Learning Forum

[muhammadyaseenkhan171]

Dear ANSYS Support Team,

I am performing a transient VOF simulation of two liquid jets intersecting at 90° and forming a liquid sheet. The final timestep of the simulation typically shows a physically correct liquid sheet, and the sheet formation and breakup behavior appear reasonable.

However, during intermediate timesteps, the α = 0.5 isosurface often appears highly fragmented and randomly dispersed throughout the domain.

I am trying to understand whether this behavior is related to Adaptive Mesh Refinement (AMR) projection or another numerical issue.

I am observing the following unexpected behavior:

• Let x1 be the final timestep of a run.

• At timestep x1, the liquid sheet appears correct and physically consistent.

Now, if I return to the solver and continue running the simulation for a few more timesteps:

• Let x2 become the new final timestep.

• At timestep x2, the liquid sheet again appears correct and physically consistent.

However:

• When I go and visualize timestep x1 (which previously showed a correct sheet before continuing the run),

• The α = 0.5 isosurface at x1 is now highly distorted, fragmented, and randomly dispersed.

In other words, a previously correct timestep becomes visually corrupted after advancing the simulation further.

[Rob]

It could also be a result of the initial contact between the two jets. In that they collide, shatter, re-collide, form an unstable sheet that eventually stabilises. AMR can cause some instabilities if mesh updates are overly frequent or the intial mesh is coarse but is generally suitable for this sort of application. 

Maybe share pictures as x1 and x2 aren't as clear as good images.