While Reynolds Stress Models appear more comprehensively formulated to capture anisotropic turbulent eddies, I keep seeing remarks here and there on the internet that they aren't necessarily as well formulated numerically, as Eddy viscosity models are.

Yes RNG K epsilon model

Thanks a lot. Any suggestion on where can I have further information/ readings?

IDK man, I've just gathered this from forums, application papers and such.. @User anyone?

Thanks :). at least I have a start point to explain the discrepancy between the two models

Is it wall-bounded turbulence?

If so: using RSM is a waste of time as you can find in many papers

If your simulations involve a lot of flow separation, then kOmega SST family is better than k-epsilon family of turbulence models

Nope, it is not wall-bounded turbulence. It is about the junction vortex in the downstream wake. Did try KOmega SST but it is not close to the experiment measured. RNG KE is way closer

Can anyone help explain what is happening at my top and bottom of the mesh? I am trying to just use the symmetry patch to "ignore" the sides

symmetry boundary condition as well

Please see #810161483903401984. It seems that you are not using symmetry at top and bottom

Hi I am using k-epsilon turbulent model for simulating pressure induced flow via simplefoam While simulating bounding k and epsilon are coming very huge and ultimately stoping simulation

How do I handle this and we're i can get more info on this?

Almost definitely inappropriate boundary conditions or poor quality mesh

A simple diagram of your set up and boundary conditions will allow people to help you

i cannot share my geometry but i will give you a analogy similar to that.(sorry for this) this is a cross-section of a cube from which a cylinder is cutted out. the flow here is perpendicular to the cylindrical axis. For inlet there is pressureinletvelocity condition for velocity and for pressure zeroGradient with p=5000 For outlet there is zeroGradient for both velocity and pressure

i have tried changing relaxation factors and initial value of K and epsilon but nothing worked

also for meshing in checkMesh it is showing ok

Do you hav eany Dirichlet pressure BC?

try changing your pressure BC at inlet to fixed value

ok i will try that

also i was thinking to try directionmixed for velocity

but my flow in original model is non orthogonal so i dont know how to work that out

> pressure zeroGradient > zeroGradient for [...] pressure

Not a well posed problem if that is true

i read that openfoam have stability issues with pressure induced flows

Every solver has problems if your solution is definitely not unique

which solver would you recommend for this?

i am using simplefoam