[katiyon]

Does anybody know how to model dam-break? i've attach the picture
there is no inlet or outlet and the volume of water is constant.

[katiyon]

it has a simple geometry but i don't know how to set boundary conditions ????????

[ghorrocks]

This looks straight-forward. Have you done the tutorial example for free surface flow over a bump? That should give you the main idea of how to do it.

[katiyon]

yea, i've done tutorials but this is special model, there is no inlet and outlet, i made my model in free surface with 2 fluids by volume fraction equation
how do you think?

[ghorrocks]

No inlet and outlet makes it much easier. You will still probably need an air opening on the top to let air go in and out.

[katiyon]

honey i've already made a simple 2-D model, without inlet and outlet; just an opening above! i don't know why it doesn't run!

[ghorrocks]

What is the error message? Post the output file.

Honey? I have been called many things in CFD but never honey

[katiyon]

ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Floating point exception: Overflow

" you dont know how your help is valued to me!"

[katiyon]

the out put file is not valid to upload

[ghorrocks]

Floating point exception means that the solver has diverged badly. You need to improve the numerical stability of the simulation. This can be caused by lots of things, but in rough order I would check:
1) That the physics is correct
2) That you time step is small enough
3) That the mesh quality is high enough
4) If using high order differencing switch to low order (but just for debugging, once things are working go back to high order differencing)

[AliTr]

Originally Posted by katiyon

ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Floating point exception: Overflow

This happens in free surface models with sharp vertical angles, with model has difficulties to start up, it needs some patience to resolve it. adding to items that Glen mentioned:
1- try to refine your mesh
2- In your intial conditons, instead of using automatic options try to provide information (e.g. velocity, hydrostatic pressure,...)
3- start with very small time step

in cases other than dam break (e.g. flow over a spillway), I found, it is much easier to fill the domain with water and let model drain out extra water while converging toward steady condition.

[katiyon]

to check anything!!!
a 2-D model; no inlet or outlet; just an opening with air;
initials: velocity components are zero; hydrostatic pressure is applied; homogeneous turbulent model, time step=.001s & total time=2s, turbulent eddy dissipation checked

and this is the result:

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Floating point exception: Divide by zero |
|

i think the mesh size is small enough!

now what? (((((

[ghorrocks]

Divide by zero errors can come from many sources - you will have to find it. Try removing the hydrostatic pressure initial condition, removing gravity, running a laminar model as a few ideas to try to find the source of the error.

[seyedashraf]

I would like to mention that the first two screen shots are showing the results from a smooth particle hydrodynamic ( SPH ) method not a VOF/FVM method