Rayleigh Taylor Instability with DualSPHysics5.0_NNewtonian
Hello DualSPHysics community! This is my first question on this forum.
I am trying to reproduce a Rayleigh Tayler instability and I was relatively successful so far, but some uncertainties remain and maybe one of you can help me out. Above in the picture you can see what I have so far.
I have three questions:
1) In the multi-phase non-Newtonian examples of DualSPHysics the “VelocityGradientType” 1:FDA is always selected. What does FDA stand for? Can someone provide a source of documentation about this parameter?
2) Computing the pressure with the PartVTK tool of a multi-phase simulation gives an unexpected result. I expect a continuous pressure gradient increasing from top to bottom, resulting from the hydrostatic pressure. However, the result is different. there are pressure gradients from top to bottom but only within a single phase. What I observe is that both phases have a pressure gradient independently, but at different pressure levels (only visible when rescaled properly in paraview). There is not one continuous pressure gradient from top to bottom for both phases. Is this a result of the PartVTK tool simply converting the particle density into pressure via the state equation? Is there a different way to do it for a multi-phase simulation?
3) How do I lower the viscosity and still obtain a physical result? In my example I use kinematic viscosities for oil and water of about 10^-4. What I want is to reduce them to 10^-6. But doing this, the fluid expands after about 0.8 s and the particles will bounce around like crazy. I noticed unphysical density fluctuation (like oscillating numerical noise) just before this phenomena occurs. Therefore, I tested around with the density diffusion formulation of Molteni and Fourtakas, but nothing influenced my simulation, like absolutely zero difference. Is there a different way for me to implement a density diffusion formulation? And coming back to the original question, is there another way how to lower the viscosities?
thank you for developeing such a nice open source software!!