|
[Sponsors] |
September 23, 2016, 14:06 |
Trouble modelling a jet fan
|
#1 |
New Member
Jose
Join Date: May 2009
Location: Tokyo
Posts: 4
Rep Power: 17 |
Hello FOAM users
I have little experience in the use of openFOAM (I have done some tutorials and some cases based on incompressible tutorials, mainly with simpleFoam) but have been reading a lot about setting a fan boundary condition for the correct modelling of a jet fan without achieving a solution to my problem. I am simulating a scale model of a jet fan inside a square duct but I have some doubts about the cyclic boundaries and the process of implementing the BC in cyclic BCs. The fan is modelled as a rectangular box in which one face is the fan inlet and the opposite face is the fan outlet. Both faces are separated 50 mm and flow in the duct is flowing in the positive Z direction. I followed the next procedure: 1. I Imported my mesh using fluent3DMeshToFoam 2. I created my createPatch file in which I set the fan inlet and fan outlet as cyclic patches (defining for the fan inlet: transform translational; separationVector (0 0 50e-3); and for the fan outlet: transform translational; separationVector (0 0 -50e-3); Is it correct to put minus (-) for the outlet patch? 3. I ran createPatch and reviewed the modified boundary file assuring the correct creation of fan inlet and outlet patches. 4. I made the setup of the initial values folder (0) for U and p as: U file FoamFile { version 2.0; format ascii; class volVectorField; location "1"; object U; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // dimensions [0 1 -1 0 0 0 0]; internalField uniform (0 0 0); boundaryField { fan_wall { type fixedValue; value uniform (0 0 0); } inlet { type zeroGradient; } outlet { type zeroGradient; } wall { type fixedValue; value uniform (0 0 0); } fanin { type cyclic; } fanout { type cyclic; } } and the p file: FoamFile { version 2.0; format ascii; class volScalarField; location "1"; object p; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // dimensions [0 2 -2 0 0 0 0]; internalField uniform 0; boundaryField { fan_wall { type zeroGradient; } inlet { type fixedValue; value uniform 0; } outlet { type fixedValue; value uniform 0; } wall { type zeroGradient; } fanin { type fan; patchType cyclic; jump uniform 0; value uniform 0; jumpTable polynomial 1((100 0)); } fanout { type fan; patchType cyclic; value uniform 0; } } 5. The case is solved by simpleFoam using k-Epsilon turb. model 6. Reviewing the results I see that the flow from the fan is not going in the jump direction but it is going out each patch (Like if the fan were a source) See the attached image for clarity. I have not used baffles and I do not know if I have to use them. From everything I have read in the forum I have the idea that I do not need to use baffles because I have already defined cyclic patches. I have been trying for 2 weeks now and I can not get the flow going from one side of the duct to the other. Any comments or ideas please???Thanks for your attention! |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
"Hand" calculation jet fan in the garage | mageno | Main CFD Forum | 3 | August 7, 2021 20:09 |
Fan boundary condition for modelling turbine | er_ijaz | FLUENT | 2 | October 6, 2014 12:56 |
fan driven flow, Fan BC validation (getting lost) | soonic | OpenFOAM Running, Solving & CFD | 0 | July 7, 2013 20:16 |
[GAMBIT] Modelling CPU radial heat sink fan | r.sirait | ANSYS Meshing & Geometry | 3 | June 4, 2012 02:57 |
Modelling Ceiling Fan | suitto | FLUENT | 1 | March 21, 2000 12:46 |