FFD Results for Winglet Generation

Yes, it works for the 3D viscous grid that Lana uses for her optimization cases! On the left side we can see the FFD hull and the CFD surface grid; and on the right side we see a slice of the CFD grid. The grid is quite good, but what we want to make sure is that grid stays good after the grid movement. So the test case is to generate a fairly large winglet.


The graphic sure look nice, It's actually not that easy to tell whether the grid is of good quality or not. To do that, we first visually take a look at grid lines at the wingtips of the original and deformed grids. The deformed grid is on the left and the original is on the right. We can actually see that the grid isn't too distorted.

slice-final slice-original

But we really have to find a more qualitative measurement of grid quality. Thankfully, my former colleague Jason Hicken had this wonderful idea of looping through each node and measuring the orthogonality of the grid lines each grid node. This is what we got from the winglet case:


So, there are the expected loss in the highest-quality nodes on the right, but overall, the numbers are quite encouraging.