Single phase solver only suitable for water simulation?

Hey guys!

I am using DualSPHysics to do a simulation with a floating object which is first moving in water and then in air. When I do it with water (rhop = 1000, gamma = 7, relativeweight=1.3) then every thing is working fine. But if I change the fluid to air I start loosing a lot of particles very fast (rhop = 1.18, gamma = 1.4, relativeweight=1101.7).

So how should I go about changing the parameters if I want to simulate air instead of water in single phase? I've tried massbody for floating, but it hasn't really worked for me, since that when I use water the massbound is: 1e-6, but when I use air it suddenly becomes 1e-9 even though I use the correct massbound. In the water sim I see that I have:

36811 particles * 0.000001 kg = 0,036811kg

And I insert this as:

Am I doing something wrong?

I found a temporary solution in using the MultiPhysics branch and just using one fluid (air) only, but wanted to know if it is possible in single phase solver to do instead.

Kind regards


  • Hi Asalih3d,

    The single phase code is based on the weakly compressible formulation which uses the Tait equation of state (EoS) and solves only the continuity and momentum equations. So, changing the parameters of gamma = 7 and rho0 = 1.1 won't really work. The speed of sound is also different and this has consequences for the CFL number. Then there are the boundary conditions which will operate differently for such vastly different densities.

    The multi-phase code has tested a simple isothermal EoS and should work.

    Finally, strictly speaking, to simulate a simple gas, the thermal energy equation (e) should also be solved with an equation of state such as p.v = rho(gamma-1)e. But this depends on whether you need to include such physics which is not in DualSPHysics.

  • Hi @ben

    Thanks, for your clarification, I guess it is not possible to use air as a fluid medium in the single phase solver then, for the above reasons. It is definitely possible to do in the Multi-Phase brance (Liquid-gas) though :-)

    I don't understand the last paragraph about having to change the equation of state, is it because that the Tait equation is only correct for liquids? So we should use something based on the ideal gas law instead? I guess that is what the Multi-Phase branch does?

    Kind regards

  • I think I answered my latest question - Tait equation is only used for liquids according to WIkipedia. Thanks again for your clarification, it really did help me
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!