[Nek5000-users] what parameters for a moving wall induced velocity problem?

[nek5000-users at lists.mcs.anl.gov]

Dear Nek5000 users,


I'm looking forward to enforce a moving wall boundary condition. My question is mainly focused on the logical switches I should turn on and off.

Until now, I managed to get the mesh moving by enforcing xm1 and ym1 in a function that is called by nekadvance() before the fluid solver. Since I know my time dependant mesh moving function, I also compute wx and wy. A recomputation of all Jacobians,etc... is then done and is followed by a topology test. All the following logical flags (IFMVBD, IFMGRID, IFUSRMV) are turned off and switching them on doesn't seem to change anything in the further described problem.

This method seems to work as long as I specify an initial velocity but it fails when I want to simulate the fluid flow induced by the wall movement itself ( therefore when I don't specify the initial velocity). I only get some random velocities values around 1e-20 despite the outer walls are moving significantly, and this regardless to the viscosity, density, or wall movement velocity. Is this phenomenon intrinsic to the code formulation and do I need to specify an initial value different from zero?

What appropriate configuration of the .rea file would you recommend to get a wall-induced fluid motion? Has anybody already succeded in doing such a thing ?

Thank you in advance for any help you could provide

Best regards,

Michael Bruckner