Learning Forum

[vidyadhar.kpm]

Hello,

I have done post processing of fluent case and data in "Paraview". I have done a basic operation using "clip" to remove certain cells.

I can export the data from paraview in any of the below types:

CGNS (.cgns)

Comma separated (OR) Tab Delimited files (.csv, .tsv, .txt)

Exodus II file (.e, .exz, .exzv2, exo, .exo II, .exoii, .g, .gen)

VTK MultiBlock (.vtm)

xdmf Data File (.xmf)

xdmf 3 Data File (.xmf)

 

My question: I want to do meshing and naming of the surfaces in ANSYS from the domain obtained in any of the above formats from Paraview.

I am trying ICEM which reads .cgns format.

I request suggestions on any other better way.

 

Thanks in advance!

vidyadhar

[DrAmine]

You have already one way.

But why are you doing that?

[Rob]

Can you export VRML format (post processing)? 

[vidyadhar.kpm]

Hello Amine,

I have done VOF simulation to obtain a static interface between a liquid and vapor. Now I want to perform Evaporation from the interface. I have written an UDF for that.

A) I am modelling (i) the evaporation mass loss as a negative source term in the liquid cells near the interface and ii) evaporation heat loss by way of applying a heat transfer coefficient. To apply heat transfer coefficient, I have to delete vapor cells above the interface and make the interface as a WALL. Since, I can not delete cells by iso-surface of VOF=0.5 in Fluent, I am deleting the vapor cells in Paraview and trying to export the residual domain back to meshing software.

If I retain vapor cells and apply the source terms both in liquid and vapor cells along with an energy source term in the liquid cells: The interface moves since liquid evaporates and also I am unable to validate my results.

Thanks & Regards,

Vidyadhar

[vidyadhar.kpm]

Hello rwoolhou,

I can export in VRML format. But, I think paraview gives images in VRML format. I need to check this again.

 

Thanks & Regards,

Vidyadhar

[DrAmine]

Still mot getting why you want to remove the cells in Fluent: are the vapor cells upsetting you? Or is it a challenge? That is normal that the interface would move: that is a dynamic response: mass transfer across the interphase is difference between interfacial normal velocity and phase normal velocity..

I guess you can export to STL the model from the Fluent (if you are able to separate the fluid zones) and use that as input in pre-processor. Same task can be done in CFD-Post

[vidyadhar.kpm]

Hello Amine,

I am interested in maintaining static interface under evaporation conditions. I am applying mass source/sink terms in liquid/vapor cells and energy source in liquid cells in the region where 0

If I chose large time step, the solution diverges giving floating point exception. Also, the interface moves quickly.

If I chose small time step (~1e-10 s) such that I get convergence in each time step, even after running for a large number of time steps, I do not see constancy/steadiness in the average interfacial temperature.

I am using the following parameters:

VOF parameters: Explicit Formulation with default Courant number of 0.25; Implicit Body Force formulation.

Liquid phase is water and Vapor phase is water-vapor. I am using PISO, Geo-Reconstruct, GreenGauss Node based for gradient discretization; PRESTO! for pressure;Third Order MUSCL for momentum and Energy discretization. First order Implicit Transient formulation.

UnderRelaxationParameters: 0.25 for pressure and momentum; and default for others.

I request your suggestions to speed up my solution.

 

Thanks & Regards,

Vidyadhar