Artificial viscosity in DSPHysics

Mustaf

Hi,

I have a question about the artificial viscosity value that we have to enter in xml file. It correspond to what ? the alpha linear coefficient (from: https://wiki.manchester.ac.uk/sphysics/images/SPHysics_v2.2.000_GUIDE.pdf) ?
I also read the reference "Simulating Free Surface Flows with SPH" of J.J.Monaghan 1992. And I can see that the second term (with beta) desappear in Sphysics. A clarification is possible ?
Thank you in advance!

Mustaf

Mustaf

I found what I was looking for with Monaghan 2005 and Monaghan and Gingold 1983, so the clarification is not needed anymore. But I still have a problem with the viscosity value that we have to implement in the xml folder. It correspond to what exactly ? Alpha in Sphysics documentation must be between 0.01 and 0.1 and we can find in the available cases (dualSphysics) a variation between 0.1 and 0.4.

mn61

As you already recognized it´s an artifical value. There is no range defined range since it is just an math parameter to stabilize the calculation. You can find useful values in the DualSPHysics validation cases.

I recommend to examine the effect of this parameter closely because you change the fluid behaviour in a non-physical way.

Alex

cases dominant by gravity are clearly dependent on ALPHA
so changes in dp imply changes in alpha
you should use the minimum alpha that stabilises your simulation

however for wave propagation a value of alpha=0.01 is correct

Mustaf

Ok It's really better now, thank you both!

TianwenDong

Hi, If I want to simulate non_Newtonian flows, Can you give me some suggestion about how to modify the code?
Thanks a lot!
Tianwen Dong