It is time again, in relation to the release of Simcenter STAR-CCM+ 2402, to look at the news related to multiphase simulations in a new version. So, what can we expect in the first release of this year? In general, the updates to existing frameworks are relatively few. Drag correction for Fluidized beds in EMP granular simulations, temperature dependent contact physics for DEM. But the largest inclusion for multiphase this time around is the SPH solver implemented in Simcenter STAR-CCM+. But let’s take a look at them one at the time.
Energy Minimization Multi-scale (EMMS) Drag Model for EMP Granular
This very long and informative name boils down to the fact that we can now simulate fluidized beds with Granular flow more accurately. The application from this is particularly fluidized beds. The Energy minimization Multi-scale (EMMS) drag model allows for a coarser grid and larger timesteps. Traditional drag models assume a homogeneous distribution of particles within the cell where there is no effect on a particle’s drag of other nearby particles. In reality one particle may be in the wake of another, and this changes its drag significantly. This contributes to clustering of particles where particles get entrained in each other’s wake. Neglecting this effect causes particles to be much more widely distributed than happens in reality. In fluidized bed applications this shortcoming leads to inaccurate results. The EMMS model incorporates drag corrections that account for the sub-grid clustering of particles and associated non-uniformities that develop giving much more accurate results.
The difference between the EMMS model and the pre-existing Gidaspow model can be seen in the example below. The Gidaspow model has a homogenous drag model, assuming that particles are only surrounded by gas and there is no effect of other particles on the particle when it comes to drag. The outcome of this is that particles experience too much drag and fluidization is heavily overestimated. The EMMS model limits the fluidization better as demonstrated in the plot below, showing a good comparison against experimental data for EMMS.
Energy Minimization Multi-scale (EMMS) Drag Model for DEM
The EMMS model is also implemented in the DEM framework. This heavily reduces the need to perform manual tuning of the drag model. Also simulating the fluidized bed with DEM, comparing EMMS and Gidaspow we can see that the Gidaspow model violently mixes the gases and solid, while the EMMS model shows a fluidization regime with steady bubble generation, which in this case is the target.
Temperature dependent contact physics (DEM)
With the new DEM feature of temperature-dependent contact physics we now get a couple of benefits for applications like rotary calciners, blast furnaces, kilns, thermal runaway etc. We can track the value of temperature at the point of contact. The capability also returns the temperature of one of two objects in a collision. This is available when “Register Contact Field functions” option is checked on for the DEM. This means for instance that we can have contact-temperature dependent cohesion when a particle hits a wall. For example, if a metallic particle is above melting temperature we can allow for the particle to adhere to the wall as it would be more sticky. The video below shows an example with thermal runaway where solid components are investigated and the deposition of particles onto the surrounding surface is investigated.
Introducing the SPH solver for Simcenter STAR-CCM+
This marks a milestone, not only have Simcenter STAR-CCM+ received a Smoothed-Particle Hydrodynamics (SPH) solver. But, also as of this day, this multiphase news series to the Volupe blog will have to include the SPH-framework as it does not fit in either the traditional Simcenter STAR-CCM+ Eulerian and Lagrangian frameworks.
In this version the SPH solver is fully integrated into the platform. In short, SPH is used to model complex transient flows with highly dynamic free surface flows. One large advantage of SPH is that it does not require a volume mesh. The particles themselves can be considered the mesh.
In this first release, the powertrain lubrication oil bath application it targeted. So far, the domain needs to be closed, it is not compatible with inlets and outlets. Currently we have an implicit incompressible SPH solver, compatible with gravity, fluid viscosity, fluid initialization and rotation for motion (no translation or other complex motions). More complexity will be added in future releases as the integration of the SPH-solver is complete.
Below is a demonstration of the main steps of setting up and post-processing when using the SPH-solver. It should be noted that the SPH-solver requires an additional license.
I hope the news has been interesting to read about and that you will have use for it in the future. As usual, reach out to support@volupe.com if you have any questions.
Author
Robin Victor
+46731473121
support@volupe.com