[ham]

Hey.

I have a few simple(?) questions.

First let me describe my situation:
I have a (2m*2m) domain with a 0.2mØ circle in the middle.
I am using simpleFoam. (is that ok?)
Left boundary = inlet.
Right boundary = outlet.
bottom boundar = upper boundary = wall (moving with same speed as inlet value)
circle = wall

Questions:
1. What should i do with the "interior" regarding U,p,k,epsilon?

2. How do I know the smallest cellsize dx if i have converted my case from fluent? (i dont have any BlockmeshDict)

3. After having guessed arbitrary answers to the above questions I get
"BICCG: Solving for Ux: solution singularity"
etc for Uy,p,epsilon,k
"time step continuity errors: sum local = nan, global = nan, cumulative = nan"


Hope you dont mind my rookie-questions.

Thanx!!

/ham

[pierre]

You can look in the tutorial for simpleFoam. typical conditions/values for walls, inlet and outlet are in there. If you're using simpleFoam, I assume you're doing a steady-state calculation. So initial internal values can be set to any guessed values, it is not really important (I think), again look at the tutorial. For bottom and upper boundary, are they really walls or you just want to set it far free stream conditions?

try running CheckMesh

Hope that helps

Pierre

[ham]

Thank you very much.
I will look into this now.
Hope Ill get it to work...

Have a nice day.

/Marcus

[ham]

Okey, now it works better.
I compared all my files against the files in the tutorials until I found the error.

I have to say its a bit embarrasing, but the kinematic viscosity was set to 0. For my defence I thought the default value of this property was set to something more logical...

Now i just have to make sure that the freestream keep on running and does not end after the first iteration

Thanx again
Marcus