[bobfck]

Hello guys,

First of all, sorry about my english.

I would like some help from you: right now I'm trying to use Fluent CFD to get Drag and Lift values of two wings, where the only difference between them is the washout angle.

I've tried using some youtube video setups, but I guess they are too simple for this kind of simulation.

Do you guys have any tutorias/online books/pdf's/anything that can guide me to do the mesh mostly?



Thank you very much for your attention and time.

[bobfck]

Anyone, please ?

[parvaz747]

Hi,
u have problem with meshing or reporting???

[Dave_88]

When you obviously have problems with meshing in general, why don't you consider a 2D-simulation before you start with a complex 3D-simulation? To mesh a two dimensional airfoil you will also find lots of explanations and tutorials - for example I suggest you to read about the c-grid meshing topology.

Otherwise you might consider describing your problem in detail and not just with a sentence or two. And images always help, too!

Regards
Dave

[bobfck]

Thanks for your answer.

I've done some 2D simulation with Profili 2D. But I got to agree with Dave that problably I should take a step back.

I'm going to describe the major doubts I have right now:

Meshing:

1) Geometry: right now I'm using tetrahedral geometry, but I've been told by some people that hexahedral could give me better results. Which is your opinion?

2) Inflation: I've created a really large enclosure for the wing (Size>5xWing spam), is there any advantage to use inflation at any specific point?

Setup:

1) Materials: My wing is made of styrofoam with a thin layer of monokote. How could I created materials which such properties to approach those I'm using?

2) Boundary conditions: I'm only applying boundary conditions to the inlet_velocity. Should I applying to all the enclosure?

3) Viscous model: I'm using spart-allmaras m. Is there any other I should take a look? I'm reading a bit about all of them at http://turbmodels.larc.nasa.gov/index.html, but tips are always welcome.

My major problem right now is the Cl values I'm getting, they are way too low when compared to the Cd.

Right now I'm trying a new simulation. When it is done, I'm going to post some pictures about the Bondary conditions and the meshing I'm using.

Thanks again for your patience
Felipe C. K.

[Dave_88]

Hi Felipe,
as for your answers/doubts/clarification:

Meshing -
1: A hexahedral mesh will almost always give better results in comparison to tetrahedral mesh. You will use a tet-mesh if you want to reduce the mesh-procedure - the main advantage is, that you can almost automatically generate a tetrahedral mesh - the main disadvantage is that this will cost you numerical accuracy (mostly due to the angles). Bottom line: if you are able to generate a hex-mesh, you should do so. If you have troubles with the procedure, either check tutorials, or perform a first shot with tetrahedrals.

2: I cannot get what you want to know there precisely. You might add a picture that explains, what you mean in detail.

___

Setup -
1 Materials: this depends on the simulation you plan to perform respectively which effects you want to calculate. For example, if you need to know boundary layer development, you need to know surface properties of your airfoil. Make sure what exactly you want to know and which properties are beeing needed for this.

2 BC: I think this is your major problem. Why would you only declare an inlet? What about your farfield, your airfoil, outlet and so on? Of course you need to put boundary conditions on every part in your model. For example wall (including surface roughness if needed), outlet (counterpressure if needed), maybe pressure- or mass-flow-inlet instead of velocity-inlet and so on. Also consider density or pressure based solver - depending on your Mach-Number, respectively on if your calculation is compressible.

3 VM: Personally I have no experience with SA-Model, but what is your value for the turbulence parameter (at velocity inlet)? I recommend to use the k-omega-model with Shear-Stress-Transport specification. This was very useful for my airfoil simulations and is also well established. You can read lots of publications on this topic. You also need to make sure, that your mesh quality fits the turbulence model. For example you need to make sure, that your y+ will be in the correct range and growth ratio.


Additionally: do you have any results that could verify your results?

Regards
Dave

[bobfck]

Hi there Dave,

"Additionally: do you have any results that could verify your results?"

I have to be honest with you that the only aerodynamic data I have is the 2D-profile Cd/Cl, which is a result I definitly could use to verify my results.

The UAV currently using this 3D-airfoil is flying very well, but it needs some optimization. The work I'm right now trying to do using this simulation is a Cd/drag comparsion between two identical 3d-airfoils, which the only difference is the washout angle(4°/6°).

Some details about the UAV:
- Flying speed: 25 m/s;
- Wing S: 21dm² (each);
- The 2d-airfoil data I got has a Reynolds between 40.000-100.000.


a) I'm going to try hexahedral geometry meshing later on my desktop.

b) Right now I just need some basic Drag data which I could trust, so boundary development shouldn't be a major problem now.

c) Boundary conditions: do you have any links/videos/etc I could use to learn?
I definitly agree that the more boundary conditions we have, the best the result will be.

d) VM: I gonna give it a try.



Thanks again Dave,
Felipe.

[LuckyTran]

Felipe my suggestion is to do some Fluent tutorials. You have some very basic problems that need to be resolved before even worrying about the answers to the actual questions that you asked.

If you don't want to do tutorials, then I suggest to play (and I mean PLAY) with Fluent and make it output a drag coefficient. After you have done this at least once, then you can come back and worry about getting the answers to your questions like meshing, models, etc.