Dear Jan Gaertner, Dear tobias martin,
I hope the message finds you well. First, I would like to appreciate your time and efforts on WENO schemes. I am using sonicFoam solver with below setting. I am using hybrid mesh which is combination of hex core and poly hederal meshes. When I use these setting my simulation seems well until few centiseconds then flow behavior changes and unconditionally it becomes unstable. but when I change the default to cellLimited leastSquares 1; I lose accuracy and wrong solution. Would you mind informing me what WENO schemes can I use for grad scheme high accurate and non oscillatory. May I use Gauss WENOUpwindFit 3 1; for all divSchemes, even to div(((rho*nuEff)*dev2(T(grad(U)))))?
ddtSchemes
{
default backward;
}
gradSchemes
{
default cellLimited leastSquares 0.9;
}
divSchemes
{
default none;
div(phi,U) Gauss limitedCubicV 1;
div(((rho*nuEff)*dev2(T(grad(U))))) Gauss linear;
}
I appreciate your time and look forward to hear from you soon.
Dear Jan Gaertner, Dear tobias martin,
I hope the message finds you well. First, I would like to appreciate your time and efforts on WENO schemes. I am using sonicFoam solver with below setting. I am using hybrid mesh which is combination of hex core and poly hederal meshes. When I use these setting my simulation seems well until few centiseconds then flow behavior changes and unconditionally it becomes unstable. but when I change the default to cellLimited leastSquares 1; I lose accuracy and wrong solution. Would you mind informing me what WENO schemes can I use for grad scheme high accurate and non oscillatory. May I use Gauss WENOUpwindFit 3 1; for all divSchemes, even to div(((rho*nuEff)*dev2(T(grad(U)))))?
ddtSchemes
{
default backward;
}
gradSchemes
{
default cellLimited leastSquares 0.9;
}
divSchemes
{
default none;
div(phi,U) Gauss limitedCubicV 1;
div(((rho*nuEff)*dev2(T(grad(U))))) Gauss linear;
}
I appreciate your time and look forward to hear from you soon.