CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Mesh a twisted wind turbine blade (http://www.cfd-online.com/Forums/main/15005-mesh-twisted-wind-turbine-blade.html)

Laurenjiongo March 31, 2008 13:49

Mesh a twisted wind turbine blade
 
Hi everyone

Has anyone know how to mesh a twisted wind turbine blade? I'm facing strong dificulties in GAMBIT to do that. The blade is made of NREL S809 airfoil profiles.

http://wind.nrel.gov/amestest/BladeDesign.pdf

I'm trying to use boundaries layer without success. When I turn on internal continuity, the volume mesh fail. When not, I have a message asking me to turn the use_3D_Boudary_layer variable off but the result is unchanged.

I know that it's very useful to split the computational domain around the blade in several blocks, but since two weeks I'am not trying hard without getting to the right blocking strategy.

Any advice will be received with a great pleasure. Thanks.

Ps. 'Sorry for my english, my translator is doing its best.' I have prefered to post this there instead of the Fluent forum because I know a little about ICEM.


eddy April 5, 2008 00:45

Re: Mesh a twisted wind turbine blade
 
I recommend you change to use AutoGrid of numeca software to mesh

laurenjiongo April 7, 2008 05:44

Re: Mesh a twisted wind turbine blade
 
Thanks for your answer

Sorry, I couldn't change. I could only use Icem CFD or Gambit.

bskim April 15, 2008 01:17

Re: Mesh a twisted wind turbine blade
 
I recommend the ICEM CFD HEXA for grid generator and it should be better to get more accurate results such as blade torque, root flap bending moment, edgewise bending moment. But, creation of the block files with high quality requires very high skill. Use o type grid in around the blade. It is also difficult to generate hexa block with high quaity abound the blade tip and hub. It is very complex problem and also actually i don't understand what do you want to know. I have ever been to do a turbulence model validation study on the NREL Phase 4 wind turbine. that time, i use ICEMCFD and CFX. Question more about what is the problem of yours exactly.


Laurenjiongo April 15, 2008 08:07

Re: Mesh a twisted wind turbine blade
 
Thanks for your time,

Actually I want to study how the wake generated by an upwind turbine affect the performance of a second one. The aim of this work is to build graphs showing the evolution of power and structural loads with the separation distance. That's why I need to generate meshes with sufficient quality to obtain good estimations.

I'm using CFX to generate an tetrahedral mesh but It still too coarse near the blade and my results (torque) are at least 3 times lower than experimental values. When I want to refine It, the volume mesher crash or when It works I obtain many high skew elements and very fine mesh where It's not need.

As you suggested,I will try O-grid. If I face some issues I will not hesitate to post again.

PS: Sorry for my english.


All times are GMT -4. The time now is 06:57.