Forces on floating bodies

1

Hello,

I’m trying to run a simple test of a floating body starting at neutral buoyancy. The floating cube is half of the density of the water. At equilibrium, the vertical force acting on the cube should be equal to the weight of the cube. However, when I plot the vertical force acting on the cube over time, it settles to about 11% higher than the weight of the cube. In other words, Fz = 1.1W, where Fz is the applied force calculated by GPUSPH/Chrono and W is the weight of the floating cube. In Paraview, I can see that the cube is not moving and is at equilibrium during the test.

I’ve tried several different simulations of this general format and have gotten similar results, the applied force is always about 10-12% higher than the weight at equilibrium. Has this type of thing happened in the past with other simulations? Is there a way that I can tell if this is a problem on the GPUSPH or Chrono side?

Thank you in advance for your help.

2

Hi @brittlyd,

do you see any change in the simulated force if you use a different value of deltap?
What boundary model are you using? If you want you could send us the problem file so that we can have a better understanding of the issue.

Vito

3

Hi @vito.zago,

Thanks for your response! I actually just ran the test case BuoyancyTest and noticed that I am getting a similar trend of unbalanced forces. In this case, the simulated vertical force acting on the floating cube is about 20% higher than the weight of the object, even when it is in equilibrium and I’m wondering if other people get similar results. It could be a convergence issue, but I wanted to check that there wasn’t anything wrong on the Chrono/Eigen side, as I had some trouble with the installation and wanted to double check that it was working properly.