DPM Fluent Manual

22.10 Discrete Phase Model (DPM) Boundary Conditions

You can define your own law other than Raoults Law for vapor concentration at theparticle surface using a user-defined function.

See Section 2.5.15: DEFINE DPM VP EQUILIB of the FLUENT UDF Manual for details.

22.10 Discrete Phase Model (DPM) Boundary Conditions

When a particle strikes a boundary face, one of several contingencies may arise:

The particle may be reflected via an elastic or inelastic collision. The particle may escape through the boundary. The particle is lost from the cal-

culation at the point where it impacts the boundary.

The particle may be trapped at the wall. Nonvolatile material is lost from thecalculation at the point of impact with the boundary; volatile material present inthe particle or droplet is released to the vapor phase at this point.

The particle may pass through an internal boundary zone, such as radiator orporous jump.

The particle may slide along the wall, depending on particle properties and impactangle.

You also have the option of implementing a user-defined function to model the particlebehavior when hitting the boundary. See the separate UDF Manual for informationabout user-defined functions.

These boundary condition options are described in detail in Section 22.13: Setting Bound-ary Conditions for the Discrete Phase.

22.11 Steps for Using the Discrete Phase Models

You can include a discrete phase in your FLUENT model by defining the initial position,velocity, size, and temperature of individual particles. These initial conditions, alongwith your inputs defining the physical properties of the discrete phase, are used to initiatetrajectory and heat/mass transfer calculations. The trajectory and heat/mass transfercalculations are based on the force balance on the particle and on the convective/radiativeheat and mass transfer from the particle, using the local continuous phase conditions asthe particle moves through the flow. The predicted trajectories and the associated heatand mass transfer can be viewed graphically and/or alphanumerically.

c Fluent Inc. September 29, 2006 22-91

Modeling Discrete Phase

The procedure for setting up and solving a problem involving a discrete phase is outlinedbelow, and described in detail in Sections 22.11.122.16. Only the steps related specifi-cally to discrete phase modeling are shown here. For information about inputs related toother models that you are using in conjunction with the discrete phase models, see theappropriate sections for those models.

1. Enable any of the discrete phase modeling options, if relevant, as described in thissection.

2. Choose a transient or steady treatment of particles as described inSection 22.11.2: Steady/Transient Treatment of Particles.

3. Enable the required physical submodels for the discrete phase model, as describedin Section 22.11.5: Physical Models for the Discrete Phase Model.

4. Set the numerics parameters and solve the problem, as described in Section 22.11.7: Nu-merics of the Discrete Phase Model and Section 22.15: Solution Strategies for theDiscrete Phase.

5. Specify the initial conditions and particle size distributions, as described in Sec-tion 22.12: Setting Initial Conditions for the Discrete Phase.

6. Define the boundary conditions, as described in Section 22.13: Setting BoundaryConditions for the Discrete Phase.

7. Define the material properties, as described in Section 22.14: Setting Material Prop-erties for the Discrete Phase.

8. Initialize the flow field.

9. For transient cases, advance the solution in time by taking the desired number oftime steps. Particle positions will be updated as the solution advances in time. Ifyou are solving an uncoupled flow, the particle position will be updated at the endof each time step. For a coupled calculation, the positions are iterated on or withineach time step.

10. Solve the coupled or uncoupled flow (Section 22.15: Solution Strategies for theDiscrete Phase).

11. Examine the results, as described in Section 22.16: Postprocessing for the DiscretePhase.

22-92 c Fluent Inc. September 29, 2006


22.11.1 Options for Interaction with the Continuous Phase

If the discrete phase interacts (i.e., exchanges mass, momentum, and/or energy) with thecontinuous phase, you should enable the Interaction with the Continuous Phase option.An input for the Number of Continuous Phase Iterations per DPM Iteration will appear,which allows you to control the frequency at which the particles are tracked and theDPM sources are updated.

For steady-state simulations, increasing the Number of Continuous Phase Iterations perDPM Iteration will increase stability but require more iterations to converge. Figure 22.11.1shows how the source term, S, when applied to the flow equations, changes with the num-ber of updates for varying under-relaxation factors. In Figure 22.11.1, S is the finalsource term for which a value is reached after a certain number of updates and S0 is theinitial source term at the start of the computation. The value of S0 is typically zero atthe beginning of the calculation.

In addition, another option exists which allows you to control the numerical treatmentof the source terms and how they are applied to the continuous phase equations. UpdateDPM Sources Every Flow Iteration is recommended when doing unsteady simulations;at every DPM Iteration, the particle source terms are recalculated. The source termsapplied to the continuous phase equations transition to the new values every flow iterationbased on Equations 22.9-6 to 22.9-8. This process is controlled by the under-relaxationfactor, specified in the Solution Controls panel, see Section 22.15.2: Under-Relaxation ofthe Interphase Exchange Terms.

Figure 22.11.1 can be applied to this option as well. Keep in mind that the DPM sourceterms are updated every continuous flow iteration.

22.11.2 Steady/Transient Treatment of Particles

The Discrete Phase Model utilizes a Lagrangian approach to derive the equations forthe underlying physics which are solved transiently. Transient numerical procedures inthe Discrete Phase Model can be applied to resolve steady flow simulations as well astransient flows.

In the Discrete Phase Model panel you have the option of choosing whether you wantto treat the particles in an unsteady or a steady fashion. This option can be chosenindependent of the settings for the solver. Thus, you can perform steady state trajectorysimulations even when selecting a transient solver for numerical reasons. You can alsospecify unsteady particle tracking when solving the steady continuous phase equations.This can be used to improve numerical stability for very large particle source terms orsimply for postprocessing purposes. Whenever you enable a breakup or collision modelto simulate sprays, the Unsteady Particle Tracking will be switched on automatically.



Figure 22.11.1: Effect of Number of Source Term Updates on Source TermApplied to Flow Equations



When Unsteady Particle Tracking is enabled, several new options appear. If steady stateequations are solved for the continuous phase, you simply enter the Particle Time StepSize and the Number of Time Steps, thus tracking particles every time a DPM iterationis conducted. When you increase the Number of Time Steps, the droplets penetrate thedomain faster.

i Note that you must enter the start and stop times for each injection.When solving unsteady equations for the continuous phase, you must decide whether youwant to use Fluid Flow Time Step to inject the particles, or whether you prefer a ParticleTime Step Size independent of the Fluid Flow Time Step. With the latter option, youcan use the Discrete Phase Model in combination with changes in the time step for thecontinuous equations, as it is done when using adaptive flow time stepping.

If you do not use Fluid Flow Time Step, you will need to decide when to inject the particlesfor a new time step. You can either Inject Particles at Particle Time step or at the flow timestep. In any case, the particles will always be tracked in such a way that they coincidewith the flow time of the continuous flow solver.

You can use a user-defined function (DEFINE DPM TIMESTEP) to change the time step forDPM particle tracking. The time step can be prescribed for special applications wherea certain time step is needed. See Section 2.5.14: DEFINE DPM TIMESTEP of the FLUENTUDF Manual for details on changing the time step size for DPM particle tracking.

i When the density-based explicit solver is used with the explicit unsteadyformulation, the particles are advanced once per time step and are calcu-lated at the start of the time step (before the flow is updated).

Additional inputs are required for each injection in the Set Injection Properties panel.For Unsteady Particle Tracking, the injection Start Time and Stop Time must be specifiedunder Point Properties. Injections with start and stop times set to zero will be injectedonly at the start of the calculation (t = 0). Changing injection settings during a transientsimulation will not affect particles currently released in the domain. At any point duringa simulation, you can clear particles that are currently in the domain by clicking on theClear Particles button in the Discrete Phase Model panel.



For unsteady simulations (with the implicit solvers), several methods can be chosen tocontrol when the particles are advanced.

If the Number of Continuous Phase Iterations per DPM Iteration is less than thenumber of iterations required to converge the continuous phase between time steps,then sub-iterations are done. Here, particles are tracked to their new positionsduring a time step and DPM sources are updated; particles are then returned totheir original state at the beginning of the time step. At the end of the timestep, particles are advanced to their new positions based on the continuous-phasesolution.

If the Number of Continuous Phase Iterations per DPM Iteration is larger than thenumber of iterations specified to converge the continuous phase between time steps,the particles are advanced at the beginning of the time step to compute the particlesource terms.

When you specify a value of Zero as the Number of Continuous Phase Iterations perDPM Iteration, the particles are advanced at the end of the time step. For thisoption, it may be better if the particle source terms are not reset at the begin-ning of the time step. This can be done with the TUI command define/models/dpm/interaction/reset-sources-at-timestep?.

In all the above cases, you must provide a sufficient number of particle source termupdates to better control when the particles are advanced, see Figure 22.11.1.

i In steady-state discrete phase modeling, particles do not interact with eachother and are tracked one at a time in the domain.

i If the collision model is used, you will not be able to set the Number of Con-tinuous Phase Iterations per DPM Iteration. Refer to Section 22.7.1: DropletCollision Model for details about this limitation.

22.11.3 Parameter Tracking for the Discrete Phase Model

You will use two parameters to control the time integration of the particle trajectoryequations:

the length scale/step length factorThis factor is used to set the time step for integration within each control volume.

the maximum number of time stepsThis factor is used to abort trajectory calculations when the particle never exitsthe flow domain.



Each of these parameters is set in the Discrete Phase Model panel (Figure 22.11.2) underTracking Parameters in the Tracking tab.

Define Models Discrete Phase...

Figure 22.11.2: The Discrete Phase Model Panel and the Tracking Parameters



Max. Number Of Steps is the maximum number of time steps used to compute a singleparticle trajectory via integration of Equations 22.2-1 and 22.15-1. When the max-imum number of steps is exceeded, FLUENT abandons the trajectory calculationfor the current particle injection and reports the trajectory fate as incomplete.The limit on the number of integration time steps eliminates the possibility of aparticle being caught in a recirculating region of the continuous phase flow fieldand being tracked infinitely. Note that you may easily create problems in whichthe default value of 500 time steps is insufficient for completion of the trajectorycalculation. In this case, when trajectories are reported as incomplete within thedomain and the particles are not recirculating indefinitely, you can increase themaximum number of steps (up to a limit of 109).

Length Scale controls the integration time step size used to integrate the equationsof motion for the particle. The integration time step is computed by FLUENTbased on a specified length scale L, and the velocity of the particle (up) and of thecontinuous phase (uc):

t =L

up + uc(22.11-1)

where L is the Length Scale that you define. As defined by Equation 22.11-1, L isproportional to the integration time step and is equivalent to the distance that theparticle will travel before its motion equations are solved again and its trajectoryis updated. A smaller value for the Length Scale increases the accuracy of thetrajectory and heat/mass transfer calculations for the discrete phase.

(Note that particle positions are always computed when particles enter/leave a cell;even if you specify a very large length scale, the time step used for integration willbe such that the cell is traversed in one step.)

Length Scale will appear in the Discrete Phase Model panel when the Specify LengthScale option is on.

Step Length Factor also controls the time step size used to integrate the equations ofmotion for the particle. It differs from the Length Scale in that it allows FLUENT tocompute the time step in terms of the number of time steps required for a particleto traverse a computational cell. To set this parameter instead of the Length Scale,turn off the Specify Length Scale option.

The integration time step is computed by FLUENT based on a characteristic timethat is related to an estimate of the time required for the particle to traverse thecurrent continuous phase control volume. If this estimated transit time is definedas t, FLUENT chooses a time step t as

t =t

(22.11-2)



where is the Step Length Factor. As defined by Equation 22.11-2, is inverselyproportional to the integration time step and is roughly equivalent to the numberof time steps required to traverse the current continuous phase control volume. Alarger value for the Step Length Factor decreases the discrete phase integration timestep. The default value for the Step Length Factor is 5. Step Length Factor willappear in the Discrete Phase Model panel when the Specify Length Scale option isoff (the default setting).

One simple rule of thumb to follow when setting the parameters above is that if you wantthe particles to advance through a domain consisting of N grid cells into the main flowdirection, the Step Length Factor times N should be approximately equal to the Max.Number Of Steps.

22.11.4 Alternate Drag Laws

There are five drag laws for the particles that can be selected in the Drag Law drop-downlist under Drag Parameters.

The spherical, nonspherical, Stokes-Cunningham, and high-Mach-number laws described inSection 22.2.1: Particle Force Balance are always available, and the dynamic-drag lawdescribed in Section 22.6: Dynamic Drag Model Theory is available only when one ofthe droplet breakup models is used in conjunction with unsteady tracking. See Sec-tion 22.11.6: Modeling Spray Breakup for information about enabling the droplet breakupmodels.

If the spherical law, the high-Mach-number law, or the dynamic-drag law is selected, nofurther inputs are required. If the nonspherical law is selected, the particle Shape Factor( in Equation 22.2-9) must be specified. The shape factor value cannot exceed 1. Forthe Stokes-Cunningham law, the Cunningham Correction factor (Cc in Equation 22.2-11)must be specified.



22.11.5 Physical Models for the Discrete Phase Model

This section provides instructions for using the optional discrete phase models available inFLUENT. All of them can be turned on in the Discrete Phase Model panel (Figure 22.11.3).

Define Models Discrete Phase...

Figure 22.11.3: The Discrete Phase Model Panel and the Physical Models



Including Radiation Heat Transfer Effects on the Particles

If you want to include the effect of radiation heat transfer to the particles (Equa-tion 13.3-13), you must turn on the Particle Radiation Interaction option under the PhysicalModels tab, in the Discrete Phase Model panel (Figure 22.11.3). You will also need todefine additional properties for the particle materials (emissivity and scattering factor),as described in Section 22.14.2: Description of the Properties. This option is availableonly when the P-1 or discrete ordinates radiation model is used.

Including Thermophoretic Force Effects on the Particles

If you want to include the effect of the thermophoretic force on the particle trajectories(Equation 22.2-14), turn on the Thermophoretic Force option under the Physical Modelstab, in the Discrete Phase Model panel. You will also need to define the thermophoreticcoefficient for the particle material, as described in Section 22.14.2: Description of theProperties.

Including Brownian Motion Effects on the Particles

For sub-micron particles in laminar flow, you may want to include the effects of Brow-nian motion (described in Section 22.2.1: Brownian Force) on the particle trajectories.To do so, turn on the Brownian Motion option under the Physical Models tab. WhenBrownian motion effects are included, it is recommended that you also select the Stokes-Cunningham drag law in the Drag Law drop-down list under Drag Parameters, and specifythe Cunningham Correction (Cc in Equation 22.2-11).

Including Saffman Lift Force Effects on the Particles

For sub-micron particles, you can also model the lift due to shear (the Saffman lift force,described in Section 22.2.1: Saffmans Lift Force) in the particle trajectory. To do this,turn on the Saffman Lift Force option under the Physical Models tab, in the Discrete PhaseModel panel.

Monitoring Erosion/Accretion of Particles at Walls

Particle erosion and accretion rates can be monitored at wall boundaries. These ratecalculations can be enabled in the Discrete Phase Model panel when the discrete phaseis coupled with the continuous phase (i.e., when Interaction with Continuous Phase is se-lected). Turning on the Erosion/Accretion option will cause the erosion and accretion ratesto be calculated at wall boundary faces when particle tracks are updated. You will alsoneed to set the Impact Angle Function (f() in Equation 22.5-1), Diameter Function (C(dp)in Equation 22.5-1), and Velocity Exponent Function (b(v) in Equation 22.5-1) in the Wallboundary conditions panel for each wall zone (as described in Section 22.13.1: DiscretePhase Boundary Condition Types).



Including the Effect of Particles on Turbulent Quantities

Particles can damp or produce turbulent eddies ([269]. In FLUENT, the work done by theturbulent eddies on the particles is subtracted from the turbulent kinetic energy usingthe formulation described in [100] and [9].

If you want to consider these effects in the chosen turbulence model, you can turn thison using Two-Way Turbulence Coupling, under the Physical Models tab.

Tracking in a Reference Frame

Particle tracking is related to a coordinate system. With Track in Absolute Frame en-abled, you can choose to track the particles in the absolute reference frame. All particlecoordinates and velocities are then computed in this frame. The forces due to frictionwith the continuous phase are transformed to this frame automatically.

In rotating flows it might be appropriate for numerical reasons to track the particles inthe relative reference frame. If several reference frames exist in one simulation, then theparticle velocities are transformed to each reference frame when they enter the fluid zoneassociated with this reference frame.

22.11.6 Options for Spray Modeling

When you enable particle tracking, the Physical Models tab in Discrete Phase Model panelwill expand to show options related to spray modeling.

Modeling Spray Breakup

To enable the modeling of spray breakup, select the Droplet Breakup option under SprayModel and then select the desired model (TAB or Wave). A detailed description of thesemodels can be found in Section 22.7.2: Droplet Breakup Models.

For the TAB model, you will need to specify a value for y0 (the initial distortion attime equal to zero in Equation 22.7-14) in the y0 field. The default value (y0 = 0) isrecommended. You will also set the number of Breakup Parcels (under Breakup Constants),to split the droplet into several child parcels, as described in Section 22.7.2: Velocity ofChild Droplets.

For the wave model, you will need to specify values for B0 and B1, where B0 is theconstant B0 in Equation 22.7-43 and B1 is the constant B1 in Equation 22.7-45. You willgenerally not need to modify the value of B0, as the default value 0.61 is acceptable fornearly all cases. A value of 1.73 is recommended for B1.

For steady-state simulations, you will also need to specify an appropriate Particle TimeStep Size and the Number of Time Steps which will control the spray density. See Sec-tion 22.11.1: Options for Interaction with the Continuous Phase for more information.



Note that you may want to use the dynamic drag law when you use one of the spraybreakup models. See Section 22.11.4: Alternate Drag Laws for information about choos-ing the drag law.

Modeling Droplet Collisions

To include the effect of droplet collisions, as described in Section 22.7.1: Droplet CollisionModel, select the Droplet Collision option under Spray Models. There are no further inputsfor this model.

22.11.7 Numerics of the Discrete Phase Model

The underlying physics of the Discrete Phase Model is described by ordinary differentialequations (ODE) as opposed to the continuous flow which is expressed in the form ofpartial differential equations (PDE). Therefore, the Discrete Phase Model uses its ownnumerical mechanisms and discretization schemes, which are completely different fromother numerics used in FLUENT.

The Numerics tab gives you control over the numerical schemes for particle tracking aswell as solutions of heat and mass equations (Figure 22.11.4).

Numerics for Tracking of the Particles

To solve equations of motion for the particles, the following numerical schemes are avail-able:

implicit uses an implicit Euler integration of Equation 22.2-1 which is unconditionallystable for all particle relaxation times.

trapezoidal uses a semi-implicit trapezoidal integration.

analytic uses an analytical integration of Equation 22.2-1 where the forces are heldconstant during the integration.

runge-kutta facilitates a 5th order Runge Kutta scheme derived by Cash and Karp [50].



Figure 22.11.4: The Discrete Phase Model Panel and the Numerics



You can either choose a single tracking scheme, or switch between higher order and lowerorder tracking schemes using an automated selection based on the accuracy to be achievedand the stability range of each scheme. In addition, you can control how accurately theequations need to be solved.

Accuracy Control enables the solution of equations of motion within a specified toler-ance. This is done by computing the error of the integration step and reducing theintegration step if the error is too large. If the error is within the given tolerance,the integration step will also be increased in the next steps.

Tolerance is the maximum relative error which has to be achieved by the trackingprocedure. Based on the numerical scheme, different methods are used to estimatethe relative error. The implemented Runge-Kutta scheme uses an embedded errorcontrol mechanism. The error of the other schemes is computed by comparing theresult of the integration step with the outcome of a two step procedure with halfthe step size.

Max. Refinements is the maximum number of step size refinements in one single inte-gration step. If this number is exceeded the integration will be conducted with thelast refined integration step size.

Automated Tracking Scheme Selection provides a mechanism to switch in an automatedfashion between numerically stable lower order schemes and higher order schemes,which are stable only in a limited range. In situations where the particle is far fromhydrodynamic equilibrium, an accurate solution can be achieved very quickly witha higher order scheme, since these schemes need less step refinements for a certaintolerance. When the particle reaches hydrodynamic equilibrium, the higher orderschemes become inefficient since their step length is limited to a stable range. Inthis case, the mechanism switches to a stable lower order scheme and facilitateslarger integration steps.

i This mechanism is only available when Accuracy Control is enabled.Higher Order Scheme can be chosen from the group consisting of trapezoidal and runge-

kutta scheme.

Lower Order Scheme consists of implicit and the exponential analytic integration scheme.

Tracking Scheme is selectable only if Automated Tracking Scheme Selection is switchedoff. You can choose any of the tracking schemes. You also can combine each of thetracking schemes with Accuracy Control.



Including Coupled Heat-Mass Solution Effects on the Particles

By default, the solution of the particle heat and mass equations are solved in a segregatedmanner. If you enable the Coupled Heat-Mass Solution option, FLUENT will solve thispair of equations using a stiff, coupled ODE solver with error tolerance control. Theincreased accuracy, however, comes at the expense of increased computational expense.

22.11.8 User-Defined Functions

User-defined functions can be used to customize the discrete phase model to includeadditional body forces, modify interphase exchange terms (sources), calculate or integratescalar values along the particle trajectory, and incorporate nonstandard erosion ratedefinitions. See the separate UDF Manual for information about user-defined functions.

In the Discrete Phase Model panel, under User-Defined Functions in the UDF tab, thereare drop-down lists labeled Body Force, Scalar Update, Source, Spray Collide Function,and DPM Time Step (Figure 22.11.5). If Erosion/Accretion is enabled under the PhysicalModels tab, there will be an additional drop-down list labeled Erosion/Accretion. Theselists will show available user-defined functions that can be selected to customize thediscrete phase model.

In addition, you can specify a Number of Scalars which are allocated to each particle andcan be used to store information when implementing your own particle models.

22.11.9 Parallel Processing for the Discrete Phase Model

FLUENT offers two modes of parallel processing for the discrete phase model: the SharedMemory and the Message Passing options under the Parallel tab, in the Discrete PhaseModel panel. The Shared Memory option is suitable for computations where the machinerunning the FLUENT host process is an adequately large, shared memory, multiprocessormachine. The Message Passing option is turned on by default and is suitable for genericdistributed memory cluster computing.

i When tracking particles in parallel, the DPM model cannot be used withany of the multiphase flow models (VOF, mixture, or Eulerian) if the SharedMemory option is enabled. (Note that using the Message Passing option,when running in parallel, enables the compatibility of all multiphase flowmodels with the DPM model.)

The Shared Memory option is implemented using POSIX Threads (pthreads) based on ashared memory model. Once the Shared Memory option is enabled, you can then selectalong with it the Workpile Algorithm and specify the Number of Threads. By default, theNumber of Threads is equal to the number of compute nodes specified for the parallelcomputation. You can modify this value based on the computational requirements of theparticle calculations. If, for example, the particle calculations require more computation



Figure 22.11.5: The Discrete Phase Model Panel and the UDFs



than the flow calculation, you can increase the Number of Threads (up to the number ofavailable processors) to improve performance. When using the Shared Memory option, theparticle calculations are entirely managed by the FLUENT host process. You must makesure that the machine executing the host process has enough memory to accommodatethe entire grid.

i Note that the Shared Memory option is not available for Windows 2000.The Message Passing option enables cluster computing and also works on shared memorymachines. With this option enabled, the compute node processes perform the particlework on their local partitions. Particle migration to other compute nodes is implementedusing message passing primitives. There are no special requirements for the host machine.Note that this model is not available if the Cloud Model option is turned on under theTurbulent Dispersion tab of the Set Injection Properties panel. When running FLUENT inparallel, by default, pathline displays are computed in serial on the host node. Pathlinedisplays may be computed in parallel on distributed memory systems if the MessagePassing parallel option is selected in the Discrete Phase Model panel.

You may seamlessly switch between the Shared Memory option and the Message Passingoption at any time during the FLUENT session.

In addition to performing general parallel processing of the Discrete Phase Model, youhave the option of implementing DPM-specific user-defined functions in parallel FLUENT.See Section 7.4: Parallelization of Discrete Phase Model (DPM) UDFs of the separateUDF Manual for details on parallelization of DPM UDFs.


22.12 Setting Initial Conditions for the Discrete Phase


For liquid sprays, a convenient representation of the droplet size distribution is the Rosin-Rammler expression. The complete range of sizes is divided into an adequate number ofdiscrete intervals; each represented by a mean diameter for which trajectory calculationsare performed. If the size distribution is of the Rosin-Rammler type, the mass fractionof droplets of diameter greater than d is given by

Yd = e(d/d)n (22.12-1)

where d is the size constant and n is the size distribution parameter. Use of the Rosin-Rammler size distribution is detailed in Section 22.12.1: Using the Rosin-Rammler Di-ameter Distribution Method.

The primary inputs that you must provide for the discrete phase calculations in FLUENTare the initial conditions that define the starting positions, velocities, and other param-eters for each particle stream. These initial conditions provide the starting values for allof the dependent discrete phase variables that describe the instantaneous conditions ofan individual particle, and include the following:

position (x, y, z coordinates) of the particle velocities (u, v, w) of the particle

Velocity magnitudes and spray cone angle can also be used (in 3D) to define theinitial velocities (see Section 22.12.1: Point Properties for Cone Injections). Formoving reference frames, relative velocities should be specified.

diameter of the particle, dp temperature of the particle, Tp mass flow rate of the particle stream that will follow the trajectory of the individual

particle/droplet, mp (required only for coupled calculations)

additional parameters if one of the atomizer models described in Section 22.8: At-omizer Model Theory is used for the injection

i When an atomizer model is selected, you will not input initial diameter,velocity, and position quantities for the particles due to the complexitiesof sheet and ligament breakup. Instead of initial conditions, the quantitiesyou will input for the atomizer models are global parameters.

These dependent variables are updated according to the equations of motion(Section 22.2: Particle Motion Theory) and according to the heat/mass transfer relations



applied (Section 22.9: One-Way and Two-Way Coupling) as the particle/droplet movesalong its trajectory. You can define any number of different sets of initial conditions fordiscrete phase particles/droplets provided that your computer has sufficient memory.

22.12.1 Injection Types

You will define the initial conditions for a particle/droplet stream by creating an injec-tion and assigning properties to it. FLUENT provides 11 types of injections:

single group cone (only in 3D) solid-cone (only in 3D) surface plain-orifice atomizer pressure-swirl atomizer flat-fan-atomizer air-blast-atomizer effervescent-atomizer file

For each nonatomizer injection type, you will specify each of the initial conditions listedin Section 22.12: Setting Initial Conditions for the Discrete Phase, the type of particlethat possesses these initial conditions, and any other relevant parameters for the particletype chosen.

You should create a single injection when you want to specify a single value for each ofthe initial conditions (Figure 22.12.1). Create a group injection (Figure 22.12.2) whenyou want to define a range for one or more of the initial conditions (e.g., a range ofdiameters or a range of initial positions). To define hollow spray cone injections in 3Dproblems, create a cone injection (Figure 22.12.3). To release particles from a surface(either a zone surface or a surface you have defined using the items in the Surface menu),you will create a surface injection. (If you create a surface injection, a particle streamwill be released from each facet of the surface. You can use the Bounded and SamplePoints options in the Plane Surface panel to create injections from a rectangular grid ofparticles in 3D (see Section 27.6: Plane Surfaces for details).



l

Figure 22.12.1: Particle Injection Defining a Single Particle Stream

l

l

ll

Figure 22.12.2: Particle Injection Defining an Initial Spatial Distribution ofthe Particle Streams

t

tt

l t

t

Figure 22.12.3: Particle Injection Defining an Initial Spray Distribution ofthe Particle Velocity



Particle initial conditions (position, velocity, diameter, temperature, and mass flow rate)can also be read from an external file if none of the injection types listed above can beused to describe your injection distribution. The file has the following form:

(( x y z u v w diameter temperature mass-flow) name )

with all of the parameters in SI units. All the parentheses are required, but the name isoptional.

The inputs for setting injections are described in detail in Section 22.12.4: Defining In-jection Properties.

Point Properties for Single Injections

For a single injection, you will define the following initial conditions for the particlestream under the Point Properties heading (in the Set Injection Properties panel):

positionSet the x, y, and z positions of the injected stream along the Cartesian axes of theproblem geometry in the X-, Y-, and Z-Position fields. (Z-Position will appear onlyfor 3D problems.)

velocitySet the x, y, and z components of the streams initial velocity in the X-, Y-, andZ-Velocity fields. (Z-Velocity will appear only for 3D problems.)

diameterSet the initial diameter of the injected particle stream in the Diameter field.

temperatureSet the initial (absolute) temperature of the injected particle stream in the Tem-perature field.

mass flow rateFor coupled phase calculations (see Section 22.15: Solution Strategies for the Dis-crete Phase), set the mass of particles per unit time that follows the trajectorydefined by the injection in the Flow Rate field. Note that in axisymmetric problemsthe mass flow rate is defined per 2pi radians and in 2D problems per unit meterdepth (regardless of the reference value for length).

duration of injectionFor unsteady particle tracking (see Section 22.11.2: Steady/Transient Treatment ofParticles), set the starting and ending time for the injection in the Start Time andStop Time fields.



Point Properties for Group Injections

For group injections, you will define the properties described in Section 22.12.1: PointProperties for Single Injections for single injections for the First Point and Last Point inthe group. That is, you will define a range of values, 1 through N , for each initialcondition by setting values for 1 and N . FLUENT assigns a value of to the ithinjection in the group using a linear variation between the first and last values for :

i = 1 +N 1N 1 (i 1) (22.12-2)

Thus, for example, if your group consists of 5 particle streams and you define a range forthe initial x location from 0.2 to 0.6 meters, the initial x location of each stream is asfollows:

Stream 1: x = 0.2 meters Stream 2: x = 0.3 meters Stream 3: x = 0.4 meters Stream 4: x = 0.5 meters Stream 5: x = 0.6 meters

i In general, you should supply a range for only one of the initial conditionsin a given groupleaving all other conditions fixed while a single conditionvaries among the stream numbers of the group. Otherwise you may find,for example, that your simultaneous inputs of a spatial distribution anda size distribution have placed the small droplets at the beginning of thespatial range and the large droplets at the end of the spatial range.

Note that you can use a different method for defining the size distribution of the particles,as discussed below.



Using the Rosin-Rammler Diameter Distribution Method

By default, you will define the size distribution of particles by inputting a diameter forthe first and last points and using the linear equation (22.12-2) to vary the diameterof each particle stream in the group. When you want a different mass flow rate foreach particle/droplet size, however, the linear variation may not yield the distributionyou need. Your particle size distribution may be defined most easily by fitting the sizedistribution data to the Rosin-Rammler equation. In this approach, the complete rangeof particle sizes is divided into a set of discrete size ranges, each to be defined by a singlestream that is part of the group. Assume, for example, that the particle size data obeysthe following distribution:

Diameter Range (m) Mass Fraction in Range07070100100120120150150180180200

0.050.100.350.300.150.05

The Rosin-Rammler distribution function is based on the assumption that an exponentialrelationship exists between the droplet diameter, d, and the mass fraction of droplets withdiameter greater than d, Yd:

Yd = e(d/d)n (22.12-3)

FLUENT refers to the quantity d in Equation 22.12-3 as the Mean Diameter and to n asthe Spread Parameter. These parameters are input by you (in the Set Injection Propertiespanel under the First Point heading) to define the Rosin-Rammler size distribution. Tosolve for these parameters, you must fit your particle size data to the Rosin-Rammlerexponential equation. To determine these inputs, first recast the given droplet size datain terms of the Rosin-Rammler format. For the example data provided above, this yieldsthe following pairs of d and Yd:

Diameter, d (m) Mass Fraction withDiameter Greater than d, Yd

70100120150180200

0.950.850.500.200.05(0.00)



A plot of Yd vs. d is shown in Figure 22.12.4.

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

50 70 90 110 130 150 170 190 210 230 250

Mas

sFr

acti

on>

d,Y

d

Diameter, d ( m)

Figure 22.12.4: Example of Cumulative Size Distribution of Particles

Next, derive values of d and n such that the data in Figure 22.12.4 fit Equation 22.12-3.The value for d is obtained by noting that this is the value of d at which Yd = e

1 0.368.From Figure 22.12.4, you can estimate that this occurs for d 131 m. The numericalvalue for n is given by

n =ln( lnYd)ln(d/d

)By substituting the given data pairs for Yd and d/d into this equation, you can obtainvalues for n and find an average. Doing so yields an average value of n = 4.52 for theexample data above. The resulting Rosin-Rammler curve fit is compared to the exampledata in Figure 22.12.5. You can input values for d and n, as well as the diameter rangeof the data and the total mass flow rate for the combined individual size ranges, usingthe Set Injection Properties panel.



This technique of fitting the Rosin-Rammler curve to spray data is used when reportingthe Rosin-Rammler diameter and spread parameter in the discrete phase summary panelin Section 22.16.8: Summary Reporting of Current Particles.

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

50 70 90 110 130 150 170 190 210 230 250

Mas

sFr

acti

on>

d,Y

d

Diameter, d ( m)

Figure 22.12.5: Rosin-Rammler Curve Fit for the Example Particle Size Data

A second Rosin-Rammler distribution is also available based on the natural logarithm ofthe particle diameter. If in your case, the smaller-diameter particles in a Rosin-Rammlerdistribution have higher mass flows in comparison with the larger-diameter particles, youmay want better resolution of the smaller-diameter particle streams, or bins. You cantherefore choose to have the diameter increments in the Rosin-Rammler distribution doneuniformly by ln d.

In the standard Rosin-Rammler distribution, a particle injection may have a diameterrange of 1 to 200 m. In the logarithmic Rosin-Rammler distribution, the same diameterrange would be converted to a range of ln 1 to ln 200, or about 0 to 5.3. In this way, themass flow in one bin would be less-heavily skewed as compared to the other bins.



When a Rosin-Rammler size distribution is being defined for the group of streams, youshould define (in addition to the initial velocity, position, and temperature) the followingparameters, which appear under the heading for the First Point:

Total Flow RateThis is the total mass flow rate of the N streams in the group. Note that inaxisymmetric problems this mass flow rate is defined per 2pi radians and in 2Dproblems per unit meter depth.

Min. DiameterThis is the smallest diameter to be considered in the size distribution.

Max. DiameterThis is the largest diameter to be considered in the size distribution.

Mean DiameterThis is the size parameter, d, in the Rosin-Rammler equation (22.12-3).

Spread ParameterThis is the exponential parameter, n, in Equation 22.12-3.

The Stochastic Rosin-Rammler Diameter Distribution Method

For atomizer injections, a Rosin-Rammler distribution is assumed for the particles exitingthe injector. In order to decrease the number of particles necessary to accurately describethe distribution, the diameter distribution function is randomly sampled for each instancewhere new particles are introduced into the domain.

The Rosin-Rammler distribution can be written as

1 Y = exp[(D

d

)n](22.12-4)

where Y is the mass fraction smaller than a given diameter D, d is the Rosin-Rammlerdiameter and n is the Rosin-Rammler exponent. This expression can be inverted bytaking logs of both sides and rearranging,

D = d ( ln(1 Y ))1/n . (22.12-5)

Given a mass fraction Y along with parameters d and n, this function will explic-itly provide a diameter, D. Diameters for the atomizer injectors described in Sec-tion 22.12.1: Point Properties for Plain-Orifice Atomizer Injections are obtained by uni-formly sampling Y in equation 22.12-5.



Point Properties for Cone Injections

In 3D problems, you can define a hollow or solid cone of particle streams using the coneor solid-cone injection type, respectively. For both types of cone injections, the inputsare as follows:

positionSet the coordinates of the origin of the spray cone in the X-, Y-, and Z-Positionfields.

diameterSet the diameter of the particles in the stream in the Diameter field.

temperatureSet the temperature of the streams in the Temperature field.

axisSet the x, y, and z components of the vector defining the cones axis in the X-Axis,Y-Axis, and Z-Axis fields.

velocitySet the velocity magnitude of the particle streams that will be oriented along thespecified spray cone angle in the Velocity Mag. field.

cone angleSet the included half-angle, , of the hollow spray cone in the Cone Angle field, asshown in Figure 22.12.6.

r origin axis

Figure 22.12.6: Cone Half Angle and Radius



radiusA nonzero inner radius can be specified to model injectors that do not emanatefrom a single point. Set the radius r (defined as shown in Figure 22.12.6) in theRadius field. The particles will be distributed about the axis with the specifiedradius.

swirl fraction (hollow cone only)Set the fraction of the velocity magnitude to go into the swirling component ofthe flow in the Swirl Fraction field. The direction of the swirl component is definedusing the right-hand rule about the axis (a negative value for the swirl fraction canbe used to reverse the swirl direction).

mass flow rateFor coupled calculations, set the total mass flow rate for the streams in the spraycone in the Total Flow Rate field.

The distribution of the particle streams for the solid cone injection is random, as seen inFigure 22.12.3. Furthermore, duplicating this injection may not necessarily result in thesame distribution, at the same location.

i For transient calculations, the spatial distribution of streams at the ini-tial injection location is recalculated at each time step. Sampling differentpossible trajectories allows a more accurate representation of a solid coneusing fewer computational parcels. For steady state calculations, the tra-jectories are initialized one time and kept the same for subsequent DPMiterations. The trajectories are recalculated when a change in the injectionpanel occurs or when a case and data file are saved. If the residuals and so-lution change when a small change is made to the injection or when a caseand data file are saved, it may mean that there are not enough trajectoriesbeing used to represent the solid cone with sufficient accuracy.

Note that you may want to define multiple spray cones emanating from the same initiallocation in order to specify a size known distribution of the spray or to include a knownrange of cone angles.

Point Properties for Surface Injections

For surface injections, you will define all the properties described in Section 22.12.1: PointProperties for Single Injections for single injections except for the initial position of theparticle streams. The initial positions of the particles will be the location of the datapoints on the specified surface(s). Note that you will set the Total Flow Rate of allparticles released from the surface (required for coupled calculations only). If you want,you can scale the individual mass flow rates of the particles by the ratio of the area of



the face they are released from to the total area of the surface. To scale the mass flowrates, select the Scale Flow Rate By Face Area option under Point Properties.

Note that many surfaces have nonuniform distributions of points. If you want to generatea uniform spatial distribution of particle streams released from a surface in 3D, you cancreate a bounded plane surface with a uniform distribution using the Plane Surface panel,as described in Section 27.6: Plane Surfaces. In 2D, you can create a rake using theLine/Rake Surface panel, as described in Section 27.5: Line and Rake Surfaces.

In addition to the option of scaling the flow rate by the face area, the normal directionof a face can be used for the injection direction. To use the face normal direction for theinjection direction, select the Inject Using Normal Direction option under Point Properties(Figure 22.12.9). Once this option is selected, you only need to specify the velocitymagnitude of the injection, not the individual components of the velocity magnitude.

i Note also that only surface injections from boundary surfaces will be movedwith the grid when a sliding mesh or a moving or deforming mesh is beingused.

A nonuniform size distribution can be used for surface injections, as described below.

Using the Rosin-Rammler Diameter Distribution Method

The Rosin-Rammler size distributions described in Section 22.12.1: Using the Rosin-Rammler Diameter Distribution Method for group injections is also available for surfaceinjections. If you select one of the Rosin-Rammler distributions, you will need to spec-ify the following parameters under Point Properties, in addition to the initial velocity,temperature, and total flow rate:

Min. DiameterThis is the smallest diameter to be considered in the size distribution.

Max. DiameterThis is the largest diameter to be considered in the size distribution.

Mean DiameterThis is the size parameter, d, in the Rosin-Rammler equation (Equation 22.12-3).

Spread ParameterThis is the exponential parameter, n, in Equation 22.12-3.

Number of DiametersThis is the number of diameters in each distribution (i.e., the number of differentdiameters in the stream injected from each face of the surface).



FLUENT will inject streams of particles from each face on the surface, with diameters de-fined by the Rosin-Rammler distribution function. The total number of injection streamstracked for the surface injection will be equal to the number of diameters in each distri-bution (Number of Diameters) multiplied by the number of faces on the surface.

Point Properties for Plain-Orifice Atomizer Injections

For a plain-orifice atomizer injection, you will define the following initial conditions underPoint Properties:

positionSet the x, y, and z positions of the injected stream along the Cartesian axes of theproblem geometry in the X-Position, Y-Position, and Z-Position fields. (Z-Positionwill appear only for 3D problems).

axis (3D only)Set the x, y, and z components of the vector defining the axis of the orifice in theX-Axis, Y-Axis, and Z-Axis fields.


mass flow rateSet the mass flow rate for the streams in the atomizer in the Flow Rate field. Notethat in 3D sectors, the flow rate must be appropriate for the sector defined by theAzimuthal Start Angle and Azimuthal Stop Angle.


vapor pressureSet the vapor pressure governing the flow through the internal orifice (pv in Ta-ble 22.8.1) in the Vapor Pressure field.

diameterSet the diameter of the orifice in the Injector Inner Diam. field (d in Table 22.8.1).

orifice lengthSet the length of the orifice in the Orifice Length field (L in Table 22.8.1).



radius of curvatureSet the radius of curvature of the inlet corner in the Corner Radius of Curv. field (rin Table 22.8.1).

nozzle parameterSet the constant for the spray angle correlation in the Constant A field (CA inEquation 22.8-17).

azimuthal anglesFor 3D sectors, set the Azimuthal Start Angle and Azimuthal Stop Angle.

See Section 22.8.1: The Plain-Orifice Atomizer Model for details about how these inputsare used.

Point Properties for Pressure-Swirl Atomizer Injections

For a pressure-swirl atomizer injection, you will specify some of the same properties as fora plain-orifice atomizer. In addition to the position, axis (if 3D), temperature, mass flowrate, duration of injection (if unsteady), injector inner diameter, and azimuthal angles(if relevant) described in Section 22.12.1: Point Properties for Plain-Orifice AtomizerInjections, you will need to specify the following parameters under Point Properties:

spray angleSet the value of the spray angle of the injected stream in the Spray Half Angle field( in Equation 22.8-26).

pressureSet the absolute pressure upstream of the injection in the Upstream Pressure field(p1 in Table 22.8.1).

sheet breakupSet the value of the empirical constant that determines the length of the ligamentsthat are formed after sheet breakup in the Sheet Constant field (ln( b

0) in Equa-

tion 22.8-31).

ligament diameterFor short waves, set the proportionality constant that linearly relates the ligamentdiameter, dL, to the wavelength that breaks up the sheet in the Ligament Constantfield (see Equations 22.8-3222.8-35).

See Section 22.8.2: The Pressure-Swirl Atomizer Model for details about how these inputsare used.



Point Properties for Air-Blast/Air-Assist Atomizer Injections

For an air-blast/air-assist atomizer, you will specify some of the same properties as for aplain-orifice atomizer. In addition to the position, axis (if 3D), temperature, mass flowrate, duration of injection (if unsteady), injector inner diameter, and azimuthal angles(if relevant) described in Section 22.12.1: Point Properties for Plain-Orifice AtomizerInjections, you will need to specify the following parameters under Point Properties:

outer diameterSet the outer diameter of the injector in the Injector Outer Diam. field. This valueis used in conjunction with the Injector Inner Diam. to set the thickness of the liquidsheet (t in Equation 22.8-23).

spray angleSet the initial trajectory of the film as it leaves the end of the orifice in the SprayHalf Angle field ( in Equation 22.8-26).

relative velocitySet the maximum relative velocity that is produced by the sheet and air in theRelative Velocity field.

sheet breakupSet the value of the empirical constant that determines the length of the ligamentsthat are formed after sheet breakup in the Sheet Constant field (ln( b

0) in Equa-

tion 22.8-31).

ligament diameterFor short waves, set the proportionality constant (CL in Equation 22.8-34) thatlinearly relates the ligament diameter, dL, to the wavelength that breaks up thesheet in the Ligament Constant field.

See Section 22.8.3: The Air-Blast/Air-Assist Atomizer Model for details about how theseinputs are used.



Point Properties for Flat-Fan Atomizer Injections

The flat-fan atomizer model is available only for 3D models. For this type of injection,you will define the following initial conditions under Point Properties:

arc positionSet the coordinates of the center point of the arc from which the fan originates inthe X-Center, Y-Center, and Z-Center fields (see Figure 22.8.6).

virtual positionSet the coordinates of the virtual origin of the fan in the X-Virtual Origin, Y-VirtualOrigin, and Z-Virtual Origin fields. This point is the intersection of the lines thatmark the sides of the fan (see Figure 22.8.6).

normal vectorSet the direction that is normal to the fan in the X-Fan Normal Vector, Y-Fan NormalVector, and Z-Fan Normal Vector fields.


mass flow rateSet the mass flow rate for the streams in the atomizer in the Flow Rate field.


spray half angleSet the initial half angle of the drops as they leave the end of the orifice in theSpray Half Angle field.

orifice widthSet the width of the orifice (in the normal direction) in the Orifice Width field.

sheet breakupSet the value of the empirical constant that determines the length of the ligamentsthat are formed after sheet breakup in the Flat Fan Sheet Constant field (see Equa-tion 22.8-31).

See Section 22.8.4: The Flat-Fan Atomizer Model for details about how these inputs areused.



Point Properties for Effervescent Atomizer Injections

For an effervescent atomizer injection, you will specify some of the same properties asfor a plain-orifice atomizer. In addition to the position, axis (if 3D), temperature, massflow rate (including both flashing and nonflashing components), duration of injection (ifunsteady), vapor pressure, injector inner diameter, and azimuthal angles (if relevant)described in Section 22.12.1: Point Properties for Plain-Orifice Atomizer Injections, youwill need to specify the following parameters under Point Properties:

mixture qualitySet the mass fraction of the injected mixture that vaporizes in the Mixture Qualityfield (x in Equation 22.8-41).

saturation temperatureSet the saturation temperature of the volatile substance in the Saturation Temp.field.

droplet dispersionSet the parameter that controls the spatial dispersion of the droplet sizes in theDispersion Constant field (Ceff in Equation 22.8-41).

spray angleSet the initial trajectory of the film as it leaves the end of the orifice in the MaximumHalf Angle field.

See Section 22.8.5: The Effervescent Atomizer Model for details about how these inputsare used.



22.12.2 Particle Types

When you define a set of initial conditions (as described in Section 22.12.4: DefiningInjection Properties), you will need to specify the type of particle. The particle typesavailable to you depend on the range of physical models that you have defined in theModels family of panels.

An inert particle is a discrete phase element (particle, droplet, or bubble) thatobeys the force balance (Equation 22.2-1) and is subject to heating or cooling viaLaw 1 (Section 22.9.2: Inert Heating or Cooling (Law 1/Law 6)). The inert type isavailable for all FLUENT models.

A droplet particle is a liquid droplet in a continuous-phase gas flow that obeysthe force balance (Equation 22.2-1) and that experiences heating/cooling via Law1 followed by vaporization and boiling via Laws 2 and 3 (Section 22.9.2: DropletVaporization (Law 2) and Section 22.9.2: Droplet Boiling (Law 3)). The droplettype is available when heat transfer is being modeled and at least two chemicalspecies are active or the nonpremixed or partially premixed combustion model isactive. You should use the ideal gas law to define the gas-phase density (in theMaterials panel, as discussed in Section 8.3.5: Density Inputs for the IncompressibleIdeal Gas Law) when you select the droplet type.

A combusting particle is a solid particle that obeys the force balance (Equa-tion 22.2-1) and experiences heating/cooling via Law 1 followed by devolatilizationvia Law 4 (Section 22.9.2: Devolatilization (Law 4)), and a heterogeneous surfacereaction via Law 5 (Section 22.9.2: Surface Combustion (Law 5)). Finally, thenonvolatile portion of a combusting particle is subject to inert heating via Law6. You can also include an evaporating material with the combusting particle byselecting the Wet Combustion option in the Set Injection Properties panel. Thisallows you to include a material that evaporates and boils via Laws 2 and 3 (Sec-tion 22.9.2: Droplet Vaporization (Law 2) and Section 22.9.2: Droplet Boiling (Law3)) before devolatilization of the particle material begins. The combusting typeis available when heat transfer is being modeled and at least three chemical speciesare active or the nonpremixed combustion model is active. You should use the idealgas law to define the gas-phase density (in the Materials panel) when you select thecombusting particle type.

A multicomponent particle is, as the name implies, a mixture of droplet particles.These particles contain more than one component, which due to its complexity ofassigning a whole particle to one process, have to be modeled by a law that inte-grates all processes of relevance in one equation. Law 7, the multicomponent law(Section 22.9.2: Multicomponent Particle Definition (Law 7)) is used for such sys-tems. You should use the volume weighted mixing law to define the particle mixturedensity (in the Materials panel) when you select the particle-mixture material type.



22.12.3 Creating, Modifying, Copying, Deleting, and Listing Injections

You will use the Injections panel (Figure 22.12.7) to create, copy, delete, and list injections.

Define Injections...

Figure 22.12.7: The Injections Panel

(You can also click on the Injections... button in the Discrete Phase Model panel to openthe Injections panel.)

Creating Injections

To create an injection, click on the Create button. A new injection will appear in theInjections list and the Set Injection Properties panel will open automatically to allowyou to set the injection properties (as described in Section 22.12.4: Defining InjectionProperties).

Modifying Injections

To modify an existing injection, select its name in the Injections list and click on the Set...button. The Set Injection Properties panel will open, and you can modify the propertiesas needed.

If you have two or more injections for which you want to set some of the same properties,select their names in the Injections list and click on the Set... button. The Set MultipleInjection Properties panel will open, which will allow you to set the common properties.



For instructions about using this panel, see Section 22.12.7: Defining Properties Commonto More than One Injection.

Copying Injections

To copy an existing injection to a new injection, select the existing injection in theInjections list and click on the Copy button. The Set Injection Properties panel will openwith a new injection that has the same properties as the injection you selected. This isuseful if you want to set another injection with similar properties.

Deleting Injections

You can delete an injection by selecting its name in the Injections list and clicking on theDelete button.

Listing Injections

To list the initial conditions for the particle streams in the selected injection, click onthe List button. FLUENT reports the initial conditions (in SI units) in the console undervarious columns:

The particle stream number is in the column headed NO. The particle type (IN for inert, DR for droplet, or CP for combusting particle) is in

the column headed TYP.

The x, y, and z positions are in the columns headed (X), (Y), and (Z). The x, y, and z velocities are in the columns headed (U), (V), and (W). The temperature is in the column headed (T). The diameter is in the column headed (DIAM). The mass flow rate in the column headed (MFLOW).

Shortcuts for Selecting Injections

FLUENT provides a shortcut for selecting injections with names that match a specifiedpattern. To use this shortcut, enter the pattern under Injection Name Pattern and thenclick Match to select the injections with names that match the specified pattern. Forexample, if you specify drop*, all injections that have names beginning with drop (e.g.,drop-1, droplet) will be selected automatically. If they are all selected already, they willbe deselected. If you specify drop?, all surfaces with names consisting of drop followedby a single character will be selected (or deselected, if they are all selected already).



22.12.4 Defining Injection Properties

Once you have created an injection (using the Injections panel, as described in Sec-tion 22.12.3: Creating, Modifying, Copying, Deleting, and Listing Injections), you willuse the Set Injection Properties panel (Figure 22.12.8) to define the injection properties.(Remember that this panel will open when you create a new injection, or when you selectan existing injection and click on the Set... button in the Injections panel.)

Figure 22.12.8: The Set Injection Properties Panel



The procedure for defining an injection is as follows:

1. If you want to change the name of the injection from its default name, enter anew one in the Injection Name field. This is recommended if you are defining alarge number of injections so you can easily distinguish them. When assigningnames to your injections, keep in mind the selection shortcut described in Sec-tion 22.12.3: Creating, Modifying, Copying, Deleting, and Listing Injections.

2. Choose the type of injection in the Injection Type drop-down list. The eleven choices(single, group, cone, solid-cone, surface, plain-orifice-atomizer, pressure-swirl-atomizer,air-blast-atomizer, flat-fan-atomizer, effervescent-atomizer, and file) are described inSection 22.12.1: Injection Types. Note that if you select any of the atomizer models,you will also need to set the Viscosity and Droplet Surface Tension in the Materialspanel.

i If you are using sliding or moving/deforming meshes in your simulation,you should not use surface injections because they are not compatible withmoving meshes.

3. If you are defining a single injection, go to the next step. For a group, cone, solid-cone, or any of the atomizer injections, set the Number of Particle Streams in thegroup, spray cone, or atomizer.

If you are defining a surface injection (see Figure 22.12.9), choose the surface(s)from which the particles will be released in the Release From Surfaces list. If youare reading the injection from a file, click on the File... button at the bottom of theSet Injection Properties panel and specify the file to be read in the resulting SelectFile dialog box. The parameters in the injection file must be in SI units.



Figure 22.12.9: Setting Surface Injection Properties



4. Select Inert, Droplet, Combusting, or Multicomponent as the Particle Type. Theavailable types are described in Section 22.12.2: Particle Types.

5. Choose the material for the particle(s) in the Material drop-down list. If this isthe first time you have created a particle of this type, you can choose from all ofthe materials of this type defined in the database. If you have already created aparticle of this type, the only available material will be the material you selectedfor that particle. You can define additional materials by copying them from thedatabase or creating them from scratch, as discussed in Section 22.14.2: SettingDiscrete-Phase Physical Properties and described in detail in Section 8.1.2: Usingthe Materials Panel.

6. If you are defining a group or surface injection and you want to change from thedefault linear (for group injections) or uniform (for surface injections) interpola-tion method used to determine the size of the particles, select rosin-rammler orrosin-rammler-logarithmic in the Diameter Distribution drop-down list. The Rosin-Rammler method for determining the range of diameters for a group injectionis described in Section 22.12.1: Using the Rosin-Rammler Diameter DistributionMethod.

7. If you have created a customized particle law using user-defined functions, turnon the Custom option under Laws and specify the appropriate laws as described inSection 22.12.6: Custom Particle Laws.

8. If your particle type is Inert, go to the next step. If you are defining Droplet particles,select the gas phase species created by the vaporization and boiling laws (Laws 2and 3) in the Evaporating Species drop-down list.

If you are defining Combusting particles, select the gas phase species created bythe devolatilization law (Law 4) in the Devolatilizing Species drop-down list, thegas phase species that participates in the surface char combustion reaction (Law5) in the Oxidizing Species list, and the gas phase species created by the surfacechar combustion reaction (Law 5) in the Product Species list. Note that if theCombustion Model for the selected combusting particle material (in the Materialspanel) is the multiple-surface-reaction model, then the Oxidizing Species and ProductSpecies lists will be disabled because the reaction stoichiometry has been definedin the mixture material.

If you are defining Multicomponent particles, law 7 will go into effect. Notice thatthe Components tab will become active when this particle type is selected. Seebelow for information on the Components tab.



9. Click the Point Properties tab (the default), and specify the point properties (posi-tion, velocity, diameter, temperature, andif appropriatemass flow rate and anyatomizer-related parameters) as described for each injection type in Sections 22.12.122.12.1.

For surface injections, you can enable the Scale Flow Rate by Face Area and youcan choose the injection direction. To use the face normal direction for the injec-tion direction, select the Inject Using Normal Direction option under Point Properties(Figure 22.12.9). Once this option is selected, you only need to specify the ve-locity magnitude of the injection, not the individual components of the velocitymagnitude.

10. If the flow is turbulent and you wish to include the effects of turbulence on the parti-cle dispersion, click the Turbulent Dispersion tab, turn on the Stochastic Model or theCloud Model, and set the related parameters as described in Section 22.12.5: Mod-eling Turbulent Dispersion of Particles.

11. If your combusting particle includes an evaporating material, click the Wet Com-bustion tab, select the Wet Combustion option, and then select the material thatis evaporating/boiling from the particle before devolatilization begins in the Liq-uid Material drop-down list. You should also set the volume fraction of the liquidpresent in the particle by entering the value of the Liquid Fraction. Finally, selectthe gas phase species created by the evaporating and boiling laws in the EvaporatingSpecies drop-down list in the top part of the panel.

12. If you include multicomponent droplets as the material in your discrete phasemodel, a Components tab will become active. In this tab, you will specify theMass Fraction of each of the components. Note that the sum of the Mass fractionsshould add up to unity, otherwise FLUENT will adjust the values such that you havea sum of 1 for the mass fraction, and will prompt you to accept the entry. UnderVaporized Species, select not-vaporizing if the component in the particle does notvaporize. Otherwise, select from the Vaporized Species drop-down list the speciesthat will be vaporized.

To change the components of a multicomponent droplet, copy the droplet materialsfrom the Fluent Database Materials panel, or define the droplet materials in theMaterials panel, then add them to the Selected Species list in the Species panel byclicking the Edit... button (in the Materials panel) next to Mixture Species.

13. If you want to use a user-defined function to initialize the injection properties, clickthe UDF tab to access the UDF inputs. You can select an Initialization functionunder User-Defined Functions to modify injection properties at the time the particlesare injected into the domain. This allows the position and/or properties of theinjection to be set as a function of flow conditions. See the separate UDF Manualfor information about user-defined functions.



22.12.5 Modeling Turbulent Dispersion of Particles

As mentioned in Section 22.12.4: Defining Injection Properties, you can choose for eachinjection stochastic tracking or cloud tracking as the method for modeling turbulentdispersion of particles.

Stochastic Tracking

For turbulent flows, if you choose to use the stochastic tracking technique, you mustenable it and specify the number of tries. Stochastic tracking includes the effect of tur-bulent velocity fluctuations on the particle trajectories using the DRW model describedin Section 22.2.2: Stochastic Tracking.

1. Click the Turbulent Dispersion tab in the Set Injection Properties panel.

2. Enable stochastic tracking by turning on the Stochastic Model under StochasticTracking.

3. Specify the Number Of Tries:

An input of zero tells FLUENT to compute the particle trajectory based on themean continuous phase velocity field (Equation 22.2-1), ignoring the effects ofturbulence on the particle trajectories.

An input of 1 or greater tells FLUENT to include turbulent velocity fluctua-tions in the particle force balance as in Equation 22.2-20. The trajectory iscomputed more than once if your input exceeds 1: two trajectory calculationsare performed if you input 2, three trajectory calculations are performed ifyou input 3, etc. Each trajectory calculation includes a new stochastic repre-sentation of the turbulent contributions to the trajectory equation.

When a sufficient number of tries is requested, the trajectories computed willinclude a statistical representation of the spread of the particle stream due toturbulence. Note that for unsteady particle tracking, the Number of Tries isset to 1 if Stochastic Tracking is enabled.



If you want the characteristic lifetime of the eddy to be random (Equation 22.2-31),enable the Random Eddy Lifetime option. You will generally not need to change theTime Scale Constant (CL in Equation 22.2-22) from its default value of 0.15, unless youare using the Reynolds Stress turbulence model (RSM), in which case a value of 0.3 isrecommended.

Figure 22.12.10 illustrates a discrete phase trajectory calculation computed via the meantracking (number of tries = 0) and Figure 22.12.11 illustrates the stochastic tracking(number of tries > 1) option.

When multiple stochastic trajectory calculations are performed, the momentum and massdefined for the injection are divided evenly among the multiple particle/droplet tracks,and are thus spread out in terms of the interphase momentum, heat, and mass transfercalculations. Including turbulent dispersion in your model can thus have a significantimpact on the effect of the particles on the continuous phase when coupled calculationsare performed.

Particle Traces Colored by Particle Time (s)

3.04e-02

2.84e-02

2.63e-02

2.43e-02

2.23e-02

2.03e-02

1.82e-02

1.62e-02

1.42e-02

1.22e-02

1.01e-02

8.10e-03

6.08e-03

4.05e-03

2.03e-03

0.00e+00

Figure 22.12.10: Mean Trajectory in a Turbulent Flow



Particle Traces Colored by Particle Time (s)

3.00e-02

2.80e-02

2.60e-02

2.40e-02

2.20e-02

2.00e-02

1.80e-02

1.60e-02

1.40e-02

1.20e-02

1.00e-02

8.00e-03

6.00e-03

4.00e-03

2.00e-03

0.00e+00

Figure 22.12.11: Stochastic Trajectories in a Turbulent Flow

Cloud Tracking

For turbulent flows, you can also include the effects of turbulent dispersion on the injec-tion. When cloud tracking is used, the trajectory will be tracked as a cloud of particlesabout a mean trajectory, as described in Section 22.2.2: Particle Cloud Tracking.

1. Click the Turbulent Dispersion tab in the Set Injection Properties panel.

2. Enable cloud tracking by turning on the Cloud Model under Cloud Tracking.

3. Specify the minimum and maximum cloud diameters. Particles enter the domainwith an initial cloud diameter equal to the Min. Cloud Diameter. The particleclouds maximum allowed diameter is specified by the Max. Cloud Diameter.

You may want to restrict the Max. Cloud Diameter to a relevant length scale for theproblem to improve computational efficiency in complex domains where the meantrajectory may become stuck in recirculation regions.



22.12.6 Custom Particle Laws

If the standard FLUENT laws, Laws 1 through 7, do not adequately describe the physicsof your discrete phase model, you can modify them by creating custom laws with user-defined functions. See the separate UDF Manual for information about user-definedfunctions. You can also create custom laws by using a subset of the existing FLUENTlaws (e.g., Laws 1, 2, and 4), or a combination of existing laws and user-defined functions.

Once you have defined and loaded your user-defined function(s), you can create a customlaw by enabling the Custom option under Laws in the Set Injection Properties panel. Thiswill open the Custom Laws panel. In the drop-down list to the left of each of the sixparticle laws, you can select the appropriate particle law for your custom law. Each listcontains the available options that can be chosen (the standard laws plus any user-definedfunctions you have loaded).

Figure 22.12.12: The Custom Laws Panel

There is a seventh drop-down list in the Custom Laws panel labeled Switching. You maywish to have FLUENT vary the laws used depending on conditions in the model. You cancustomize the way FLUENT switches between laws by selecting a user-defined functionfrom this drop-down list.

An example of when you might want to use a custom law might be to replace the stan-dard devolatilization law with a specialized devolatilization law that more accuratelydescribes some unique aspects of your model. After creating and loading a user-definedfunction that details the physics of your devolatilization law, you would visit the CustomLaws panel and replace the standard devolatilization law (Law 2) with your user-definedfunction.



22.12.7 Defining Properties Common to More than One Injection

If you have a number of injections for which you want to set the same properties, FLUENTprovides a shortcut so that you do not need to visit the Set Injection Properties panel foreach injection to make the same changes.

As described in Section 22.12.4: Defining Injection Properties, if you select more thanone injection in the Injections panel, clicking the Set... button will open the Set MultipleInjection Properties panel (Figure 22.12.13) instead of the Set Injection Properties panel.

Figure 22.12.13: The Set Multiple Injection Properties Panel

Depending on the type of injections you have selected (single, group, atomizers, etc.),there will be different categories of properties listed under Injections Setup. The namesof these categories correspond to the headings within the Set Injection Properties panel(e.g., Particle Type and Stochastic Tracking). Only those categories that are appropriatefor all of your selected injections (which are shown in the Injections list) will be listed. Ifall of these injections are of the same type, more categories of properties will be availablefor you to modify. If the injections are of different types, you will have fewer categoriesto select from.



Modifying Properties

To modify a property, perform the following steps:

1. Select the appropriate category in the Injections Setup list. For example, if you wantto set the same flow rate for all of the selected injections, select Point Properties.The panel will expand to show the properties that appear under that heading inthe Set Injection Properties panel.

2. Set the property (or properties) to be modified, as described below.

3. Click Apply. FLUENT will report the change in the console window.

i You must click Apply to save the property settings within each category. If,for example, you want to modify the flow rate and the stochastic trackingparameters, you will need to select Point Properties in the Injections Setuplist, specify the flow rate, and click Apply. You would then repeat theprocess for the stochastic tracking parameters, clicking Apply again whenyou are done.

There are two types of properties that can be modified using the Set Multiple InjectionProperties panel.

The first type involves one of the following actions:

selecting a value from a drop-down list choosing an option using a radio button

The second type involves one of the following actions:

entering a value in a field turning an option on or off

Setting the first type of property works the same way as in the Set Injection Propertiespanel. For example, if you select Particle Type in the Injections Setup list, the panel willexpand to show the portion of the Set Injection Properties panel where you choose theparticle type. You can simply choose the desired type and click Apply.

Setting the second type of property requires an additional step. If you select a categoryin the Injections Setup list that contains this type of property, the expanded portion of thepanel will look like the corresponding part of the Set Injection Properties panel, with theaddition of Modify check buttons (see Figure 22.12.13). To change one of the properties,



first turn on the Modify check button to its left, and then specify the desired status orvalue.

For example, if you would like to enable stochastic tracking, first turn on the Modifycheck button to the left of Stochastic Model. This will make the property active so youcan modify its status. Then, under Property, turn on the Stochastic Model check button.(Be sure to click Apply when you are done setting stochastic tracking parameters.)

If you would like to change the value of Number of Tries, select the Modify check buttonto its left to make it active, and then enter the new value in the field. Make sure youclick Apply when you have finished modifying the stochastic tracking properties.

i The setting for a property that has not been activated with the Modifycheck button is not relevant, because it will not be applied to the selectedinjections when you click Apply. After you turn on Modify for a particularproperty, clicking Apply will modify that property for all of the selectedinjections, so make sure that you have the settings the way that you wantthem before you do this. If you make a mistake, you will have to returnto the Set Injection Properties panel for each injection to fix the incorrectsetting, if it is not possible to do so in the Set Multiple Injection Propertiespanel.

Modifying Properties Common to a Subset of Selected Injections

Note that it is possible to change a property that is relevant for only a subset of theselected injections. For example, if some of the selected injections are using stochastictracking and some are not, enabling the Random Eddy Lifetime option and clicking Applywill turn this option on only for those injections that are using stochastic tracking. Theother injections will be unaffected.

22.13 Setting Boundary Conditions for the Discrete Phase

When a particle reaches a physical boundary (e.g., a wall or inlet boundary) in yourmodel, FLUENT applies a discrete phase boundary condition to determine the fate of thetrajectory at that boundary. The boundary condition, or trajectory fate, can be definedseparately for each zone in your FLUENT model.


22.13 Setting Boundary Conditions for the Discrete Phase

22.13.1 Discrete Phase Boundary Condition Types

The available boundary conditions, as noted in Section 22.10: Discrete Phase Model(DPM) Boundary Conditions, include the following:

reflectThe particle rebounds the off the boundary in question with a change in its mo-mentum as defined by the coefficient of restitution. (See Figure 22.13.1.)

coefficientofrestitution =

V2,nV1,n

1 2

Figure 22.13.1: Reflect Boundary Condition for the Discrete Phase

The normal coefficient of restitution defines the amount of momentum in the di-rection normal to the wall that is retained by the particle after the collision withthe boundary [366]:

en =v2,nv1,n

(22.13-1)

where vn is the particle velocity normal to the wall and the subscripts 1 and 2 referto before and after collision, respectively. Similarly, the tangential coefficient ofrestitution, et, defines the amount of momentum in the direction tangential to thewall that is retained by the particle.

A normal or tangential coefficient of restitution equal to 1.0 implies that the particleretains all of its normal or tangential momentum after the rebound (an elasticcollision). A normal or tangential coefficient of restitution equal to 0.0 implies thatthe particl

DPM Fluent Manual

Documents

Transcript of DPM Fluent Manual