[casesam]

Hello all,
I am simulating 2D water flow through pipes that are ~100 microns in width (microfluidics). In some of the channels I have placed obstacles (cylinders). I solve the system using simpleFoam with turbulence turned off. so basically steady, laminar, incompressible flow.
My solutions converge even with very low tolerances.
I have done many simulations and have the results. What I want to verify now is weather or not there could be Karman vortex streets (vortex shedding) behind these obstacle.

So my questions, I'm still very new to CFD.
1. If the flow through my system were truly unsteady, would simpleFoam even converge to a solution?
2. To check if my flow is unsteady, can I run icoFoam with the initial conditions set as the final results from simpleFoam and observing whether the flow changes?

Thanks very much!

[PHS-]

Hi,

For your first question :
I think it depends on the case.
I already ran flows over cylinders and other bluff-bodies with steady solvers and got converged solution even though these flows are of unsteady nature with alternating vortex shedding. You just get a wake similar to the mean flow with a larger recirculation region.
On the other hand, (for a complete different type of flow) I was once unable to converge a steady simulation and unsteady solver gave me good results.

For the second question:
yes, if the flow is unsteady with von Karman vortex streets, you should see alternating vortical structures appearing after a while. Initialising the unsteady solver with steady solution will save you a lot of computational time.
From my experience, in the case of 2D cylinders if you start the unsteady solver from the initial uniform velocity flow-field it will first converge to the steady solution with a large wake that will then start oscillating until you get von-Karman vortex streets.

[casesam]

Great!
Thank you very much for the clear explanation.