[dancfd]

Hello all,

I am in the process of setting up a case using k omega SST as a turbulence model (OF 1.7.x), and I am having trouble selecting boundary conditions. Menter's 1994 paper does not speak to k at the wall, but NASA () suggests it should be 0. This makes sense, since the k omega SST is a blend of k omega and k epsilon, using pure k omega at the wall. However, looking at tutorials that use k omega SST (wingMotion2D, verticalChannel, motorBike, mixer, throttle), this is not always the case. At first glance, verticalChannel appears to be the only case that follows Menter 1994 (low k throughout domain, 0 at walls), however wingMotion2D is much higher.

I understand that the k omega SST implementation in OF uses Menter 2001, and I have not seen it yet (it seems to be hard to find that conference paper), I cannot imagine it would differ on the selection of k and omega, especially k at the wall. Can anybody shed some light on this?

On a related note, Menter 1994 leaves some room for tweaking in the equations for k_inf and omega_inf: ranges of the power of 10 that can be chosen. Can anyone recommend which value is best? I plan to benchmark a few cases to see which is best and proceed from there.

Thanks,

Dan