CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   wrong y+ (http://www.cfd-online.com/Forums/main/2532-wrong-y.html)

jurek August 26, 2000 05:23

wrong y+
 
Hy, we know, that y+ (standard wall-model) should be between 40 and 400 (about..). What happens, if it is much larger (say 2000) or much smaller ? Does it have only a local effect or is the whole solution "wrong", even if it is converged ? Jurek

John C. Chien August 26, 2000 21:01

Re: wrong y+
 
(1). The best place to start is the velocity plot of the law of the wall in traditional semi-log scale. (2). The law of the wall log-profile will grow linearly away from the wall in this scale. While the actual velocity profile with adverse pressure gradient effect will have the wake portion of the profile. (3). In other words, if the Y+ is too large and into the wake portion of the velocity profile, the matching with the log-profile is no longer valid. (4). The main flow region solution at this Y+ will not give you the log-profile behavior. So, matching its solution to the log-profile in the extended region will not give you the correct answer. (5). Take an extreme condition for example, for thin boundary layer, with the Y+ placed outside the boundary layer, the main flow region solution will indicate that the turbulence shear stress is zero there. Since in using the law of the wall, you have assumed that the shear stress is constant from the wall and in the law of the wall region, right away, you will have trouble in matching to the law of the wall log-profile. (6). In general, if the Y+ is outside the range, the main flow region solution will not give you the law of the wall log-profile solution, and matching will be incorrect. It is also hard to know whether it will converge or not. (7). By the way, it is also very important to show that the solution in the main flow region will show the law of the wall log-profile behavior with the particular turbulence model used. You can not assume that all turbulence models have this law of the wall log-profile solution behavior. (8). Also, the range of the log-profile will become very small when you have adverse pressure gradient. (9). So, it is very important to know that the matching range is a function of the adverse pressure gradient in the flow.


All times are GMT -4. The time now is 14:24.