# Tumblewing Meets CFD

Although a sheet of paper can be made to fly by spinning, it isn't always reliable. By adding winglets and flaps with 4 simple folds we can improve the stability of the wing and in so doing produce a tumblewing. How do the aerodynamics of a tumblewing compare with a sheet of paper? It just so happens Computational Fluid Dynamics (CFD) can help, just as it did with the simulation of the spinning sheet of paper.

CFD Simulation of a Rotating TumblewingPressure iso-surfaces at 90 degrees

### CFD Model

The only difference in the model for the tumblewing compared to the paper sheet was the more complex geometry, which was fully enclosed in a cylindrical Moving Reference Frame (MRF). It only required a half model due to symmetry.

CFD MRF Symmetry 3D Model of a Tumblewing

### Results

To obtain flow results at various angles of the tumblewing I used automation via a Python script to rotate the geometry and run a series of CFD simulations. The lift and drag values are collated in the plots below.

Velocity Vectors for Rotating Tumblewing at 0 Degrees

Velocity Vectors for Rotating Tumblewing at 90 Degrees

Pressure Iso-Surfaces for Rotating Tumblewing at 0 Degrees

Pressure Iso-Surfaces for Rotating Tumblewing at 90 Degrees

Lift for a Rotating TumblewingCompared to a rotating paper sheet

Drag for a Rotating TumblewingCompared to a rotating paper sheet

### Conclusion

Broadly speaking there is a similar distribution of lift and drag relative to the angle of the tumblewing and the paper sheet. However, throughout most of that range the tumblewing has lower lift and lower drag. In a pure performance assessment the paper sheet has higher lift and when it flies straight it tends to fly further than the tumblewing. However, what the tumblewing lacks in lift it makes up for in stability.

### Notes

• Build your own tumblewing with instructions from Science Toy Maker.
• The CFD simulation was performed in Caedium Professional using the MRF option for the incompressible, steady-state RANS solver, and the k-omega SST turbulence model.