[Thomas MADELEINE]

Hi everybody,


I have some issues with one of my simulation with fluent.

I run a transient simulation of a tank, half-filled with liquid (other half is gas)


I use VOF model with geo-reconstruct method
K-epsilon turbulence model
pressure velocity coupled (try also with SIMPLE, PISO and SIMPLEC but didn't help)
second order for turbulence energy
no inlet nor outlet
some gravity to maintain the liquide down.

I run this case for some timestep (10 ts, 1ms each) with no problem.


After that I want to see what happens in quite a instant time so I run one more timestep but with a value of 1e-7.

it converges quite well (max is 1e-8) and everythings look good except the static pressure which went quite strange... (pressure gradient doesn't only depend on the gravity but create a bulb, some points with high or low values appear too).

Moreover the shorter the timestep is, the smaller the difference should be between two timestep... I would think but this is clearly not the case here...

I don't understand why since I have tried with bigger timestep and this effect is much more damped and become null with timestep close to the initialisation value.


In fact it seems that I can repeat this error and it clearly depends on my timestep value (shorter timestep means bigger deformation).
also this error is quite converged : it is the same with another try and/or with more/less iteration (so not random value).

If somebody has any idea, he or she will make my day.

thanks in advance,

Thomas