CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Defining time step size (https://www.cfd-online.com/Forums/main/6864-defining-time-step-size.html)

 kilcoo316 November 27, 2003 10:11

Defining time step size

OK, this is a simple problem, that I haven't got my head round,

I have a test case to model, which has a characteristic height 0.066m (as used for E in K-E). For running the URANS sims for vortex shedding, in kelvin-helmhotz (K-H) form. I have taken the size of the first K-H instability, and calculated the time the free-stream velocity takes to pass over it, gave it 30 steps over the motion, which results in a timestep of 0.000006 sec. Is that too small?

The Cd and Cl traces are extremely unsteady, varying between 10 and 0 for Cl( high as 30 - -15 near start) and 5 and 0 for Cd (30 - -5 near start).

Length of 1st K-H approx = 0.005423m (0.000180 sec/periodic motion) Length of test case = 0.371m vel = 30m/s (Re approx 9*10^5)