[comet]

Dear SU2 Community,

I have come across a discrepancy in the transient simulation of a harmonically oscillating airfoil, and I don't know whether this is an input error on my part or whether something is wrong in the solver.

Attached, you find the CL response of a harmonically oscillating airfoil using deforming grid with the pyhton interface. The only difference is the multigrid level 1V vs 2V, which leads to entirely different results. Convergence criteria and timestep are identical.

In addition, for comparison, the solution of a rigid body translation (same grid, BC, and solver settings) leads to a similar result as that of deforming grid 1V independent of the multigrid level.

Has anyone had a similar experience or knows what I am doing wrong?

I also noticed that ROE discretization is unstable with the deforming grid option, but not with rigid mesh translation. At the moment, I can only obtain solutions with JST.

Every hint and exchange of experience is very welcome.