A question on the periodic-hill case

[esaka-forever]

Hi,
I have tried to run a DNS for periodic hills using this code. Unlike the other cases, there is no example code for the periodic hills for which IBM should be used. So I just copy the input.i3d from the cylinder example case (which I believe IBM is also used), and change the flow type to period hills. After I run the code, the velocity diverges in the first step, regardless of whether I set iibm=1 or 2 (iibm is the Flag for immersed boundary method) or how I adjust the timesteps or Reynolds number. I highly appreciate it if anyone can offer any help or hint on what else I should change.

[slaizet]

Hi,
We will aim to provide an example code for the periodic hills. There are few changes to be made with respect to the cylinder case, for example the boundary conditions & the initial condition. I will create an issue so that we can start working on an example code for the periodic hills.
Best,
Sylvain

[esaka-forever]

Glad to know that. Thank you!