Is there a specific reason laminar viscosity is connected to SPS in DualSPHysics?

edited January 2023 in DualSPHysics v5.0

I was wondering this because I don't really understand the point of using SPS when applying the physical viscosity.

If I understand it correctly, the molecular viscosity is not strong enough to dampen the oscillations in the density and pressure fields, such that you should use the delta-SPH formulation to directly diffuse the density. However, where is the point of using an additional SPS viscosity? Is this another stability measure (as it of course introduces an extra viscosity) ?

Maybe I am also misunderstanding the laminar+SPS approach and it was introduced to allow turbulence modelling rather then giving a more intuitive viscosity treatment than the artificial viscosity. In that case I understand that it makes sense to add laminar viscosity (with Morris operator) to the model.

However, I would then still be interested in why they are coupled, and why there is not an option to activate the physical viscosity treatment (with DDT) without the SPS model.

If someone has an opinion about that I am curious to hear it.


  • edited May 2023

    I am also confused by the fact that DualSPHyscis does not provide a solution with only laminar term. This is important when testing the efficiency of SPS turbulence model.

    One solution I could think of for this issue is to choose the artificial viscous term solution in Case 1 and set alpha to 0 to simulate the situation without the SPS turbulence model; then choose the laminar + SPS solution in Case 2 and set alpha to 0 again to simulate the situation with the SPS model; and finally the performance of the SPS model solution is discussed by comparing two Cases.

    Is this approach feasible?

Sign In or Register to comment.