures, … and their direct enforced values from boundary conditions of type Velocity Inlet, especially in a coarse grid.
I think it can simply be resolved in the Tecplot. As I know when the fluent exports the results for the Ansys EnSight, this point is obviously noted.
2-In addition, my solution specialty is that all fluid and solid properties are supposed to be variable and temperature dependent, including specific heats, Cp and Cv. So the Enthalpy calculated by the Tecplot facilities in the form of h=Cp.T have led to a wrong result for me. A correct Enthalpy calculation consists of numerical integration of ∫Cp.dT, which necessitate implementation of a fully-described macro.