December 31, 2016, 09:42
|
Problem with convergence using 2nd order scheme for multiple elements car wing
|
#1
|
New Member
Stan
Join Date: Dec 2015
Posts: 1
Rep Power: 0
|
Hello,
firstly, Happy New Year for everyone  I have a problem with convergence in calculations of multiple element front wing for car. I use k-omega SST turbulence model, velocity value 20m/s. First, I used first order for all equations exclude pressure (2nd order). After I got convergence (point 1 on picture) I changeg for 2nd order for everything. Unfortunately I can’t get converge solution using 2nd order scheme. Residuals stuck and only oscillates about some average value, but the Cl and Cd takes aprox. state values. I tried decrease under-relaxation factors (point 2 on the picture one), but it doesn’t help. I want to be implement the wall law so I must have y plus between values <30,300>, right? I calculated estimated y plus value and used an inflation option in meshing (the mesh is on the picture, I’m using ANSYS mesher). When I plot x-y plot of y plus value some values falls behind 30 (for plot I choose all surfaces of wing, is it right or maybe it should be done different?). I tried also double precision, but nothing changed, residuals got stuck. My question is, what can I do for getting convergence result, is this maybe problem of mesh (max. orthogonality about 0,78) or maybe even the residuals got stuck the solution is proper? Also how can I control the mesh for getting everywhere y plus value above 30? And the last question is that in the k-omega model wall law is implemented automatically? Please give me some advices.
Regards!
|
|
|