[ArathoN]

Please can someone help me here? I'm simulating the backward facing step with and without the fully developed flow at inlet. My problem is i'm getting High value of the cumulative continuity errors, especially for the case with the fully developed flow. As a consequence the fields data are really strange see my post here.

In case of the fully developed flow i used the "setdiscreteFields" to map the inlet BCs. I'll upload my case,please take a look at it and see if you can find what i'm doing wrong, i tried everything and changed almost everything with no success.

I changed also the mesh, varied the yplus value, the grid spacing (uniform and non uniform) used different programs to create the mesh but all in vain.

here the zip with the case:
https://www.dropbox.com/s/76zz543hb5p3808/BFS.zip

Hoping someone would help me this time.

[ArathoN]

I tried to find out the culprit of the high cumulative error and the weird behavior of the field around the step, I simplified the mesh and i've considered a uniform mesh with the fully developed flow imposed at the inlet and the residual have a really strange behavior and i don't know what is the cause of it i started with the mesh at yplus 30 then i started to refine (x4) but i observed the same pattern.

I uploaded an image with the plot of the residual done with pyFoam.


P.S how can i save the plots done by pyFoam?

[ArathoN]

I increased the cell number up to 400000 from around 12000, but th residuals have the save pattern. Now i'll try to change the initial condition and see if the problem lies there. Can the schemes or fvsolution settings be the reason of this?

[wc34071209]

I am faced with almost the same problem and I don't know why.

[ArathoN]

Originally Posted by wc34071209

I am faced with almost the same problem and I don't know why.

I've gonna do more tests, but i narrowed it down to teh k/omega/epsilon initial values or the schemes used in fvchemes.

I decided to simplify the mesh using an uniform with the same spacing in the x and y direction, then i increased the cell points and simulated it (up to about 500000 cells) and all of them had the same behavior of the residuals. The residuals tend to go down at each iteration then they explode when they are around e-5, they manifest an oscillatory behavior with a large amplitude and i'm still trying to find out the root of the problem, for sure the initial value of the velocity (being a fully developed flow or not) ans the mesh aren't the cause of it.

[ArathoN]

I refined the mesh, changed the geometry and considered different types of BCs but i couldn't trace the root of the problem. With paraFoam i can't see the vortex (with the streamline tool) and the residual after some time will go crazy. Please can someone help me here.