3d Wing in ICEM : "Imaginary Mesh Extension After Wing Tips"
 

Hello,

Q1. Can someone please explain on why do i see in tutorial videos on ICEM CFD where people are extending the blocking after the wingtips.

Eg1. https://www.youtube.com/watch?v=UO8RJWsIV40&t=883s (21.06/31.57)
Eg2. https://www.youtube.com/watch?v=LmRiNeGfx9w&t=5s (9.48/49.18)

Q2. Should this approach remain the same if i have multiple winglets?

Q3. Should this be extended all the way if i have a fairly large "semi bullet shaped farfield" where Radius : 10x Wingspan, Wake 25x Wingspan.

Any useful insight will be much appreciated.

Ans 1 : Simply the intended approach is a 3d grid. In other words a 2d grid being extended along the 3rd axis. Theres no way to remove all the densed distribution in the spanwise direction, thus it needs to be extruded all the way to the end of the farfield.

The unnecessary extension can be limited with the introduction of a hybrid mesh where the blocking is done up to the tip or slightly beyond. Then, a tet/hex mesh can be used after to diffuse the highly densed region outward in the spanwise direction.

This approach however is not yet tested so it cant be compared on which method gives better result.

Ans 2 : Yes, a good example was given in the 2nd link (open trailing edge). He used a double C grid in the chordwise direction which was useful to counter the "triangle shaped" block right after the trailing edge. This in my opinion gives a better mesh when compared to the typical approach seen in many tutorial videos.

As for split winglet problems, a double C grid can be used in the spanwise direction after building it in the chordwise direction.

Ans 3 : This approach i believe is not economical and slighly tricky to block. One should try with a typical C grid such in the case of Simon's 3 part youtube tutorial only difference is that it is extruded spanwise.

P/s : I did a 2d comparison between a normal cuboid shaped farfield and an extruded C grid. Better results was seen in the C shaped.