Transient model pattern changing point

Lugia_06

Hi everyone,

I found something interesting in my simulation pattern:

As I use an element size >=6e-5 and < 6e-5, the simulation patterns were completely different:

However, within their own group, simulation patterns were the same:

element size 5.5e-6 vs 5.75e-6

I am using the same original fluent file and only changing the mesh quality to export, all the settings should be the same. So I wonder if that's a problem associated with the mesh quality such as mesh orthogonality? would a change in bias factor in y-direction cause this issue?

cells with orthogonality>0.75 (element size 6e-6)

cells with orthogonality>0.75 (element size 5.75e-6)

the images of cells with skewness <0.8 and aspect ratio<=2 are the same with different element size.

Just wondering if anyone knows what could cause the difference in the simulation patterns?

Best regards,

Lugia_06

Kremella

Hello,

When you change your grid, do you also change your time-step to maintain the same CFL number?

Also, what are your VOF settings? Explicit formulation with sharp interface modeling?

What discretization schemes are you using?

Karthik

Lugia_06

Hi Kremella,

Good to see you on the forum again!😀 I kept the same number of time step and step size(1e-6) as I run the models, however, sometimes the fluent system itself auto changed my step size into a lower number like 9.999998e-7 (not sure if that matters).

Please see below my VOF settings and discretisation schemes:

explicit with sharp interface modelling

Best regards,

Lugia_06

YasserSelima

no it doesn't matter ... I always have this issue 😁