Sticking particles
Hi
I'm running a two liquid phases simulation which performs fairly well. But, one of the liquids tends to stick to the boundaries... randomly. When a particle touches a wall, it may either follow its way, or get stuck in contact of the surface. Only one of the liquids does this. What could be the reason ?
I'm using DualSPHysics3.4_LiquidSediment_win64.exe, with a two newtonian liquid case in 3D. The liquids have different volumic mass (7000 and 2500 respectively), and different viscosities (0.007 and 2.9 respectively).
Regards
I'm running a two liquid phases simulation which performs fairly well. But, one of the liquids tends to stick to the boundaries... randomly. When a particle touches a wall, it may either follow its way, or get stuck in contact of the surface. Only one of the liquids does this. What could be the reason ?
I'm using DualSPHysics3.4_LiquidSediment_win64.exe, with a two newtonian liquid case in 3D. The liquids have different volumic mass (7000 and 2500 respectively), and different viscosities (0.007 and 2.9 respectively).
Regards
Comments
Just to follow on the same subject, I have noticed that this behaviour seems to be correlated to particles with strange densities : following the initial step, some particles seem to have acquired a density far from normal... They tend to keep this density along with time. You can see that on the figure below : normal density is dark blue, other colors range up to twice the normal density.
A particle with a strange density repulses the other particles (this keeps the average density constant in space), but one gets disturbed surfaces on postprocessing.
Please note that at the beginning, the two liquids are not in contact.
Regards
Interesting application,
If you are running two fluids, I assume you have all the parameters related to the yield strength equal to zero?
could you please forward me the xml with one or two outputs (*.vtk)?
Thanks George
In fact I'm trying to simulate a two non-miscible liquid setting. With the Liquid-Sediment version of DualSPHysics, it works. I have described two newtonian fluids and the global behaviour is just as expected. The yield strength is 0 for both.
The problem seems to come from the the walls which have a fixed density. When filling au tank with a density equal for the walls and the liquids, everything is OK. When filling a tank with a low density liquid compared to the walls, the liquid violently shrinks at the very beginning, then stabilizes. It behaves normally afterwards. When you put a high density liquid in a low density tank... it explodes : most of the liquid is ejected through the walls.
In my case I'm working with a density of 7000 for the walls and one of the liquids. The other liquid has a density of 2500. The sticking problem comes from this light liquid, not from the other one. Few particles are concerned, and I can handle this little problem. But I would prefer to have a nice clean wall surface.
NB : How do I proceed to send you an *.xml ?
Regards.