Droplet transport in an industrial plant


  • Inclusion Transport
  • Particle & Droplet Transport in an Industrial Setting
  • Particle & Droplet Transport in the Environment

Capabilities covered in this demonstration

  • Multi-fluid
  • Fixed Size Droplet Transport (FSDT)
  • Algebraic Slip
  • Dispersed Particle/Droplet Phase

Description & Performance Data

A 3D SST Detached Eddy Simulation of incompressible, isothermal air with a droplet phase injected at the inlet. This example illustrates fixed size droplet transport using the FSDT model in EXN/Aero. The FSDT model employs an alegbraic slip model (with Stokes drag and gravity forces) to estimate the mean droplet velocity relative to the continuous phase (in this example a gas). The model also, if selected, activates a turbulent diffusion model (Ryan Model [1]) that includes droplet/particle inertia and time limit effects. The FSDT model is limited to dilute particle/droplet flows in a mildly accelerating continuous phase (due to the Stokes drag approximation) with a long-time limit set as the default for the turbulent diffusion model (i.e. it no longer depends on time). Under these conditions the model has been applied to air/(particle/droplet) flows with droplet sizes up to ~150 microns [1]. As an alternative to the Ryan model, the user can also estimate the effect of turbulent dispersion by entering a Schmidt number which is applied as a scaling parameter to the turbulent viscosity. The laminar droplet diffusion coefficients can be assigned directly or it can be based on the Brownian motion theory [1].

Mesh Topology:

Multi-block structured

Data Type:


No. Control Volumes:

~1 Million



GPU Type:

Nvidia K80

No. GPU Devices:


GPU memory usage:


CPU Type:

Intel Xeon

No. CPU Devices:



Droplets of three different sizes, and volume fractions, enter the domain via a boundary source area centred at the inlet of a pipe. The injected volume flow rate of the dispersed phase is governed by the inflow volume fractions and the shared continuous phase velocity (the droplet slip at the inlet is assumed to be zero). A volume fraction equation is solved for each of the three size classes. The distribution and trajectory of the droplet phase (represented by the volume fractions) can deviate from the continous phase and will be most significant for the larger droplet sizes. The forces acting on the droplet to affect this motion is drag and gravity. Diameter, volume fraction and physical property values for each of the three size classes are shown in the table below. When boundary conditions are specified a separate inlet boundary family must be specified for the portion of the inlet that supplies the droplet sources (called a multifluid inlet). In this example the gas phase inlet conditions are constant across each inlet at a velocity of -10.0 m/s for the inner multifluid inlet and -15.0 m/s for the outer inlet area. Initial velocity in the domain is likewise set to 0.0 m/s. A gravitational acceleration of -9.81 m/s2 in the z-direction is applied, which draws droplets downward on average over time. For the results presented a turbulent dispersion model based on a Schmidt number of 0.1 is used.

EXN/View: Mesh, Physics and Solver Settings

This is a navigable instance of the actual EXN/Aero user interface, showing the mesh at right and the solver control tree at left. Expand nodes in the tree to see an how we set up the physics models and solver parameters. The demonstration case shown here can be reconstructed on any terminal where EXN/Aero is installed, using the files found below.

Data Files



Tutorial sessions can be found at:


Diffusion coefficient, mean velocity magnitude, volume fraction and relative droplet velocity (z-direction) are shown for each droplet size class.


  • Fixed Size Droplet Transport (FSDT)
  • Dispersed Phase Particle Transport
  • Multi-fluid

[1] Ryan SD, Gerber AG, Holloway AGL, A time-dependent Eulerian model of droplet diffusion in turbulent flow. Computers and Fluids 2016;131:1-15. [2] Maxey MR . The gravitational settling of aerosol particles in homogeneous turbulence and random flow fields. J Fluid Mech 1987;174:441-65. [3] Shotorban B , Balachandar S . Particle concentration in homogeneous shear turbulence simulated via Lagrangian and Equilibrium Eulerian approaches. Phys Fluids 2006;18:65-105.

Last modified 14 months ago Last modified on Jun 1, 2017 11:12:46 AM

Attachments (11)

Download all attachments as: .zip