True to tradition, whenever there is a new release of Simcenter STAR-CCM+, there is a new set of multiphase news that we look at in our blog. And this time is no different. We’ll be looking at the Eulerian and Lagrangian multiphase news, starting with Eulerian. Simcenter STAR-CCM+ version 2502 was released a few weeks ago, and last week my colleague Christoffer wrote about news related to CAD and mesh features, presenting the new motion feature of virtual body. [New CAD And Mesh Features In Simcenter STAR-CCM+ 2502 – Volupe.com]
Improved slip modeling robustness for MMP
When doing MMP (Mixture Multi Phase) it is important to get the slip between phases correct, as we are simulating the entire Eulerian field as a mixture, the only thing that separates the phases spatially is the velocity difference between them. When this difference is large, robustness can become a problem, and the improvement in 2502 is there to address these problems. This should lead to better convergence and more accurate results. This should also reduce the need to artificially limit the slip velocity to maintain stability.
Together with numerical improvements a coupled of default has been changed with the intent of making the MMP simulations easier to run with default values. First, the Body Force Smoothing Iterations, used to smoothen the distribution of specific body forces, has been changed from 0 (deactivated) to 2. For the interaction length scale, that decides the interaction area density between phases, both the first and second dispersed regime interactions are set to 1e-4 m.
The graphs below show the difference in mass monitoring between the versions. The case includes conversion between MMP to Lagrangian with a slip limiter of 20 m/s. In 2410, this simulation did not run well, and experienced mass loss, while in 2502, the simulation is more stable and can maintain a releatively constant mass of the liquid phase.
S-gamma Sub-Grid Stripping model for EMP-LSI
We can now accurately predict droplet populations stripped from free surfaces with the S-gamma stripping model for EMP-LSI. It is basically the reverse of the entrainment model, where we now instead of feeding bubbles from jets into the liquid phase, can better feed droplets to the gaseous phase and provide that information to the population balance. The sub-models that come with this feature are models for stripping rate (Brocchini and Peregrine) and stripping diameter (Limited Hinze). The example below shows the stripping rate prediction from the free surface in a spillway.
Aside from these specific improvements with changes or additions to the a model, the general robustness of EMP granular has also Improved. This includes general robustness for simulations that contain the EMMS or Gidaspow drag models.
Wall-bound to fluid film transition model for LMP
The next addition is a Lagrangian one. Typically for water management simulations, a general speed up is now possible, as the wall bound phase can transition into both VOF and fluid film. The transition between the wall-bound phase and the Fluid film is based on a user-provided value of Equivalent film thickness. The animation below shows an example of simulating rainwater flow on the surface of the side mirror of a moving car. The droplets moving in the volume region, also known as free-stream Lagrangian droplets, are colored in blue, while wall-bound droplets moving in the shell region of the side mirror geometry are colored in grey. The new transition model here provides the formation of the fluid film where the water quantity is higher, near the edges of the mirror.
Flat fan nozzle injector for LMP
The flat fan nozzle injector type is a selection for Atomizer type when LISA atomization is selected as a primary atomization. It uses similar models to the Linear Instability Sheet Atomization (LISA) breakup model. This option is useful in several applications such as cleaning and degreasing, coating and painting, cooling and humidifying and dust control. The animation below shows an example of the flat fan atomizer in effect in the cleaning and cooling of a solar panel.
Native table option for the latent heat of vaporization
Version 2502 introduces the native table option for specifying latent heat of vaporization. This is a useful feature for all internal combustion engine simulations, since heat of vaporization has a temperature dependance that affects the evaporation rate of fuel droplets. Without accounting for this temperature dependence, the prediction of droplet sizes is often inaccurate. The previous workaround to this was rather complicated and included user-defined field functions. This new option provides a simpler and more intuitive workflow for this type of simulation.
I hope this blog has been useful to read, and I also hope it provides ideas for what is possible to do with your multiphase simulations. As usual, if you have any questions reach out to support@volupe.com.
Author
Robin Victor
+46731473121
support@volupe.com