hi,

i am writin a code for incomp. k-e high reynolds number model. my two questions are 1) y+ is evaluated as y+ =rho*Ut*delta/viscos. what is viscos. is it addition of laminar and turbulent value of viscosity or only the laminar value. 2) what should be ideal value of y+ (i mean first node distance from boundary ) i know that it should be between 11.6 and 400. but what is the range of y+ values where most of the production of turbulence takes place? thanx

(1). Check the previous messages on wall functions. (2). Check out the books by Hinze (Turbulence) and Schlichting (Boundary Layer Theory) or the like, for Y+ definition. Y+ is used to scale the velocity profile in semi-log scale to give the famous log-law of the wall profile. (3). There is no such thing as the laminar viscosity. (4). The viscosity used is the molecular viscosity of the fluid. (5). In the semi-log scale, the law of the wall is represented by a segment of straight line. You can pick a point on the straight line, on the left hand side of the region to the wall, this analytical law of the wall equation is used. On the right hand side of the region, the law of the wall segment of the straight line is computed by the cfd code. The unknown is the wall shear stress. So, it must be derived iteratively. (6). SO, the matching point can be anywhere on this segment of the straight line (log-profile). And the converged wall shear stress will determine the proper matching of the right hand side straight line with the left hand side straight line. The end result is the complete line segment of the law of the wall joined smoothly at the matching point. (7). In some models, the tke (k) near the wall is used to derive the wall shear stress. There is another book you can check out, Turbulence modeling for cfd by David C.Wilcox. Check the resources section for his address.

1. In y+ formula, viscos is molecular viscosity (remember y+ is a definition of near-wall turbulent Reynolds number).

2. Normally, the near wall turbulent region is divided into viscous sublayer, buffer zone and log-law region. Numerically, one can ignore the buffer zone. So in viscous sublayer, y+=u+; in log-law region, u+=log(Ey+). Thus the minimum y+ can be derived by setting y+=log(Ey+) - that is you can sovle this simple equation (iteratively) to get starting point of y+.

sorry, in the u+ formula, it should be u+=1/k log(ey+), where k is normally 0.42.