Stationary 2D square dispersing

Hi! I am trying to build a horizontal 2D simulation, but it seems that DesignSPHysics 2D setup is always vertical and I suppose it applies to DualSPHysics as well. I figured I can't just set gravity to 0, but got ahead by the CaseForces example, adding an equal and opposite force.

However, something strange happens. It seems that most particles on the top edge disappear after a few timesteps, and an area of less dense fluid starts developing in the center.

I'd be interested in any insights on why this might happen, and how I could mitigate the issue! Eventually I want a model where matter could be at least mostly stationary in the absence of forces. Here's a video:


  • I tried switching the integration kernel to from Wendland to Cubic spline, and that stopped the density change. The fluid is still rather restless, in a different way, and I'd like to know why, but it still feels like an improvement!

  • Using Wendland and Laminar + SPS viscosity it goes completely crazy :D

  • Cubic spline kernel with Laminar + SPS viscosity makes for a violent explosion of particles, almost all of which leave the domain.

Sign In or Register to comment.