yuwen 0 Report post Posted June 2, 2022 Hi there, I am trying to mesh a wing using the meshengine. I want the mesh to be cosine distribution so it is more clustered towards the wing leading and trailing edge. I realize that with the mapping factor, if put in negative value, it will sort of give the result I want (see attached). And the larger the negative value , -1 --> -0.7 ----> -0.5, the more clustered the mesh will be in both ends. May I know what is the underlying formula that calculates the meshing spacing given the number of panels and mapping factor ? So I can understand better what goes on. Thank you. Yuwen Share this post Link to post Share on other sites
Mr. Heinrich von Zadow 9 Report post Posted June 2, 2022 Dear Yuwen, I'd have to ask a developer to take a look into the code to answer that question. However, I think it might be enough for you to know that the distribution (if not set to be uniform) is dependent on the parametrization of the underlying surface. Hence, by parametrizing the surface differently, you can freely control the distribution of cells. Cheers, Heinrich Share this post Link to post Share on other sites
yuwen 0 Report post Posted June 3, 2022 20 hours ago, Mr. Heinrich von Zadow said: Dear Yuwen, I'd have to ask a developer to take a look into the code to answer that question. However, I think it might be enough for you to know that the distribution (if not set to be uniform) is dependent on the parametrization of the underlying surface. Hence, by parametrizing the surface differently, you can freely control the distribution of cells. Cheers, Heinrich Dear Heinch, Thanks for the reply. The surfaces I generated is created from the Fsubsurface feature where I give four surface edges and the parent surface, and it will return me a four edge parametrized surface, ready for meshing. Hence, I dont have control over how the surface can be parametrized differently. If you look at the attached images. I have a wing that is connected to a fuselage. At the intersection where the wing is attached to the fuselage, the wing and the fuselage joint patches have same mapping factor, so their boundaries edges will be aligned. However, if you look at the surface in front of the fuselage joint patch, I need to manually tweak the mapping factor (-0.7) so that I can have smooth transition in the size of the mesh from the fuselage joint to the fuselage nose. It would be great if I can know for example, given the length and the width of a patch, and also number of panels in the u and v direction, what mapping factor I need to give so that I can make sure the size the mesh at the boundary of two neighboring patch, will have similar length and width, or aspect ratio. Again, if I simply set the distribution to be uniform, this won't be a problem, but if I want to have non-uniform distribution, I will probably need the equation in order to know what would be the right mapping factor to put in. Please let me know if you have any suggestion. And If it is not of too much trouble, could you help ask the developer to look for the equation ? If it is too hard to find, I guess I can create a simple rectangular patch and try to reverse engineering the equation on it. Cheers, Yuwen Share this post Link to post Share on other sites
Mr. Heinrich von Zadow 9 Report post Posted June 3, 2022 Dear Yuwen, the code is not that straight forward to explain. Mostly it's sine/cosine distribution functions, but there are quite a few additions here and there. Would you not rather need a topology like outlined in the attached project, anyway? Cheers, Heinrich gridDemo.cdb Share this post Link to post Share on other sites
yuwen 0 Report post Posted June 4, 2022 It looks lovely ! Thanks for sharing the demo. Have a great weekend. Cheers, Yuwen Share this post Link to post Share on other sites