nCloth paper confetti - twirling

[Duncan]

This is an interesting problem. The air model used by nCloth will handle leaf like motions, but a small stiff rectangle of paper will spin like a turbine, and it is nearly impossible to get this sort of motion with the normal settings.

However I found a workaround you might be able to use. The pressure attribute pushes the cloth along the triangle normal. If one makes each bit of confetti out of two faces side by side but rotated 180 deg then one can use pressure to make the piece spin.

For the confetti example you could start with a plane(say with 10x10 subdivisions) then use detach component to break up the faces. Then select the faces and do transform component, scaling the faces to less than half their size( at least along one axis). Duplicate this and rotate it 180 degrees, translating it so the edges touch then combine with the original and do merge verts. One can then make this nCloth and use pressure to control the spinning.

Duncan

[ranxx]

thanks Duncan, I will try that. In my exploration of this problem I thought that the spinning in reality would be caused by a lift along one edge of the confetti and this could perhaps be represented in the nCloth solver by having the ability to map the drag coefficient of the surface (have one edge cause more drag than the other, thus causing an instability). I looked in the docs and unfortunately drag is not a paintable attribute, so I tried using mass but the effect was undesirable.

thanks again for taking the time to read my post, I really appreciate it and can't wait to try it out. Thankfully the art direction for the shot has changed and the flipping isn't so important, but I want to implement it anyway for my own understanding.

cheers

-ranxx

[Duncan]

If you find any good explanations of this phenomena post a link. I've been looking for one, but have been unable to find anything that explains the spinning of thin paper strips.

I don't thing that mapping lift along one edge would provide the rotation, but it is hard to know for sure. Mapping mass would be equivalent to mapping lift and drag together.

It might be something to do with the asymmetry of the leading edge (bottom most edge that is cutting into the wind) and the trailing edge. Perhaps vortices along the trailing edge reduce drag there and increase it near the leading edge.

Note that with the pressure based spinning I found that it is important to get a good balance on the tangentialDrag (the degree to which the drag is uniform rather than using the surface normal). A pressure around .2 combined with a tangentialDrag around .2 can work OK (depending on the drag setting). As one lowers the tangential drag the bits will tend to flop and fly around, while for higher values they will fall more uniformly as they spin.

Duncan

[Aikiman]

This maybe of some help...

Control of Leading-Edge Vortices With Suction.
http://www.dtic.mil/srch/doc?collection=t3&id=ADA320167

Reading quickly, the leading edge creates vortices and therefore suction which I assume would suck the rest of the surface in the direction of the vortex and bring the trailing edge to the lead and so it goes on. Could this be done utilising the velocity swirl generated in Maya fluids and taking that info back into nucleus?!

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.