This action might not be possible to undo. Are you sure you want to continue?
Septe mber 29,
2006
Chapter23. ModelingMultiphaseFlows
This chapter discusses the general multiphase models that are available in FLUENT.
Section 23.1: Introduction provides a brief introduction to multiphase modeling, Chap
ter 22: Modeling Discrete Phase discusses the Lagrangian dispersed phase model, and
Chapter 24: Modeling Solidification and Melting describes FLUENT’s model for solidifi
cation and melting.
• Section 23.1: Introduction
• Section 23.2: Choosing a General Multiphase Model
• Section 23.3: Volume of Fluid (VOF) Model Theory
• Section 23.4: Mixture Model Theory
• Section 23.5: Eulerian Model Theory
• Section 23.6: Wet Steam Model Theory
• Section 23.7: Modeling Mass Transfer in Multiphase Flows
• Section 23.8: Modeling Species Transport in Multiphase Flows
• Section 23.9: Steps for Using a Multiphase Model
• Section 23.10: Setting Up the VOF Model
• Section 23.11: Setting Up the Mixture Model
• Section 23.12: Setting Up the Eulerian Model
• Section 23.13: Setting Up the Wet Steam Model
• Section 23.14: Solution Strategies for Multiphase Modeling
• Section 23.15: Postprocessing for Multiphase Modeling
ModelingMultiphaseFlows
232 c Flue nt Inc. Septe mber 29,
2006
23.1 In troduction
A large number of flows encountered in nature and technology are a mixture of phases.
Physical phases of matter are gas, liquid, and solid, but the concept of phase in a mul
tiphase flow system is applied in a broader sense. In multiphase flow, a phase can be
defined as an identifiable class of material that has a particular inertial response to and
interaction with the flow and the potential field in which it is immersed. For example,
differentsized solid particles of the same material can be treated as different phases be
cause each collection of particles with the same size will have a similar dynamical response
to the flow field.
23.1.1 MultiphaseFlow Regimes
Multiphase flow regimes can be grouped into four categories: gasliquid or liquidliquid
flows; gassolid flows; liquidsolid flows; and threephase flows.
GasLiquidor LiquidLiquidFlows
The following regimes are gasliquid or liquidliquid flows:
• Bubbly flow: This is the flow of discrete gaseous or fluid bubbles in a
continuous fluid.
• Droplet flow: This is the flow of discrete fluid droplets in a continuous gas.
• Slug flow: This is the flow of large bubbles in a continuous fluid.
• Stratified/freesurface flow: This is the flow of immiscible fluids separated
by a clearlydefined interface.
See Figure 23.1.1 for illustrations of these regimes.
GasSolidFlows
The following regimes are gassolid flows:
• Particleladen flow: This is flow of discrete particles in a continuous gas.
• Pneumatic trans port: This is a flow pattern that depends on factors such as
solid loading, Reynolds numbers, and particle properties. Typical patterns are
dune flow, slug flow, packed beds, and homogeneous flow.
• Fluidized bed: This consists of a vertical cylinder containing particles, into
which a gas is introduced through a distributor. The gas rising through the bed
suspends the particles. Depending on the gas flow rate, bubbles appear and rise
through the bed, intensifying the mixing within the bed.
233 c Flue nt Inc. Septe mber 29,
2006
23.1 In t r oduction
See Figure 23.1.1 for illustrations of these regimes.
LiquidSolidFlows
The following regimes are liquidsolid flows:
• Slurry flow: This flow is the trans port of particles in liquids. The
fundame ntal behavior of liquidsolid flows varies with the properties of the solid
particles relati ve to those of the liquid. In slurry flows, the Stokes number (see
Equation 23.24) is normally less than 1. When the Stokes number is larger than
1, the characteristic of the flow is liquidsolid fluidization.
• Hydrotrans port: This describes denselydistributed solid particles in a
continuous liquid
• Sedimentation: This describes a tall column initially containing a uniform
dispersed mixture of particles. At the bottom, the particles will slow down and
form a sludge layer. At the top, a clear interface will appear, and in the middle a
constant settling zone will exist.
See Figure 23.1.1 for illustrations of these regimes.
ThreePhase Flows
Threephase flows are combinations of the other flow regimes listed in the previous sec
tions.
234 c Flue nt Inc. Septe mber 29,
2006
ModelingMultiphaseFlows
slug flow
bubbl y, droplet, or
particleladen flow
stratified/f reesurface flow pneumatic transport,
hydrotransport, or slurry flow
sedimentation fluidized bed
Figure 23.1.1: Multiphase Flow Regimes
23.2 Choosinga GeneralMultiphaseModel
235 c Flue nt Inc. Septe mber 29,
2006
23.1.2 Examplesof MultiphaseSystems
Specific examples of each regime described in Section 23.1.1: Multiphase Flow Regimes
are listed below:
• Bubbly flow examples include absorbers, aeration, air lift pumps, cavitation,
evap orators, flotation, and scrubbers.
• Droplet flow examples include absorbers, atomizers, combustors, cryogenic
pump ing, dryers, evaporation, gas cooling, and scrubbers.
• Slug flow examples include large bubble motion in pipes or tanks.
• Stratified/freesurface flow examples include sloshing in offshore separator
devices and boiling and condensation in nuclear reactors.
• Particleladen flow examples include cyclone separators, air classifiers, dust
collec tors, and dustladen environme ntal flows.
• Pneumatic trans port examples include trans port of cement, grains, and metal
pow ders.
• Fluidized bed examples include fluidized bed reactors and circulating fluidized
beds.
• Slurry flow examples include slurry trans port and mineral processing
• Hydrotrans port examples include mineral processing and biomedical and
physio chemical fluid systems
• Sedimentation examples include mineral processing.
23.2 Choosing a GeneralMultiphaseModel
The first step in solving any multiphase problem is to determine which of the regimes
provides some broad guidelines for determining appropriate models for each regime, and
how to determine the degree of interphase coupling for flows involving bubbles, droplets,
or particles, and the appropriate model for different amounts of coupling.
236 c Flue nt Inc. Septe mber 29,
2006
ModelingMultiphaseFlows
23.2.1 Approaches to MultiphaseModeling
Advances in computational fluid mechanics have provided the basis for further insight
into the dynamics of multiphase flows. Currently there are two approaches for the nu
merical calculation of multiphase flows: the EulerLagrange approach (discussed in Sec
tion 22.1.1: Overview) and the EulerEuler approach (discussed in the following section).
The EulerEulerApproach
In the EulerEuler approach, the different phases are treated mathematically as inter
penetrating continua. Since the volume of a phase cannot be occupied by the other
phases, the concept of phasic volume fraction is introduced. These volume fractions are
assumed to be continuous functions of space and time and their sum is equal to one.
Conservation equations for each phase are derived to obtain a set of equations, which
have similar structure for all phases. These equations are closed by providing constituti ve
relations that are obtained from empirical information, or, in the case of granular flows,
by application of kinetic theor y.
In FLUENT, three different EulerEuler multiphase models are available: the volume
of fluid (VOF) model, the mixture model, and the Eulerian model.
TheVOFModel
The VOF model (described in Section 23.3: Volume of Fluid (VOF) Model Theory) is
a surfacetracking technique applied to a fixed Eulerian mesh. It is designed for two or
more immiscible fluids where the position of the interface between the fluids is of interest.
In the VOF model, a single set of momentum equations is shared by the fluids, and the
volume fraction of each of the fluids in each computational cell is tracked throughout the
domain. Applications of the VOF model include stratified flows, freesurface flows,
filling, sloshing, the motion of large bubbles in a liquid, the motion of liquid after a dam
break, the prediction of jet breakup (surface tension), and the steady or transie nt
tracking of any liquidgas interface.
TheMixtureModel
The mixture model (described in Section 23.4: Mixture Model Theory) is designed for two
or more phases (fluid or particulate). As in the Eulerian model, the phases are treated as
interpenetrating continua. The mixture model solves for the mixture momentum
equation and prescribes relative velocities to describe the dispersed phases.
Applications of the mixture model include particleladen flows with low loading, bubbly
flows, sedimentation, and cyclone separators. The mixture model can also be used
without relative velocities for the dispersed phases to model homogeneous multiphase
flow.
237 c Flue nt Inc. Septe mber 29,
2006
23.2 Choosinga GeneralMultiphaseModel
TheEulerianModel
The Eulerian model (described in Section 23.5: Eulerian Model Theory) is the most com
plex of the multiphase models in FLUENT. It solves a set of n momentum and continuity
equations for each phase. Coupling is achieved through the pressure and interphase ex
change coefficients. The manner in which this coupling is handled depends upon the type
of phases involved; granular (fluidsolid) flows are handled differently than nongranular
(fluidfluid) flows. For granular flows, the properties are obtained from application of ki
netic theory. Momentum exchange between the phases is also dependent upon the type
of mixture being modeled. FLUENT’suserdefined functions allow you to customize the
calculation of the momentum exchange. Applications of the Eulerian multiphase model
include bubble columns, risers, particle suspension, and fluidized beds.
23.2.2 ModelComparisons
In general, once you have determined the flow regime that best represents your multiphase
system, you can select the appropriate model based on the following guidelines:
• For bubbly, droplet, and particleladen flows in which the phases mix
and/or dispersedphase volume fractions exceed 10%, use either the mixture
model (de scribed in Section 23.4: Mixture Model Theory) or the Eulerian model
(descri bed in Section 23.5: Eulerian Model Theory).
• For slug flows, use the VOF model. See Section 23.3: Volume of Fluid (VOF)
Model
Theory for more information about the VOF model.
• For stratified/freesurface flows, use the VOF model. See Section 23.3: Volume
of
Fluid (VOF) Model Theory for more information about the VOF model.
• For pneumatic trans port, use the mixture model for homogeneous flow
(descri bed in Section 23.4: Mixture Model Theory) or the Eulerian model for
granular flow (described in Section 23.5: Eulerian Model Theory).
• For fluidized beds, use the Eulerian model for granular flow. See Section 23.5:
Eu lerian Model Theory for more information about the Eulerian model.
• For slurry flows and hydrotrans port, use the mixture or Eulerian model
(descri bed, respectively, in Sections 23.4 and 23.5).
• For sedimentation, use the Eulerian model. See Section 23.5: Eulerian
Model
Theory for more information about the Eulerian model.
• For general, complex multiphase flows that involve multiple flow regimes,
select the aspect of the flow that is of most interest, and choose the model that
is most appropriate for that aspect of the flow. Note that the accuracy of results
will not be as good as for flows that involve just one flow regime, since the model
you use will be valid for only part of the flow you are modeling.
238 c Flue nt Inc. Septe mber 29,
2006
ModelingMultiphaseFlows
As discussed in this section, the VOF model is appropriate for stratified or freesurface
flows, and the mixture and Eulerian models are appropriate for flows in which the phases
mix or separate and/or dispersedphase volume fractions exceed 10%. (Flows in which
the dispersedphase volume fractions are less than or equal to 10% can be modeled using
the discrete phase model described in Chapter 22: Modeling Discrete Phase.)
To choose between the mixture model and the Eulerian model, you should consider the
following guidelines:
• If there is a wide distribution of the dispersed phases (i.e., if the particles
vary in size and the largest particles do not separate from the primary flow
field), the mixture model may be preferable (i.e., less computationally
expensive). If the
dispersed phases are concentrated just in portions of the domain, you should use
the Eulerian model instead.
• If interphase drag laws that are applicable to your system are available
(either within FLUENT or through a userdefined function), the Eulerian model can
usually provide more accurate results than the mixture model. Even though you
can apply the same drag laws to the mixture model, as you can for a nongranular
Eulerian simulation, if the interphase drag laws are unknown or their applicability
to your system is questionable, the mixture model may be a better choice. For
most cases with spherical particles, then the SchillerNaumann law is more than
adequate. For cases with nonspherical particles, then a userdefined function can
be used.
• If you want to solve a simpler problem, which requires less computational effort,
the mixture model may be a better option, since it solves a smaller number of
equations than the Eulerian model. If accuracy is more important than
computational effort, the Eulerian model is a better choice. Keep in mind,
however, that the complexi ty of the Eulerian model can make it less
computationally stable than the mixture model.
FLUENT’s multiphase models are compatible with FLUENT’s dynamic mesh modeling
feature. For more information on the dynamic mesh feature, see Section 11: Modeling
Flows Using Sliding and Deforming Meshes. For more information about how other FLU
ENT models are compatible with FLUENT’s multiphase models, see Appendix A: FLUENT
Model Compatibili ty.
DetailedGuidelines
For stratified and slug flows, the choice of the VOF model, as indicated in Section 23.2.2:
Model Comparisons, is straightforward. Choosing a model for the other types of flows is
less straightforward. As a general guide, there are some parameters that help to identify the
appropriate multiphase model for these other flows: the particulate loading, β, and the
Stokes number, St. (Note that the word “particle” is used in this discussion to refer to
a particle, droplet, or bubble.)
γ
d
239 c Flue nt Inc. Septe mber 29,
2006
23.2 Choosinga GeneralMultiphaseModel
TheEffectof ParticulateLoading
Particulate loading has a major impact on phase interactions. The particulate loading is
defined as the mass density ratio of the dispersed phase (d) to that of the carrier phase
(c):
The material density ratio
β =
α
d
ρ
d
α
c
ρ
c
(23.21)
γ =
ρ
d
ρ
c
(23.22)
is greater than 1000 for gassolid flows, about 1 for liquidsolid flows, and less than 0.001
for gasliquid flows.
Using these parameters it is possible to estimate the average distance between the indi
vidual particles of the particulate phase. An estimate of this distance has been given by
Crowe et al. [68]:
L
π 1 + κ
1/3
=
(23.23)
d
d
6 κ
where κ =
β
. Information about these parameters is important for determining how the
dispersed phase should be treated. For example, for a gasparticle flow with a particulate
loading of 1, the interparticle space
L
d
is about 8; the particle can therefore be treated
as isolated (i.e., very low particulate loading).
Depending on the particulate loading, the degree of interaction between the phases can
be divided into the following three categories:
• For very low loading, the coupling between the phases is oneway (i.e., the
fluid carrier influences the particles via drag and turbulence, but the particles
have no influence on the fluid carrier). The discrete phase (Chapter 22: Modeling
Discrete Phase), mixture, and Eulerian models can all handle this type of problem
correctl y. Since the Eulerian model is the most expensive, the discrete phase or
mixture model is recommended.
• For intermediate loading, the coupling is twoway (i.e., the fluid carrier
influences the particulate phase via drag and turbulence, but the particles in turn
influence the carrier fluid via reduction in mean momentum and turbulence). The
discrete phase(Chapter 22: Modeling Discrete Phase) , mixture, and Eulerian
models are all applicable in this case, but you need to take into account other
factors in order to decide which model is more appropriate. See below for
information about using the Stokes number as a guide.
d
Vs
• For high loading, there is twoway coupling plus particle pressure and
viscous stresses due to particles (fourway coupling). Only the Eulerian model will
handle this type of problem correctl y.
The Significanceof the Stokes Number
For systems with intermediate particulate loading, estimating the value of the Stokes
number can help you select the most appropriate model. The Stokes number can be
defined as the relation between the particle response time and the system response time:
τ
d
St =
t
s
(23.24)
where τ
d
=
ρ
d
d
2
18µ
c
and t
s
is based on the characteristic length (L
s
) and the characteristic
velocity (V
s
) of the system under investigation: t
s
=
L
s
.
For St 1.0, the particle will follow the flow closely and any of the three models
(discrete phase(Chapter 22: Modeling Discrete Phase) , mixture, or Eulerian) is
applicable; you can therefore choose the least expensive (the mixture model, in most
cases), or the most appropriate considering other factors. For St > 1.0, the particles will
move independently of the flow and either the discrete phase model (Chapter 22:
Modeling Discrete Phase) or the Eulerian model is applicable. For St ≈ 1.0, again
any of the three models is applicable; you can choose the least expensive or the most
appropriate considering other factors.
Examples
For a coal classifier with a characteristic length of 1 m and a characteristic velocity of
10 m/s, the Stokes number is 0.04 for particles with a diameter of 30 microns, but 4.0
for particles with a diameter of 300 microns. Clearly the mixture model will not be
applicable to the latter case.
For the case of mineral processing, in a system with a characteristic length of 0.2 m and a
characteristic velocity of 2 m/s, the Stokes number is 0.005 for particles with a diameter
of 300 microns. In this case, you can choose between the mixture and Eulerian models.
(The volume fractions are too high for the discrete phase model (Chapter 22: Modeling
Discrete Phase), as noted below.)
OtherConsiderations
Keep in mind that the use of the discrete phase model (Chapter 22: Modeling Discrete
Phase) is limited to low volume fractions. Also, the discrete phase model is the only mul
tiphase model that allows you to specify the particle distribution or include combustion
modeling in your simulation.
∆t
23.2.3 TimeSchemes in MultiphaseFlow
In many multiphase applications, the process can vary spatially as well as temporally. In
order to accurately model multiphase flow, both higherorder spatial and time discretiza
tion schemes are necessary. In addition to the firstorder time scheme in FLUENT, the
secondorder time scheme is available in the Mixture and Eulerian multiphase models,
and with the VOF Implicit Scheme.
i
The secondorder time scheme cannot be used with the VOF Explicit
Schemes.
The secondorder time scheme has been adapted to all the trans port equations, includ
ing mixture phase momentum equations, energy equations, species trans port equations,
turbulence models, phase volume fraction equations, the pressure correction equation,
and the granular flow model. In multiphase flow, a general trans port equation (similar
to that of Equation 25.315) may be written as
∂(αρφ)
∂t
+ ∇ · (αρV
~
φ) = ∇ · τ + S
φ
(23.25)
Where φ is either a mixture (for the mixture model) or a phase variable, α is the phase
volume fraction (unity for the mixture equation), ρ is the mixture phase density, V
~
is
the mixture or phase velocity (depending on the equations), τ is the diffusion term, and
S
φ
is the source term.
As a fully implicit scheme, this secondorder timeaccurate scheme achieves its accuracy
by using an Euler backward approximation in time (see Equation 25.317). The general
trans port equation, Equation 23.25 is discretized as
3(α
p
ρ
p
φ
p
V ol)
n+1
− 4(α
p
ρ
p
φ
p
V ol)
n
+ (α
p
ρ
p
φ
p
)
n−1
2∆t
= (23.26)
X
[A
nb
(φ
nb
− φ
p
)]
n+1
+ S
U
n+1
− S
p
n+1
φ
p
n+1
Equation 23.26 can be written in simpler form:
A
p
φ
p
=
X
A
n
bφ
n
b + S
φ
(23.27)
where
A
p
=
P
A
nb
n+1
+ S
p
n+1
+
1.5(αp ρp V
ol)
n
n+1
n−1
S
φ
= S
U
n+1
+
2(α
p
ρ
p
φ
p
V
ol)
−0.5(αp ρp φp V ol)
∆t
This scheme is easily implemented based on FLUENT’sexisting firstorder Euler scheme.
It is unconditionally stable, however, the negative coefficient at the time level t
n−1
, of
the threetime level method, may produce oscillatory solutions if the time steps are
large.
This problem can be eliminated if a bounded secondorder scheme is introduced. How
ever, oscillating solutions are most likely seen in compressible liquid flows. Therefore,
in this version of FLUENT, a bounded secondorder time scheme has been implemented
for compressible liquid flows only. For single phase and multiphase compressible liquid
flows, the secondorder time scheme is, by default, the bounded scheme.
23.2.4 Stabilityand Convergence
The process of solving a multiphase system is inherently difficult, and you may encounter
some stability or convergence problems. If a timedependent problem is being solved, and
patched fields are used for the initial conditions, it is recommended that you perform a
few iterations with a small time step, at least an order of magnitude smaller than the
characteristic time of the flow. You can increase the size of the time step after performing
a few time steps. For steady solutions it is recommended that you start with a small
underrelaxation factor for the volume fraction, it is also recommended not to start with
a patch of volume fraction equal to zero. Another option is to start with a mixture
multiphase calculation, and then switch to the Eulerian multiphase model.
Stratified flows of immiscible fluids should be solved with the VOF model (see Sec
tion 23.3: Volume of Fluid (VOF) Model Theory). Some problems involving small
volume fractions can be solved more efficiently with the Lagrangian discrete phase
model (see Chapter 22: Modeling Discrete Phase).
Many stability and convergence problems can be minimized if care is taken during the
setup and solution processes (see Section 23.14.4: Eulerian Model).
23.3 Volumeof Fluid(VOF)ModelTheory
23.3.1 Overviewand Limitationsof the VOFModel
Overview
The VOF model can model two or more immiscible fluids by solving a single set of
momentum equations and tracking the volume fraction of each of the fluids throughout
the domain. Typical applications include the prediction of jet breakup, the motion of
large bubbles in a liquid, the motion of liquid after a dam break, and the steady or
transie nt tracking of any liquidgas interface.
23.3Volumeof Fluid(VOF)ModelTheory
Limitations
The following restrictions apply to the VOF model in
FLUENT:
• You must use the pressurebased solver. The VOF model is not available
with either of the densitybased solvers.
• All control volumes must be filled with either a single fluid phase or a
combination of phases. The VOF model does not allow for void regions where no
fluid of any type is present.
• Only one of the phases can be defined as a compressible ideal gas. There is
no limitation on using compressible liquids using userdefined functions.
• Streamwise periodic flow (either specified mass flow rate or specified pressure
drop)
cannot be modeled when the VOF model is used.
• The secondorder implicit timestepping formulation cannot be used with the
VOF
explicit scheme.
• When tracking particles in parallel, the DPM model cannot be used with the
VOF model if the shared memory option is enabled (Section 22.11.9: Parallel
Processing for the Discrete Phase Model). (Note that using the message passing
option, when running in parallel, enables the compatibili ty of all multiphase flow
models with the DPM model.)
SteadyState and TransientVOFCalculations
The VOF formulation in FLUENTis generally used to compute a timedependent solution,
but for problems in which you are concerned only with a steadystate solution, it is
possible to perform a steadystate calculation. A steadystate VOF calculation is sensible
only when your solution is independent of the initial conditions and there are distinct
inflow boundaries for the individual phases. For example, since the shape of the free
surface inside a rotating cup depends on the initial level of the fluid, such a problem
must be solved using the timedependent formulation. On the other hand, the flow of
water in a channel with a region of air on top and a separate air inlet can be solved with
the steadystate formulation.
The VOF formulation relies on the fact that two or more fluids (or phases) are not
interpenetrating. For each additional phase that you add to your model, a variable is
introduced: the volume fraction of the phase in the computational cell. In each control
volume, the volume fractions of all phases sum to unity. The fields for all variables and
properties are shared by the phases and represent volumeaveraged values, as long as
the volume fraction of each of the phases is known at each location. Thus the variables
and properties in any given cell are either purely represe ntative of one of the phases, or
ModelingMultiphaseFlows
represe ntative of a mixture of the phases, depending upon the volume fraction values.
In other words, if the q
th
fluid’s volume fraction in the cell is denoted as α
q
, then the
following three conditions are possible:
• α
q
= 0: The cell is empty (of the q
th
fluid).
• α
q
= 1: The cell is full (of the q
th
fluid).
• 0 < α
q
< 1: The cell contains the interface between the q
th
fluid and one or
more other fluids.
Based on the local value of α
q
, the appropriate properties and variables will be assigned
to each control volume within the domain.
23.3.2 VolumeFractionEquation
The tracking of the interface(s) between the phases is accomplished by the solution of a
continuity equation for the volume fraction of one (or more) of the phases. For the q
th
phase, this equation has the following form:
1
¿
∂
(α
q
ρ
q
) + ∇ · (α
q
ρ
q
~v
q
) = S
α
q
+
n
Q
X
(m˙
pq
− m˙
qp
)
\
(23.31)
ρ
q

∂t
p=1
where m˙
qp
is the mass transfer from phase q to phase p and m˙
pq
is the mass transfer
from phase p to phase q. By default, the source term on the righthand side of Equation
23.31, S
α
q
, is zero, but you can specify a constant or userdefined mass source for each
phase. See Section 23.7: Modeling Mass Transfer in Multiphase Flows for more
information on the modeling of mass transfer in FLUENT’sgeneral multiphase models.
The volume fraction equation will not be solved for the primary phase; the primaryphase
volume fraction will be computed based on the following constrai nt:
n
X
α
q
= 1 (23.32)
q=1
The volume fraction equation may be solved either through implicit or explicit time
discretization.
TheImplicitScheme
When the implicit scheme is used for time discretization, FLUENT’s standard finite
difference interpolation schemes, QUICK, Second Order Upwind and First Order
Upwind, and the Modified HRIC schemes, are used to obtain the face fluxes for all cells,
including those near the interface.
α
n+1
n+1 n n
¿
n
Q
q
ρ
q
− α
q
ρ
q
V +
X
(ρ
n+1 n+1 n+1
X
∆t
f
q
U
f
α
q,f
) =

S
αq
+
p=1
(m˙
pq
− m˙
qp
)
\
V (23.33)
Since this equation requires the volume fraction values at the current time step (rather
than at the previous step, as for the explicit scheme), a standard scalar trans port equation
is solved iteratively for each of the secondaryphase volume fractions at each time step.
The implicit scheme can be used for both timedependent and steadystate calculations.
See Section 23.10.1: Choosing a VOF Formulation for details.
The ExplicitScheme
In the explicit approach, FLUENT’s standard finitedifference interpolation schemes are
applied to the volume fraction values that were computed at the previous time step.
α
n+1
n+1 n n
¿
n
Q
q
ρ
q
− α
q
ρ
q
V +
X
(ρ U
n
α
n
)
=
X
(
m˙
m˙ ) + S V (23.34)
∆t
f
q
f q,f

p=1
pq
−
qp αq
\
where n + 1
n
=
=
index for new (curre nt) time step
index for previous time step
α
q,f
= face value of the q
th
volume fraction, computed from the first
or secondorder upwind, QUICK, modified HRIC, or CICSAM scheme
V = volume of cell
U
f
= volume flux through the face, based on normal velocity
This formulation does not require iterative solution of the trans port equation during each
time step, as is needed for the implicit scheme.
i
When the explicit scheme is used, a timedependent solution must be
com puted.
When the explicit scheme is used for time discretization, the face fluxes can be interpo
lated either using interface reconstruction or using a finite volume discretization scheme
(Section 23.3.2: Interpolation near the Interface). The reconstruction based schemes
available in FLUENT are GeoReconstruct and DonorAcceptor. The discretization schemes
available with explicit scheme for VOF are First Order Upwind, Second Order Upwind,
CICSAM,Modified HRIC, and QUICK.
Interpolationnear the Interface
FLUENT’scontrolvolume formulation requires that convection and diffusion fluxes through
the control volume faces be computed and balanced with source terms within the control
volume itself.
In the geometric reconstruction and donoracceptor schemes, FLUENT applies a spe
cial interpolation treatme nt to the cells that lie near the interface between two phases.
Figure 23.3.1 shows an actual interface shape along with the interfaces assumed during
computation by these two methods.
actual interface shape
interface shape represented by
the geometric reconstruction
(piecewiselinear) scheme
interface shape represented by
the dono racceptor scheme
Figure 23.3.1: Interface Calculations
The explicit scheme and the implicit scheme treat these cells with the same interpo
lation as the cells that are completely filled with one phase or the other (i.e., using
the standard upwind (Section 25.3.1: FirstOrder Upwind Scheme), secondorder (Sec
tion 25.3.1: SecondOrder Upwind Scheme), QUICK (Section 25.3.1: QUICK Scheme,
modified HRIC (Section 25.3.1: Modified HRIC Scheme), or CICSAM scheme), rather
than applying a special treatme nt.
The GeometricReconstructionScheme
In the geometric reconstruction approach, the standard interpolation schemes that are
used in FLUENT are used to obtain the face fluxes whenever a cell is completely filled
with one phase or another. When the cell is near the interface between two phases, the
geometric reconstruction scheme is used.
The geometric reconstruction scheme represents the interface between fluids using a
piecewiselinear approach. In FLUENTthis scheme is the most accurate and is applicable
for general unstructured meshes. The geometric reconstruction scheme is generalized
for unstructured meshes from the work of Youngs [411]. It assumes that the interface
between two fluids has a linear slope within each cell, and uses this linear shape for
calculation of the advection of fluid through the cell faces. (See Figure 23.3.1.)
The first step in this reconstruction scheme is calculating the position of the linear in
terface relative to the center of each partiallyfilled cell, based on information about
the volume fraction and its derivatives in the cell. The second step is calculating the
advecting amount of fluid through each face using the computed linear interface repre
sentation and information about the normal and tangential velocity distribution on the
face. The third step is calculating the volume fraction in each cell using the balance of
fluxes calculated during the previous step.
i
When the geometric reconstruction scheme is used, a timedependent
solu tion must be computed. Also, if you are using a conformal grid (i.e.,
if the
grid node locations are identical at the boundaries where two subdomains
meet), you must ensure that there are no twosided (zerothickness) walls
within the domain. If there are, you will need to slit them, as described in
Section 6.8.6: Slitting Face Zones.
TheDonorAcceptorScheme
In the donoracceptor approach, the standard interpolation schemes that are used in
FLUENT are used to obtain the face fluxes whenever a cell is completely filled with
one phase or another. When the cell is near the interface between two phases, a “donor
acceptor” scheme is used to determine the amount of fluid advected through the face
[144]. This scheme identifies one cell as a donor of an amount of fluid from one phase
and another (neighbor) cell as the acceptor of that same amount of fluid, and is
used to prevent numerical diffusion at the interface. The amount of fluid from one
phase that can be convected across a cell boundary is limited by the minimum of two
values: the filled volume in the donor cell or the free volume in the acceptor cell.
The orientation of the interface is also used in determining the face fluxes. The interface
orientation is either horizontal or vertical, depending on the direction of the volume
fraction gradient of the q
th
phase within the cell, and that of the neighbor cell that shares
the face in question. Depending on the interface’s orientation as well as its motion, flux
values are obtained by pure upwinding, pure downwinding, or some combination of the
two.
i
When the donoracceptor scheme is used, a timedependent solution
must be computed. Also, the donoracceptor scheme can be used
only with
quadrilateral or hexahedral meshes. In addition, if you are using a con
formal grid (i.e., if the grid node locations are identical at the boundaries
where two subdomains meet), you must ensure that there are no twosided
(zerothickness) walls within the domain. If there are, you will need to slit
them, as described in Section 6.8.6: Slitting Face Zones.
The CompressiveInterfaceCapturingScheme for Arbitrary Meshes (CICSAM)
The compressive interface capturing scheme for arbitrary meshes (CICSAM), based on
the Ubbink’s work [376], is a high resolution differencing scheme. The CICSAM scheme is
particularly suitable for flows with high ratios of viscosities between the phases. CICSAM
is implemented in FLUENT as an explicit scheme and offers the advantage of producing
an interface that is almost as sharp as the geometric reconstruction scheme.
∂t
23.3.3 Material Properties
The properties appearing in the trans port equations are determined by the presence of
the component phases in each control volume. In a twophase system, for example,
if the phases are represented by the subscripts 1 and 2, and if the volume fraction of
the second of these is being tracked, the density in each cell is given by
ρ = α
2
ρ
2
+ (1 − α
2
)ρ
1
(23.35)
In general, for an nphase system, the volumefraction averaged density takes on the
following form:
ρ =
X
α
q
ρ
q
(23.36)
All other properties (e.g., viscosity) are computed in this manner.
23.3.4 MomentumEquation
A single momentum equation is solved throughout the domain, and the resulting velocity
field is shared among the phases. The momentum equation, shown below, is dependent
on the volume fractions of all phases through the properties ρ and µ.
∂
(ρ~v) + ∇ · (ρ~v~v) = −∇p + ∇ ·
h
µ
∇~v + ∇~v
T
i
+
ρ~g + F
~
(23.37)
One limitation of the sharedfields approximation is that in cases where large velocity
differences exist between the phases, the accuracy of the velocities computed near the
interface can be adversely affected.
Note that if the viscosity ratio is more than 1x10
3
, this may lead to convergence diffi
culties. The compressive interface capturing scheme for arbitrary meshes (CICSAM)
(Section 23.3.2: The Compressive Interface Capturing Scheme for Arbitrary Meshes
(CICSAM)) is suitable for flows with high ratios of viscosities between the phases, thus
solving the problem of poor convergence.
n
23.3.5 Energy Equation
The energy equation, also shared among the phases, is shown below.
∂
∂t
(ρE) + ∇ · (~v(ρE + p)) = ∇ · (k
eff
∇T ) + S
h
(23.38)
The VOF model treats energy, E, and temperature, T , as massaveraged variables:
n
X
α
q
ρ
q
E
q
E =
q=1
X
α
q
ρ
q
q=1
(23.39)
where E
q
for each phase is based on the specific heat of that phase and the shared
temperature.
The properties ρ and k
eff
(effective thermal conductivi ty) are shared by the phases. The
source term, S
h
, contains contributions from radiation, as well as any other volumetric
heat sources.
As with the velocity field, the accuracy of the temperature near the interface is limited in
cases where large temperature differences exist between the phases. Such problems also
arise in cases where the properties vary by several orders of magnitude. For example, if a
model includes liquid metal in combination with air, the conductivities of the materials
can differ by as much as four orders of magnitude. Such large discrepancies in properties
lead to equation sets with anisotropic coefficients, which in turn can lead to convergence
and precision limitations.
23.3.6 Additional Scalar Equations
Depending upon your problem definition, additional scalar equations may be involved in
your solution. In the case of turbulence quantities, a single set of trans port equations is
solved, and the turbulence variables (e.g., k and or the Reynolds stresses) are shared
by the phases throughout the field.
23.3.7 TimeDependence
For timedependent VOF calculations, Equation 23.31 is solved using an explicit time
marching scheme. FLUENTautomatically refines the time step for the integration of the
volume fraction equation, but you can influence this time step calculation by modifying
the Courant number. You can choose to update the volume fraction once for each time
step, or once for each iteration within each time step. These options are discussed in
more detail in Section 23.10.4: Setting TimeDependent Parameters for the VOF Model.
R R
23.3.8 Surface Tension and Wall
Adhesion
The VOF model can also include the effects of surface tension along the interface between
each pair of phases. The model can be augmented by the additional specification of the
contact angles between the phases and the walls. You can specify a surface tension
coefficient as a consta nt, as a function of temperature, or through a UDF. The solver
will include the additional tangential stress terms (causing what is termed as Marangoni
convection) that arise due to the variation in surface tension coefficient. Variable surface
tension coefficient effects are usually important only in zero/nearzero gravity conditions.
SurfaceTension
Surface tension arises as a result of attracti ve forces between molecules in a fluid. Con
sider an air bubble in water, for example. Within the bubble, the net force on a molecule
due to its neighbors is zero. At the surface, however, the net force is radially inward, and
the combined effect of the radial components of force across the entire spherical surface
is to make the surface contract, thereby increasing the pressure on the concave side of
the surface. The surface tension is a force, acting only at the surface, that is required
to maintain equilibrium in such instances. It acts to balance the radially inward inter
molecular attracti ve force with the radially outward pressure gradient force across the
surface. In regions where two fluids are separated, but one of them is not in the form
of spherical bubbles, the surface tension acts to minimize free energy by decreasing the
area of the interface.
The surface tension model in FLUENT is the continuum surface force (CSF) model pro
posed by Brackbill et al. [39]. With this model, the addition of surface tension to the
VOF calculation results in a source term in the momentum equation. To understand the
origin of the source term, consider the special case where the surface tension is constant
along the surface, and where only the forces normal to the interface are considered. It can
be shown that the pressure drop across the surface depends upon the surface tension co
efficient, σ, and the surface curvature as measured by two radii in orthogonal directions,
R
1
and R
2
:
1 1
p
2
− p
1
= σ +
1 2
(23.310)
where p
1
and p
2
are the pressures in the two fluids on either side of the interface.
In FLUENT, a formulation of the CSF model is used, where the surface curvature is
computed from local gradients in the surface normal at the interface. Let n be the
surface normal, defined as the gradient of α
q
, the volume fraction of the q
th
phase.
n = ∇α
q
(23.311)
1
The curvature, κ, is defined in terms of the divergence of the unit normal, nˆ [39]:
κ = ∇ · nˆ (23.312)
where
nˆ =
n
n
(23.313)
The surface tension can be written in terms of the pressure jump across the surface. The
force at the surface can be expressed as a volume force using the divergence theorem. It
is this volume force that is the source term which is added to the momentum equation.
It has the following form:
F
vol
=
X
pairs ij, i<j
σ
ij
α
i
ρ
i
κ
j
∇α
j
+ α
j
ρ
j
κ
i
∇α
i
2
(ρ
i
+ ρ
j
)
(23.314)
This expression allows for a smooth superposition of forces near cells where more than
two phases are present. If only two phases are present in a cell, then κ
i
= −κ
j
and
∇α
i
= −∇α
j
, and Equation 23.314 simplifies to
ρκ
i
∇α
i
F
vol
= σ
ij
1
2
(ρ
i
+ ρ
j
)
(23.315)
where ρ is the volumeaveraged density computed using Equation 23.36. Equation 23.315
shows that the surface tension source term for a cell is proportional to the average densi ty
in the cell.
Note that the calculation of surface tension effects on triangular and tetrahedral meshes
is not as accurate as on quadrilateral and hexahedral meshes. The region where surface
tension effects are most important should therefore be meshed with quadrilaterals or
hexahedra.
WhenSurfaceTension EffectsAreImportant
The importance of surface tension effects is determined based on the value of two di
mensionless quantities: the Reynolds number, Re, and the capillary number, Ca; or the
Reynolds number, Re, and the Weber number, We. For Re 1, the quantity of
interest is the capillary number:
µU
Ca =
σ
(23.316)
and for Re 1, the quantity of interest is the Weber number:
We =
ρLU
2
σ
(23.317)
where U is the freestream velocity. Surface tension effects can be neglected if Ca 1
or We 1.
Several surface tension options are provided through the text user interface (TUI) using
the solve/set/surfacetension command:
solve −→ set −→surfacetension
The surfacetension command prompts you for the following information:
• whether you require nodebased smoothing
The default value is no indicating that cellbased smoothing will be used for the
VOF calculations.
• the number of smoothings
The default value is 1. A higher value can be used in case of tetrahedral
and triangular meshes in order to reduce any spurious velocities.
• the smoothing relaxation factor
The default is 1. This is useful in the cases where VOF smoothing causes a problem
(e.g., liquid enters through the inlet with wall adhesion on).
• whether you want to use VOF gradients at the nodes for curvature calculations
With this option, FLUENTuses VOF gradients directly from the nodes to calculate
the curvature for surface tension forces. The default is yes which produces better
results with surface tension compared to gradients that are calculated at the cell
centers.
Wall Adhesion
An option to specify a wall adhesion angle in conjunction with the surface tension model
is also available in the VOF model. The model is taken from work done by Brackbill et
al. [39]. Rather than impose this boundary condition at the wall itself, the contact angle
that the fluid is assumed to make with the wall is used to adjust the surface normal in
cells near the wall. This socalled dynamic boundary condition results in the adjustme nt
of the curvature of the surface near the wall.
If θ
w
is the contact angle at the wall, then the surface normal at the live cell next to the
wall is
nˆ = nˆ
w
cos θ
w
+ t
ˆ
w
sin θ
w
(23.318)
where nˆ
w
and t
ˆ
w
are the unit vectors normal and tangential to the wall,
respectively. The combination of this contact angle with the normally calculated
surface normal one cell away from the wall determine the local curvature of the surface,
and this curvature is used to adjust the body force term in the surface tension
calculation.
The contact angle θ
w
is the angle between the wall and the tangent to the interface
at the wall, measured inside the phase listed in the left column under Wall Adhesion in
the Momentum tab of the Wall panel. For example, if you are setting the contact angle
between the oil and air phases in the Wall panel shown in Figure 23.3.2, θ
w
is measured
inside the oil phase, as seen in Figure 23.3.3.
Figure 23.3.2: The Wall Panel for a Mixture in a VOF Calculation with Wall
Adhesion
o
interface
AIR
OIL
OR
θW
= 30
AIR
θ = 30
o
interface
OIL
wall
W
wall
Figure 23.3.3: Measuring the Contact Angle
23.3.9 OpenChannelFlow
FLUENT can model the effects of open channel flow (e.g., rivers, dams, and surface
piercing structures in unbounded stream) using the VOF formulation and the open chan
nel boundary condition. These flows involve the existence of a free surface between the
flowing fluid and fluid above it (generally the atmosphere). In such cases, the wave prop
agation and free surface behavior becomes important. Flow is generally governed by the
forces of gravity and inertia. This feature is mostly applicable to marine applications
and the analysis of flows through drainage systems.
Open channel flows are characterized by the dimensionless Froude Number, which is
defined as the ratio of inertia force and hydrostatic force.
V
F r = √
gy
(23.319)
where V is the velocity magnitude, g is gravity, and y is a length scale, in this case,
the distance from the bottom of the channel to the free surface. The denominator in
Equation 23.319 is the propagation speed of the wave. The wave speed as seen by the
fixed observer is defined as
V
w
= V ±
√
gy (23.320)
Based on the Froude number, open channel flows can be classified in the following three
categories:
• When F r < 1, i.e., V <
√
gy (thus V
w
< 0 or V
w
> 0), the flow is known to
be subcritical where disturbances can travel upstream as well as downstream. In
this case, downstream conditions might affect the flow upstream.
• When F r = 1 (thus V
w
= 0), the flow is known to be critical, where
upstream propagating waves remain stationar y. In this case, the character of the
flow changes.
• When F r > 1, i.e., V >
√
gy (thus V
w
> 0), the flow is known to be
supercriti cal where disturbances cannot travel upstream. In this case,
downstream conditions do not affect the flow upstream.
UpstreamBoundaryConditions
There are two options available for the upstream boundary condition for open channel
flows:
• pressure inlet
• mass flow rate
PressureInlet
The total pressure p
0
at the inlet can be given as
1
2
p
0
=
2
(ρ − ρ
0
)V
+ (ρ − ρ
0
)
−→
g (gˆ · (
−→
b −
−→
a )) (23.3
21)
where
−→
b and
−→
a are the position vectors of the face centroid and any point on
the free
surface, respectively, Here, free surface is assumed to be horizontal and normal to the
direction of gravity.
−→
g is the gravity vector, 
−→
g  is the gravity magnitude, gˆ is
the unit
vector of gravity, V is the velocity magnitude, ρ is the density of the mixture in the cell,
and ρ
0
is the reference densi ty.
From this, the dynamic pressure q is
and the static pressure p
s
is
q =
ρ − ρ
0
2
V
2
(23.322)
p
s
= (ρ − ρ
0
)
−→
g (gˆ · (
−→
b −
−→
a )) (23.323)
which can be further expanded to
p
s
= (ρ − ρ
0
)
−→
g ((gˆ ·
−→
b ) + y
local
) (23.324)
where the distance from the free surface to the reference position, y
local
, is
y
local
= −(
−→
a · gˆ) (23.325)
MassFlow Rate
The mass flow rate for each phase associated with the open channel flow is defined by
m˙
phase
= ρ
phase
(Area
phase
)(V elocity) (23.326)
VolumeFractionSpecification
In open channel flows, FLUENT internally calculates the volume fraction based on the
input parameters specified in the Boundary Conditions panel, therefore this option has
been disabled.
For subcritical inlet flows (Fr < 1), FLUENTreconstructs the volume fraction values on
the boundary by using the values from the neighboring cells. This can be accomplished
using the following procedure:
• Calculate the node values of volume fraction at the boundary using the cell
values.
• Calculate the volume fraction at the each face of boundary using the
interpolated node values.
For supercritical inlet flows (Fr > 1), the volume fraction value on the boundary can
be calculated using the fixed height of the free surface from the bottom.
DownstreamBoundaryConditions
PressureOutlet
Determining the static pressure is dependent on the Pressure Specification Method. Using
the Free Surface Level, the static pressure is dictated by Equation 23.323 and Equa
tion 23.325, otherwise you must specify the static pressure as the Gauge Pressure.
For subcritical outlet flows (Fr < 1), if there are only two phases, then the pressure is
taken from the pressure profile specified over the boundary, otherwise the pressure is
taken from the neighboring cell. For supercritical flows (Fr >1), the pressure is always
taken from the neighboring cell.
23.4 MixtureModelTheory
Outfl ow Boundary
Outflow boundary conditions can be used at the outlet of open channel flows to model
flow exits where the details of the flow velocity and pressure are not known prior to
solving the flow problem. If the conditions are unknown at the outflow boundaries, then
FLUENTwill extrapolate the required information from the interior.
It is important, however, to understand the limitations of this boundary type:
• You can only use single outflow boundaries at the outlet, which is achieved by
set ting the flow rate weighting to 1. In other words, outflow splitting is not
permitted
in open channel flows with outflow boundaries.
• There should be an initial flow field in the simulation to avoid convergence
issues due to flow reversal at the outflow, which will result in an unreliable
solution.
• An outflow boundary condition can only be used with mass flow inlets. It is
not compatible with pressure inlets and pressure outlets. For example, if you
choose the inlet as pressureinlet, then you can only use pressureoutlet at the outlet.
If you choose the inlet as massflowinlet, then you can use either outflow or
pressureoutlet boundary conditions at the outlet. Note that this only holds true
for open channel flow.
• Note that the outflow boundary condition assumes that flow is fully
developed in the direction perpendicular to the outflow boundary surface.
Therefore, such surfaces should be placed accordingly.
Backflow VolumeFractionSpecification
FLUENTinternally calculates the volume fraction values on the outlet boundary by using
the neighboring cell values, therefore, this option is disabled.
23.4 MixtureModelTheory
23.4.1 Overviewand Limitationsof the MixtureModel
Overview
The mixture model is a simplified multiphase model that can be used to model multiphase
flows where the phases move at different velocities, but assume local equilibrium
over short spatial length scales. The coupling between the phases should be strong. It
can also be used to model homogeneous multiphase flows with very strong coupling and
the phases moving at the same velocity. In addition, the mixture model can be
used to calculate nonNewtonian viscosity.
ModelingMultiphaseFlows
The mixture model can model n phases (fluid or particulate) by solving the momentum,
continuity, and energy equations for the mixture, the volume fraction equations for the
secondary phases, and algebraic expressions for the relative velocities. Typical applica
tions include sedimentation, cyclone separators, particleladen flows with low loading,
and bubbly flows where the gas volume fraction remains low.
The mixture model is a good substitute for the full Eulerian multiphase model in several
cases. A full multiphase model may not be feasible when there is a wide distribution of
the particulate phase or when the interphase laws are unknown or their reliability can
be questioned. A simpler model like the mixture model can perform as well as a full
multiphase model while solving a smaller number of variables than the full multiphase
model.
The mixture model allows you to select granular phases and calculates all properties of
the granular phases. This is applicable for liquidsolid flows.
Limitations
The following limitations apply to the mixture model in FLUENT:
• You must use the pressurebased solver. The mixture model is not available
with either of the densitybased solvers.
• Only one of the phases can be defined as a compressible ideal gas. There is
no limitation on using compressible liquids using userdefined functions.
• Streamwise periodic flow with specified mass flow rate cannot be modeled
when the mixture model is used (the user is allowed to specify a pressure drop).
• Solidification and melting cannot be modeled in conjunction with the
mixture model.
• The LES turbulence model cannot be used with the mixture model if the
cavitation model is enabled.
• The relative velocity formulation cannot be used in combination with the MRF
and mixture model (see Section 10.3.1: Limitations ).
• The mixture model cannot be used for inviscid flows.
• The shell conduction model for walls cannot be used with the mixture model.
• When tracking particles in parallel, the DPM model cannot be used with the
mix ture model if the shared memory option is enabled (Section 22.11.9: Parallel
Pro cessing for the Discrete Phase Model). (Note that using the message passing
option, when running in parallel, enables the compatibili ty of all multiphase flow
models with the DPM model.)
23.4 MixtureModelTheory
The mixture model, like the VOF model, uses a singlefluid approach. It differs from the
VOF model in two respects:
• The mixture model allows the phases to be interpenetrating. The volume
fractions
α
q
and α
p
for a control volume can therefore be equal to any value between 0 and
1, depending on the space occupied by phase q and phase p.
• The mixture model allows the phases to move at different velocities, using
the concept of slip velocities. (Note that the phases can also be assumed to
move at the same velocity, and the mixture model is then reduced to a
homogeneous multiphase model.)
The mixture model solves the continuity equation for the mixture, the momentum equa
tion for the mixture, the energy equation for the mixture, and the volume fraction equa
tion for the secondary phases, as well as algebraic expressions for the relative velocities
(if the phases are moving at different velocities).
23.4.2 ContinuityEquation
The continuity equation for the mixture is
∂
∂t
(ρ
m
) + ∇ · (ρ
m
~v
m
) = 0 (23.41)
where ~v
m
is the massaveraged velocity:
P
n
α
k
ρ
k
~v
k
and ρ
m
is the mixture densi ty:
~v
m
=
k =1
ρ
m
(23.42)
n
ρ
m
=
X
α
k
ρ
k
(23.43)
k=1
α
k
is the volume fraction of phase k.
m
k k
ρ
k
∂
∂t
23.4.3 MomentumEquation
The momentum equation for the mixture can be obtained by summing the individual
momentum equations for all phases. It can be expressed as
(ρ
m
~v
m
) + ∇ · (ρ
m
~v
m
~v
m
) = −∇p + ∇ ·
h
µ
m
∇~v
m
+ ∇~v
T
i
+
ρ
m
~g + F
~
+ ∇ ·
n
!
X
α
k
ρ
k
~v
dr,k
~v
dr,k k=1 (23.44)
where n is the number of phases, F
~
is a body force, and µ
m
is the viscosity of the
mixture:
n
µ
m
=
X
α
k
µ
k
(23.45)
k=1
~v
dr,k
is the drift velocity for secondary phase k:
~v
dr,k
= ~v
k
− ~v
m
(23.46)
23.4.4 Energy Equation
The energy equation for the mixture takes the following form:
∂
n n
X
(α
k
ρ
k
E
k
) + ∇ ·
X
(α
k
~v
k
(ρ
k
E
k
+ p)) = ∇ · (k
eff
∇T ) + S
E
(23.47)
∂t
k=1 k=1
where k
eff
is the effective conductivity (
P
α
k
(k
k
+ k
t
)), where k
t
is the turbule nt thermal
conductivi ty, defined according to the turbulence model being used). The first term on
the righthand side of Equation 23.47 represents energy transfer due to conduction. S
E
includes any other volumetric heat sources.
In Equation 23.47,
E = h −
p
k
v
2
+ (23.48)
2
for a compressible phase, and E
k
= h
k
for an incompressible phase, where h
k
is the
sensible enthalpy for phase k.
p
23.4.5 Relative(Slip)Velocityand the DriftVelocity
The relative velocity (also referred to as the slip velocity) is defined as the velocity of a
secondary phase (p) relative to the velocity of the primary phase (q):
~v
pq
= ~v
p
− ~v
q
(23.49)
The mass fraction for any phase (k) is defined as
c
k
=
α
k
ρ
k
ρ
m
(23.410)
The drift velocity and the relative velocity (~v
qp
) are connected by the following
expression:
n
~v
dr,p
= ~v
pq
−
X
c
k
~v
qk
(23.411)
k=1
FLUENT’smixture model makes use of an algebraic slip formulation. The basic assump
tion of the algebraic slip mixture model is that to prescribe an algebraic relation for the
relative velocity, a local equilibrium between the phases should be reached over
short spatial length scale. Following Manninen et al. [229], the form of the relative
velocity is given by:
τ
p
~v
pq
=
f
(ρ
p
− ρ
m
)
ρ
~a (23.412)
drag p
where τ
p
is the particle relaxation time
τ
p
=
ρ
p
d
2
18µ
q
(23.413)
d is the diameter of the particles (or droplets or bubbles) of secondary phase p, ~a is
the secondaryphase particle’s acceleration. The default drag function f
drag
is taken
from Schiller and Naumann [320]:
f
drag
=
(
1 + 0.15 Re
0.687
Re ≤ 1000
0.0183 Re Re > 1000
(23.414)
and the acceleration ~a is of the form
~a = ~g − (~v
m
·
∇)~v
m
−
∂~v
m
(23.415)
∂t
p
~a
−
∂
∂t
The simplest algebraic slip formulation is the socalled drift flux model, in which the ac
celeration of the particle is given by gravity and/or a centrifugal force and the particulate
relaxation time is modified to take into account the presence of other particles.
In turbule nt flows the relative velocity should contain a diffusion term due to the
disper sion appearing in the momentum equation for the dispersed phase. FLUENT
adds this dispersion to the relative velocity:
~v
pq
=
(ρ
p
− ρ
m
)d
2
18µ
q
f
drag
ν
m
α
p
σ
D
∇α
q
(23.416)
where (ν
m
) is the mixture turbule nt viscosity and (σ
D
) is a Prandtl dispersion coefficient.
When you are solving a mixture multiphase calculation with slip velocity, you can directly
prescribe formulations for the drag function. The following choices are available:
• SchillerNaumann (the default formulation)
• MorsiAlexander
• symmetric
• constant
• userdefined
See Section 23.5.4: Interphase Exchange Coefficients for more information on these drag
functions and their formulations, and Section 23.11.1: Defining the Phases for the Mixture
Model for instructions on how to enable them.
Note that, if the slip velocity is not solved, the mixture model is reduced to a homogeneous
multiphase model. In addition, the mixture model can be customized (using userdefined
functions) to use a formulation other than the algebraic slip method for the slip velocity.
See the separate UDF Manual for details.
23.4.6 VolumeFractionEquationfor the SecondaryPhases
From the continuity equation for secondary phase p, the volume fraction equation for
secondary phase p can be obtained:
n
(α
p
ρ
p
) + ∇ · (α
p
ρ
p
~v
m
) = −∇ · (α
p
ρ
p
~v
dr,p
) +
X
(m˙
qp
− m˙
pq
) (23.417)
q=1
5
s
23.4.7 GranularProperties
Since the concentration of particles is an important factor in the calculation of the effec
tive viscosity for the mixture, we may use the granular viscosity (see section on Eulerian
granular flows) to get a value for the viscosity of the suspension. The volume weighted
averaged for the viscosity would now contain shear viscosity arising from particle mo
mentum exchange due to translation and collision.
The collisional and kinetic parts, and the optional frictional part, are added to give the
solids shear viscosity:
µ
s
= µ
s,col
+ µ
s,kin
+ µ
s,fr
(23.418)
Collisional Viscosity
The collisional part of the shear viscosity is modeled as [119, 363]
KineticViscosity
4
µ
s,col
=
5
α
s
ρ
s
d
s
g
0,ss
(1 + e
ss
)
Θ
s
1/2
π
(23.419)
FLUENTprovides two expressions for the kinetic viscosity.
The default expression is from Syamlal et al. [363]:
α
s
d
s
ρ
s
√
Θ
s
π
2
µ
s,kin
=
6 (3 − e
ss
)
1 + (1 + e
ss
) (3e
ss
− 1) α
s
g
0,ss
(23.420)
The following optional expression from Gidaspow et al. [119] is also available:
10ρ
s
d
s
√
Θ
s
π 4
2
µ
s,kin
=
96α (1 + e
ss
) g
0,ss
1 + g
0,ss
α
s
(1 + e
ss
)
5
(23.421)
ss
23.4.8 GranularTemperature
The viscosities need the specification of the granular temperature for the s
th
solids phase.
Here we use an algebraic equation derived from the trans port equation by neglecting
convection and diffusion and takes the form [363]
where
0 = (−p
s
I + τ
s
) : ∇~v
s
− γ
Θ
s
+ φ
ls
(23.422)
(−p
s
I + τ
s
) : ∇~v
s
= the generation of energy by the solid stress tensor
γ
Θ
s
= the collisional dissipation of energy
φ
ls
= the energy exchange between the l
th
fluid or solid phase and the s
th
solid phase
The collisional dissipation of energy, γ
Θ
s
, represents the rate of energy dissipation within
the s
th
solids phase due to collisions between particles. This term is represented by the
expression derived by Lun et al. [221]
γ
Θm
=
12(1 − e
2
)g
0,ss
ρ
s
α
2
Θ
3/2
(23.423)
d
s
√
π
s s
The transfer of the kinetic energy of random fluctuations in particle velocity from the s
th
solids phase to the l
th
fluid or solid phase is represented by φ
ls
[119]:
φ
ls
= −3K
ls
Θ
s
(23.424)
FLUENTallows you to solve for the granular temperature with the following options:
• algebraic formulation (the default)
This is obtained by neglecting convection and diffusion in the trans port equation
(Equation 23.422) [363].
• constant granular temperature
This is useful in very dense situations where the random fluctuations are small.
• UDF for granular temperature
23.4.9 Solids Pressure
The total solid pressure is calculated and included in the mixture momentum equations:
N
P
s,total
=
X
p
q
(23.425)
q=1
where p
q
is presented in the section for granular flows by equation Equation 23.548
23.5EulerianModelTheory
23.5 EulerianModelTheory
Details about the Eulerian multiphase model are presented in the following subsections:
• Section 23.5.1: Overview and Limitations of the Eulerian Model
• Section 23.5.2: Volume Fractions
• Section 23.5.3: Conservation Equations
• Section 23.5.4: Interphase Exchange Coefficients
• Section 23.5.5: Solids Pressure
• Section 23.5.6: Maximum Packing Limit in Binary Mixtures
• Section 23.5.7: Solids Shear Stresses
• Section 23.5.8: Granular Temperature
• Section 23.5.9: Description of Heat Transfer
• Section 23.5.10: Turbulence Models
• Section 23.5.11: Solution Method in FLUENT
23.5.1 Overviewand Limitationsof the EulerianModel
Overview
The Eulerian multiphase model in FLUENT allows for the modeling of multiple sepa
rate, yet interacting phases. The phases can be liquids, gases, or solids in nearly any
combination. An Eulerian treatme nt is used for each phase, in contrast to the Eulerian
Lagrangian treatme nt that is used for the discrete phase model.
With the Eulerian multiphase model, the number of secondary phases is limited only
by memory requirements and convergence behavior. Any number of secondary phases
can be modeled, provided that sufficient memory is available. For complex multiphase
flows, however, you may find that your solution is limited by convergence behavior.
See Section 23.14.4: Eulerian Model for multiphase modeling strategies.
FLUENT’sEulerian multiphase model does not distinguish between fluidfluid and fluid
solid (granular) multiphase flows. A granular flow is simply one that involves at least
one phase that has been designated as a granular phase.
ModelingMultiphaseFlows
The FLUENTsolution is based on the following:
• A single pressure is shared by all phases.
• Momentum and continuity equations are solved for each phase.
• The following parameters are available for granular phases:
– Granular temperature (solids fluctuating energy) can be calculated for each
solid phase. You can select either an algebraic formulation, a consta nt, a
userdefined function, or a partial differential equation.
– Solidphase shear and bulk viscosities are obtained by applying kinetic the
ory to granular flows. Frictional viscosity for modeling granular flow is also
available. You can select appropriate models and userdefined functions for
all properties.
• Several interphase drag coefficient functions are available, which are
appropriate for various types of multiphase regimes. (You can also modify the
interphase drag coefficient through userdefined functions, as described in the
separate UDF Man ual.)
• All of the k turbulence models are available, and may apply to all phases or
to the mixture.
Limitations
All other features available in FLUENT can be used in conjunction with the Eulerian
multiphase model, except for the following limitations:
• The Reynolds Stress turbulence model is not available on a per phase basis.
• Particle tracking (using the Lagrangian dispersed phase model) interacts only
with the primary phase.
• Streamwise periodic flow with specified mass flow rate cannot be modeled
when the Eulerian model is used (the user is allowed to specify a pressure drop).
• Inviscid flow is not allowed.
• Melting and solidification are not allowed.
• When tracking particles in parallel, the DPM model cannot be used with the
Eule rian multiphase model if the shared memory option is enabled (Section
22.11.9: Par allel Processing for the Discrete Phase Model). (Note that using the
message pass ing option, when running in parallel, enables the compatibili ty of
all multiphase flow models with the DPM model.)
23.5EulerianModelTheory
To change from a singlephase model, where a single set of conservation equations for
momentum, continuity and (optionally) energy is solved, to a multiphase model, addi
tional sets of conservation equations must be introduced. In the process of introduc
ing additional sets of conservation equations, the original set must also be modified.
The modifications involve, among other things, the introduction of the volume fractions
α
1
, α
2
, . . . α
n
for the multiple phases, as well as mechanisms for the exchange of
momen tum, heat, and mass between the phases.
23.5.2 VolumeFractions
The description of multiphase flow as interpenetrating continua incorporates the concept
of phasic volume fractions, denoted here by α
q
. Volume fractions represent the
space occupied by each phase, and the laws of conservation of mass and momentum are
satisfied by each phase individually. The derivation of the conservation equations can
be done by ensemble averaging the local instantaneous balance for each of the phases
[10] or by using the mixture theory approach [36].
The volume of phase q, V
q
, is defined by
Z
V
q
=
V
α
q
dV (23.51)
where
n
X
α
q
= 1 (23.52)
q=1
The effective density of phase q is
ρˆ
q
= α
q
ρ
q
(23.53)
where ρ
q
is the physical density of phase q.
q
∂
∂t
23.5.3 ConservationEquations
The general conservation equations from which the equations solved by FLUENT are
derived are presented in this section, followed by the solved equations themsel ves.
Equationsin GeneralForm
Conservationof Mass
The continuity equation for phase q is
n
(α
q
ρ
q
) + ∇ · (α
q
ρ
q
~v
q
) =
X
(m˙
pq
− m˙
qp
) + S
q
(23.54)
p=1
where ~v
q
is the velocity of phase q and m˙
pq
characterizes the mass transfer from the
p
th
to q
th
phase, and m˙
qp
characterizes the mass transfer from phase q to phase p, and
you are able to specify these mechanisms separatel y.
By default, the source term S
q
on the righthand side of Equation 23.54 is zero, but you
can specify a constant or userdefined mass source for each phase. A similar term appears
in the momentum and enthalpy equations. See Section 23.7: Modeling Mass Transfer in
Multiphase Flows for more information on the modeling of mass transfer in FLUENT’s
general multiphase models.
Conservationof Momentum
The momentum balance for phase q yields
∂
∂t
(α
q
ρ
q
~v
q
) + ∇ · (α
q
ρ
q
~v
q
~v
q
) = −α
q
∇p + ∇ · τ
q
+ α
q
ρ
q
~g+
n
X
(R
~
pq
+ m˙
pq
~v
pq
− m˙
qp
~v
qp
) + (F
~
q
+ F
~
lift,q
+ F
~
vm,q
)
(23.55)
p=1
where τ
q
is the q
th
phase stressstrain tensor
τ
q
= α
q
µ
q
(∇~v
q
+ ∇~v
T
) + α
q
(λ
q
−
2
3
µ
q
)∇ · ~v
q
I (23.56)
Here µ
q
and λ
q
are the shear and bulk viscosity of phase q, F
~
q
is an external body force,
F
~
lift,q
is a lift
force,
F
~
vm,q
is a virtual mass
force,
R
~
pq
is an interaction force
between
phases, and p is the pressure shared by all phases.
~v
pq
is the interphase velocity, defined as follows. If m˙
pq
> 0 (i.e., phase p mass is
being transferred to phase q), ~v
pq
= ~v
p
; if m˙
pq
< 0 (i.e., phase q mass is being
transferred to phase p), ~v
pq
= ~v
q
. Likewise, if m˙
qp
> 0 then v
qp
= v
q
, if m˙
qp
< 0
then v
qp
= v
p
.
Equation 23.55 must be closed with appropriate expressions for the interphase force R
~
pq
. This force depends on the friction, pressure, cohesion, and other effects, and is
subject to the conditions that R
~
pq
= −R
~
qp
and R
~
qq
= 0.
FLUENTuses a simple interaction term of the following form:
n n
X
R
~
pq
=
X
K
pq
(~v
p
− ~v
q
) (23.57)
p=1 p=1
where K
pq
(= K
qp
) is the interphase momentum exchange coefficient (described in Sec
tion 23.5.4: Interphase Exchange Coefficients).
Lift Forces
For multiphase flows, FLUENTcan include the effect of lift forces on the secondary phase
particles (or droplets or bubbles). These lift forces act on a particle mainly due to velocity
gradients in the primaryphase flow field. The lift force will be more significant for larger
particles, but the FLUENT model assumes that the particle diameter is much smaller
than the interparticle spacing. Thus, the inclusion of lift forces is not appropriate for
closely packed particles or for very small particles.
The lift force acting on a secondary phase p in a primary phase q is computed from [88]
F
~
lift
= −0.5ρ
q
α
p
(~v
q
− ~v
p
) × (∇ × ~v
q
) (23.58)
The lift force F
~
lift
will be added to the righthand side of the momentum equation
for both phases (F
~
lift,q
= −F
~
lift ,p
).
In most cases, the lift force is insignificant compared to the drag force, so there is no
reason to include this extra term. If the lift force is significant (e.g., if the phases separate
quickly), it may be appropriate to include this term. By default, F
~
lift
is not
included. The lift force and lift coefficient can be specified for each pair of phases, if
desired.
i
It is important that if you include the lift force in your calculation,
you need not include it everywhere in the computational domain since
it is
computationally expensive to converge. For example, in the wall boundary
layer for turbule nt bubbly flows in channels, the lift force is significant
when the slip velocity is large in the vicinity of high strain rates for the
primary phase.
dt
Virtual Mass Force
For multiphase flows, FLUENT includes the “virtual mass effect” that occurs when a
secondary phase p accelerates relative to the primary phase q. The inertia of the primary
phase mass encountered by the accelerating particles (or droplets or bubbles) exerts a
“virtual mass force” on the particles [88]:
F
~
vm
= 0.5α
p
ρ
q
d
q
~v
q
dt
d
p
~v
p
!
−
dt
(23.59)
The term
d
q
denotes the phase material time derivative of the form
d
q
(φ)
=
∂(φ)
+
(~v
· ∇)φ (23.510)
dt ∂t
q
The virtual mass force F
~
vm
will be added to the righthand side of the momentum
equation for both phases (F
~
vm,q
= −F
~
vm,p
).
The virtual mass effect is significant when the secondary phase density is much smaller
than the primary phase density (e.g., for a transie nt bubble column). By default, F
~
vm
is not included.
Conservationof Energy
To describe the conservation of energy in Eulerian multiphase applications, a separate
enthalpy equation can be written for each phase:
∂
∂t
(α
q
ρ
q
h
q
) + ∇ · (α
q
ρ
q
~u
q
h
q
) =
−α
q
∂p
q
∂t
n
+ τ
q
: ∇~u
q
− ∇ ·~q
q
+ S
q
+
X
(Q
pq
+ m˙
pq
h
pq
−
m˙
qp
h
qp
)
p=1
(23.511)
where h
q
is the specific enthalpy of the q
th
phase, ~q
q
is the heat flux, S
q
is a source
term that includes sources of enthalpy (e.g., due to chemical reaction or radiation),
Q
pq
is the intensity of heat exchange between the p
th
and q
th
phases, and h
pq
is the
interphase enthalpy (e.g., the enthalpy of the vapor at the temperature of the droplets,
in the case of evaporation). The heat exchange between phases must comply with the
local balance conditions Q
pq
= −Q
qp
and Q
qq
= 0.
EquationsSolved by FLUENT
The equations for fluidfluid and granular multiphase flows, as solved by FLUENT, are
presented here for the general case of an nphase flow.
ContinuityEquation
The volume fraction of each phase is calculated from a continuity equation:
1
"
∂
n
ü
ρ
rq
,
∂t
(α
q
ρ
q
) + ∇ · (α
q
ρ
q
~v
q
) =
X
(m˙
pq
− m˙
qp
)
ü
(23.512)
p=1
where ρ
rq
is the phase reference density, or the volume averaged density of the q
th
phase
in the solution domain.
The solution of this equation for each secondary phase, along with the condition that the
volume fractions sum to one (given by Equation 23.52), allows for the calculation of the
primaryphase volume fraction. This treatme nt is common to fluidfluid and granular
flows.
FluidFluidMomentumEquations
The conservation of momentum for a fluid phase q is
∂
∂t
(α
q
ρ
q
~v
q
) + ∇ · (α
q
ρ
q
~v
q
~v
q
) = −α
q
∇p + ∇ · τ
q
+ α
q
ρ
q
~g +
n
X
(K
pq
(~v
p
− ~v
q
) + m˙
pq
~v
pq
− m˙
qp
~v
qp
) +
p=1
(F
~
q
+ F
~
lift,q
+ F
~
vm,q
) (23.513)
Here ~g is the acceleration due to gravity and τ
q
, F
~
q
, F
~
lift,q
, and F
~
vm,q
are as
defined for
Equation 23.55.
FluidSolidMomentumEquations
Following the work of [7, 51, 79, 119, 198, 221, 267, 363], FLUENT uses a multifluid
granular model to describe the flow behavior of a fluidsolid mixture. The solidphase
stresses are derived by making an analogy between the random particle motion arising
from particleparticle collisions and the thermal motion of molecules in a gas, taking into
account the inelasticity of the granular phase. As is the case for a gas, the intensity of the
particle velocity fluctuations determines the stresses, viscosity, and pressure of the solid
phase. The kinetic energy associated with the particle velocity fluctuations is represented
p
by a “pseudothermal” or granular temperature which is proportional to the mean square
of the random motion of particles.
The conservation of momentum for the fluid phases is similar to Equation 23.513, and
that for the s
th
solid phase is
∂
∂t
(α
s
ρ
s
~v
s
) + ∇ · (α
s
ρ
s
~v
s
~v
s
) = −α
s
∇p − ∇p
s
+ ∇ · τ
s
+ α
s
ρ
s
~g +
N
X
(K
ls
(~v
l
− ~v
s
) + m˙
ls
~v
ls
− m˙
sl
~v
sl
) +
l=1
(F
~
s
+ F
~
lift ,s
+ F
~
vm,s
) (23.514)
where p
s
is the s
th
solids pressure, K
ls
= K
sl
is the momentum exchange coefficient
between fluid or solid phase l and solid phase s, N is the total number of phases, and
F
~
q
, F
~
lift,q
, and F
~
vm,q
are as defined for Equation 23.55.
Conservationof Energy
The equation solved by FLUENTfor the conservation of energy is Equation 23.511.
23.5.4 InterphaseExchange Coefficients
It can be seen in Equations 23.513 and 23.514 that momentum exchange between the
phases is based on the value of the fluidfluid exchange coefficient K
pq
and, for granular
flows, the fluidsolid and solidsolid exchange coefficients K
ls
.
FluidFluidExchange Coefficient
For fluidfluid flows, each secondary phase is assumed to form droplets or bubbles. This
has an impact on how each of the fluids is assigned to a particular phase. For example,
in flows where there are unequal amounts of two fluids, the predomina nt fluid should be
modeled as the primary fluid, since the sparser fluid is more likely to form droplets or
bubbles. The exchange coefficient for these types of bubbly, liquidliquid or gasliquid
mixtures can be written in the following general form:
K
pq
=
α
q
α
p
ρ
p
f
τ
p
(23.515)
where f , the drag function, is defined differently for the different exchange
coefficient models (as described below) and τ
p
, the “particulate relaxation time”, is
defined as
τ
p
=
ρ
p
d
2
18µ
q
(23.516)
where d
p
is the diameter of the bubbles or droplets of phase p.
Nearly all definitions of f include a drag coefficient (C
D
) that is based on the relati ve
Reynolds number (Re). It is this drag function that differs among the exchangecoefficient
models. For all these situations, K
pq
should tend to zero whenever the primary phase is
not present within the domain. To enforce this, the drag function f is always multiplied
by the volume fraction of the primary phase q, as is reflected in Equation 23.515.
• For the model of Schiller and Naumann [320]
where
f =
C
D
Re
24
(23.517)
C
D
=
(
24(1 + 0.15 Re
0.687
)/Re Re ≤ 1000
0.44 Re > 1000
(23.518)
and Re is the relative Reynolds number. The relative Reynolds number for the
primary phase q and secondary phase p is obtained from
Re =
ρ
q
~v
p
− ~v
q
d
p
µ
q
(23.519)
The relative Reynolds number for secondary phases p and r is obtained from
Re =
ρ
rp
~v
r
− ~v
p

d
rp
µ
rp
(23.520)
where µ
rp
= α
p
µ
p
+ α
r
µ
r
is the mixture viscosity of the phases p and r.
The Schiller and Naumann model is the default method, and it is acceptable for
general use for all fluidfluid pairs of phases.
• For the Morsi and Alexander model [252]
where
f =
C
D
Re
24
(23.521)
a
2
a
3
C
D
= a
1
+
Re
+
Re
2
(23.522)
0
,
,
,
−
,
,
,
§
2
and Re is defined by Equation 23.519 or 23.520. The a’s are defined as follows:
a
1
, a
2
, a
3
=
)
0, 24, 0 0 < Re < 0.1
, 3.690, 22.73, 0.0903 0.1 < Re < 1
,
1.222, 29.1667, −3.8889 1 < Re < 10
,
§
0.6167, 46.50, −116.67 10 < Re < 100
0.3644, 98.33, 2778 100 < Re < 1000
,
,
0.357, 148.62, −47500 1000 < Re < 5000
,
0.46, −490.546, 578700 5000 < Re < 10000
.5191, −1662.5, 5416700 Re ≥
10000
(23.523)
The Morsi and Alexander model is the most complete, adjusting the function def
inition frequently over a large range of Reynolds numbers, but calculations with
this model may be less stable than with the other models.
• For the symmetric model
where
K
pq
=
α
p
(α
p
ρ
p
+ α
q
ρ
q
)f
τ
pq
(23.524)
and
(α
p
ρ
p
+ α
q
ρ
q
)(
d
p
+d
q
)
2
τ
pq
=
2
18(α
p
µ
p
+ α
q
µ
q
)
(23.525)
where
f =
C
D
Re
24
(23.526)
C
D
=
(
24(1 + 0.15 Re
0.687
)/Re Re ≤ 1000
0.44 Re > 1000
(23.527)
and Re is defined by Equation 23.519 or 23.520. Note that if there is only one
dispersed phase, then d
p
= d
q
in Equation 23.525.
The symmetric model is recommended for flows in which the secondary (dispersed)
phase in one region of the domain becomes the primary (continuous) phase in
another. Thus for a single dispersed phase, d
p
= d
q
and
(d
p
+d
q
)
= d
p
. For
example, if air is injected into the bottom of a container filled halfway with
water, the air is the dispersed phase in the bottom half of the container; in the
top half of the container, the air is the continuous phase. This model can also
be used for the interaction between secondary phases.
s
24v
2
You can specify different exchange coefficients for each pair of phases. It is also possible
to use userdefined functions to define exchange coefficients for each pair of phases. If the
exchange coefficient is equal to zero (i.e., if no exchange coefficient is specified), the flow
fields for the fluids will be computed independently, with the only “interaction” being
their complementary volume fractions within each computational cell.
FluidSolidExchangeCoefficient
The fluidsolid exchange coefficient K
sl
can be written in the following general form:
K
sl
=
α
s
ρ
s
f
τ
s
(23.528)
where f is defined differently for the different exchangecoefficient models (as described
below), and τ
s
, the “particulate relaxation time”, is defined as
ρ
s
d
2
τ
s
=
18µ
l
where d
s
is the diameter of particles of phase s.
(23.529)
All definitions of f include a drag function (C
D
) that is based on the relative Reynolds
number (Re
s
). It is this drag function that differs among the exchangecoefficient models.
• For the SyamlalO’Brien model [362]
f =
C
D
Re
s
α
l
r,s
(23.530)
where the drag function has a form derived by Dalla Valle [73]
"
4.8
ü
2
C
D
=
,
0.63 +
q
ü
Re
s
/v
r,s
(23.531)
This model is based on measurements of the terminal velocities of particles in
fluidized or settling beds, with correlations that are a function of the volume fraction
and relative Reynolds number [305]:
Re
s
=
ρ
l
d
s
~v
s
−
~v
l

µ
l
(23.532)
where the subscript l is for the l
th
fluid phase, s is for the s
th
solid phase, and d
s
is
the diameter of the s
th
solid phase particles.
r,s
A = α
4.14
l
B = α
2.65
α
D
The fluidsolid exchange coefficient has the form
K
sl
=
3α
s
α
l
ρ
l
C
4v
2
d
s
Re
s
!
v
r,s
~v
s
− ~v
l
 (23.533)
where v
r,s
is the terminal velocity correlation for the solid phase [113]:
v
r,s
= 0.5
A − 0.06 Re
s
+
q
2
(0.06 Re
s
)
+ 0.12 Re
s
(2B − A) +
A
2
(23.534)
with
l
(23.535)
and
for α
l
≤ 0.85,
and
B = 0.8α
1.28
(23.536)
for α
l
> 0.85.
l
(23.537)
This model is appropriate when the solids shear stresses are defined according to
Syamlal et al. [363] (Equation 23.564).
• For the model of Wen and Yu [396], the fluidsolid exchange coefficient is of
the following form:
where
3
K
sl
=
4
C
D
s
α
l
ρ
l
~v
s
−
~v
l

d
s
−2.65
l
(23.538)
24
h
1 + 0.15(α Re )
0.687
i
(23.539)
C
D
=
α Re
l s
l s
and Re
s
is defined by Equation 23.532.
This model is appropriate for dilute systems.
α
2
• The Gidaspow model [119] is a combination of the Wen and Yu model [396]
and the Ergun equation [96].
When α
l
> 0.8, the fluidsolid exchange coefficient K
sl
is of the following form:
where
3
K
sl
=
4
C
D
s
α
l
ρ
l
~v
s
−
~v
l

d
s
−2.65
l
(23.540)
24
h
1 + 0.15(α Re )
0.687
i
(23.541)
C
D
=
α Re
l s
l s
When α
l
≤ 0.8,
K
sl
= 150
α
s
(1 − α
l
)µ
l
α d
2
+ 1.75
ρ
l
α
s
~v
s
−
~v
l

d
(23.542)
l
s
s
This model is recommended for dense fluidized beds.
SolidSolidExchangeCoefficient
The solidsolid exchange coefficient K
ls
has the following form [361]:
3 (1 + e
ls
)
π
+ C
fr,ls
π
α ρ α ρ
(d
+ d )
2
g
2 8
s s l l l s 0,ls
K
ls
=
2π (ρ d
3
+ ρ d
3
)
~v
l
− ~v
s
 (23.543)
l l s s
where
e
ls
= the coefficient of restitution
C
fr,ls
= the coefficient of friction between the l
th
and s
th
solidphase particles (C
fr, ls
= 0)
d
l
= the diameter of the particles of solid l
g
0,ls
= the radial distribution coefficient
Note that the coefficient of restitution is described in Section 23.5.5: Solids Pressure
and the radial distribution coefficient is described in Section 23.5.5: Radial Distribution
Function.
s
s
qp
23.5.5 Solids Pressure
For granular flows in the compressible regime (i.e., where the solids volume fraction is less
than its maximum allowed value), a solids pressure is calculated independently and used
for the pressure gradient term, ∇p
s
, in the granularphase momentum equation. Because
a Maxwellian velocity distribution is used for the particles, a granular temperature is
introduced into the model, and appears in the expression for the solids pressure and
viscosities. The solids pressure is composed of a kinetic term and a second term due to
particle collisions:
p
s
= α
s
ρ
s
Θ
s
+ 2ρ
s
(1 + e
ss
)α
2
g
0,ss
Θ
s
(23.544)
where e
ss
is the coefficient of restitution for particle collisions, g
0,ss
is the radial distribu
tion function, and Θ
s
is the granular temperature. FLUENT uses a default value of 0.9
for e
ss
, but the value can be adjusted to suit the particle type. The granular temperature
Θ
s
is proportional to the kinetic energy of the fluctuating particle motion, and will be
described later in this section. The function g
0,ss
(described below in more detail) is a
distribution function that governs the transition from the “compressible” condition with
α < α
s, max
, where the spacing between the solid particles can continue to decrease, to
the “incompressible” condition with α = α
s, max
, where no further decrease in the spacing
can occur. A value of 0.63 is the default for α
s, max
, but you can modify it during the
problem setup.
Other formulations that are also available in FLUENTare [363]
p
s
= 2ρ
s
(1 + e
ss
)α
2
g
0,ss
Θ
s
(23.545)
and [226]
1
p
s
= α
s
ρ
s
Θ
s
[(1 + 4α
s
g
0,ss
) +
2
[(1 + e
ss
)(1 − e
ss
+ 2µ
f ric
)]] (23.546)
When more than one solids phase are calculated, the above expression does not take into
account the effect of other phases. A derivation of the expressions from the Boltzman
equations for a granular mixture are beyond the scope of this manual, however there is
a need to provide a better formulation so that some properties may feel the presence of
other phases. A known problem is that N solids phases with identical properties should be
consistent when the same phases are described by a single solids phase. Equations derived
empirically may not satisfy this property and need to be changed accordingly without
deviating significantly from the original form. From [118], a general solids pressure
formulation in the presence of other phases could be of the form
N
π
p
q
= α
q
ρ
q
Θ
q
+
X
g
0,pq
d
3
n
q
n
p
(1 + e
qp
)f (m
p
, m
q
, Θ
p
, Θ
q
) (23.547)
p=1
3
2
d
3
d
3
where d
pq
=
d
p
+d
q
is the average diameter, n
p
, n
q
are the number of particles, m
p
and m
q
are the masses of the particles in phases p and q, and f is a function of the masses of the
particles and their granular temperatures. For now, we have to simplify this expression
so that it depends only on the granular temperature of phase q
N
p
q
= α
q
ρ
q
Θ
q
+
X
2
p=1
d
3
pq
(1 + e
pq
)g
0,pq
α
q
α
p
ρ
q
Θ
q
(23.548)
q
Since all models need to be cast in the general form, it follows that
N
p
q
= α
q
ρ
q
Θ
q
+ (
X
p=1
d
3
pq
p
c,qp
)ρ
q
Θ
q
(23.549)
q
where p
c,qp
is the collisional part of the pressure between phases q and p.
The above expression reverts to the one solids phase expression when N = 1 and q = p
but also has the property of feeling the presence of other phases.
RadialDistributionFunction
The radial distribution function, g
0
, is a correction factor that modifies the probabili ty
of collisions between grains when the solid granular phase becomes dense. This function
may also be interpreted as the nondimensional distance between spheres:
g
0
=
s + d
p
(23.550)
s
where s is the distance between grains. From Equation 23.550 it can be observed that
for a dilute solid phase s → ∞, and therefore g
0
→ 1. In the limit when the solid phase
compacts, s → 0 and g
0
→ ∞. The radial distribution function is closely connected
to the factor χ of Chapman and Cowling’s [51] theory of nonuniform gases. χ is equal
to 1 for a rare gas, and increases and tends to infinity when the molecules are so close
together that motion is not possible.
1
1
α
s,max
)
1 −
1
s
2
k
d
N
In the literature there is no unique formulation for the radial distribution function. FLU
ENT has a number of options:
• For one solids phase, use [267]:
¿
α
!
−
3
Q
s
g
0
=

1 −
α
s,max
\
(23.551)
This is an empirical function and does not extends easily to n phases. For two
identical phases with the property that α
q
= α
1
+ α
2
, the above function is not
consistent for the calculation of the partial pressures p
1
and p
2
, p
q
= p
1
+ p
2
. In
order to correct this problem, FLUENTuses the following consistent formulation:
¿
α
!
−1
1
N
α
g
0,ll
=

1 −
α
3
Q
\
+ d
l
X
(23.552)
s, max
k=1
k
where
and k are solids phases only.
N
α
s
=
X
α
k
(23.553)
k=1
• The following expression is also available [151]:
1
g
0,ll
=
(1 −
α
s
3
+
2
d
l
X
α
k
k=1
d
k
(23.554)
• Also available [226], slightly modified for n solids phases, is the following:
1 + 2.5α
s
+ 4.59α
2
+
4.52α
3
1
N
α
k
g
0,ll
=
s s
l
X
α s
αs, max
3
0.678
+
2
d
k=1
d
k
(23.555)
• The following equation [363] is
available:
1
3(
P
N α
k
)
k=1
d
k
g
0,kl
=
(1 − α )
+
(1 − α )
2
(d
d
k
d
l
(23.556)
+ d )
s s j k
X
1
=
(α
d1
α
d1
When the number of solid phases is greater than 1, Equation 23.552, Equation 23.554
and Equation 23.555 are extended to
d
m
g
0,ll
+ d
l
g
0,mm
g
0,lm
=
d
m
+ d
l
(23.557)
It is interesting to note that equations Equation 23.554 and Equation 23.555 compare
well with [6] experimental data, while Equation 23.556 reverts to the [47] derivation.
23.5.6 Maxi mum Packing Limitin BinaryMixtures
The packing limit is not a fixed quantity and may change according to the number of
particles present within a given volume and the diameter of the particles. Small particles
accumulate in between larger particles increasing the packing limit. For a binary mixture
FLUENTuses the correlations proposed by [99].
For a binary mixture with diameters d
1
> d
2
, the mixture composition is defined as
α1
α1 +α2
where
X
1
<=
1,max
α
1,max
+ (1 − α
1,max
)α
2,max
(23.558)
)
The maximum packing limit for the mixture is given by
α
s,max
= (α
1,max
− α
2,max
+ [1
−
s
d2
](1 − α
1,max
)α
2,max
) (23.559)
X
1
∗(α
1,max
+ (1 − α
1,max
)α
2,max
)
1,max
+α
2,max
(23.560)
otherwise, the maximum packing limit for the binary mixture is
[1 −
s
d2
](α
1,max
+ (1 − α
1,max
)α
2,max
)(1 − X
1
) + α
1,max
(23.561)
The packing limit is used for the calculation of the radial distribution function.
5
s
23.5.7 Solids Shear Stresses
The solids stress tensor contains shear and bulk viscosities arising from particle momen
tum exchange due to translation and collision. A frictional component of viscosity can
also be included to account for the viscousplastic transition that occurs when particles
of a solid phase reach the maximum solid volume fraction.
The collisional and kinetic parts, and the optional frictional part, are added to give the
solids shear viscosity:
µ
s
= µ
s,col
+ µ
s,kin
+ µ
s,fr
(23.562)
Collisional Viscosity
The collisional part of the shear viscosity is modeled as [119, 363]
KineticViscosity
4
µ
s,col
=
5
α
s
ρ
s
d
s
g
0,ss
(1 + e
ss
)
Θ
s
1/2
π
(23.563)
FLUENTprovides two expressions for the kinetic part.
The default expression is from Syamlal et al. [363]:
α
s
d
s
ρ
s
√
Θ
s
π
2
µ
s,kin
=
6 (3 − e
ss
)
1 + (1 + e
ss
) (3e
ss
− 1) α
s
g
0,ss
(23.564)
The following optional expression from Gidaspow et al. [119] is also available:
10ρ
s
d
s
√
Θ
s
π 4
2
µ
s,kin
=
96α (1 + e
ss
) g
0,ss
1 + g
0,ss
α
s
(1 + e
ss
)
5
(23.565)
BulkViscosity
The solids bulk viscosity accounts for the resistance of the granular particles to compres
sion and expansion. It has the following form from Lun et al. [221]:
4
λ
s
=
3
α
s
ρ
s
d
s
g
0,ss
(1 + e
ss
)
Θ
s
1/2
π
(23.566)
Note that the bulk viscosity is set to a constant value of zero, by default. It is also
possible to select the Lun et al. expression or use a userdefined function.
τ
f riction
= −P
f r
~
~
f riction s s
FrictionalViscosity
In dense flow at low shear, where the secondary volume fraction for a solid phase nears
the packing limit, the generation of stress is mainly due to friction between particles.
The solids shear viscosity computed by FLUENT does not, by default, account for the
friction between particles.
If the frictional viscosity is included in the calculation, FLUENT uses Schaeffer’s [318]
expression:
µ
s,fr
=
p
s
sin φ
2
√
I
2D
(23.567)
where p
s
is the solids pressure, φ is the angle of internal friction, and I
2D
is the second
invariant of the deviatoric stress tensor. It is also possible to specify a constant or user
defined frictional viscosity.
In granular flows with high solids volume fraction, instantaneous collisions are less im
portant. The application of kinetic theory to granular flows is no longer relevant since
particles are in contact and the resulting frictional stresses need to be taken into account.
FLUENTextends the formulation of the frictional viscosity and employs other models, as
well as providing new hooks for UDFs. See the separate UDF Manual for details.
The frictional stresses are usually written in Newtonian form:
icti on
I + µ (∇~u + (∇~u )
T
) (23.568)
The frictional stress is added to the stress predicted by the kinetic theory when the solids
volume fraction exceeds a critical value. This value is normally set to 0.5 when the flow
is threedimensional and the maximum packing limit is about 0.63. Then
P
S
= P
kinetic
+ P
f riction
(23.569)
µ
S
= µ
kinetic
+ µ
f riction
(23.570)
The derivation of the frictional pressure is mainly semiempirical, while the frictional vis
cosity can be derived from the first principles. The application of the modified Coulomb
law leads to an expression of the form
µ
f riction
=
P
f riction
sin φ
2
√
I
2D
(23.571)
Where φ is the angle of internal friction and I
2D
is the second invariant of the deviatoric
stress tensor.
(α
Two additional models are available in FLUENT: the Johnson and Jackson [165] model
for frictional pressure and Syamlal et al [363].
The Johnson and Jackson [165] model for frictional pressure is defined as
(α
s
− α
s,mi n
)
n
P
f riction
= F r
s,max
(23.572)
− α
s
)
p
With coefficient Fr = 0.05, n=2 and p = 3 [266]. The critical value for the solids volume
fraction is 0.5. The coefficient Fr was modified to make it a function of the volume
fraction:
F r = 0.1α
s
(23.573)
The frictional viscosity for this model is of the form
µ
f riction
= P
f riction
sin φ (23.574)
The second model that is employed is Syamlal et al [363], described in Equation 23.564.
Comparing the two models results in the frictional normal stress differing by orders of
magnitude.
The radial distribution function is an important parameter in the description of the solids
pressure resulting from granular kinetic theory. If we use the models of Lun et al. [221] or
Gidaspow [118] the radial function tends to infinity as the volume fraction tends to the
packing limit. It would then be possible to use this pressure directly in the calculation
of the frictional viscosity, as it has the desired effect. This approach is also available in
FLUENTby default.
i
The introduction of the frictional viscosity helps in the description of
fric tional flows, however a complete description would require the
introduction
of more physics to capture the elastic regime with the calculation of the
yield stress and the use of the flowrule. These effects can be added by the
user via UDFs to model static regime. Small time steps are required to get
good convergence behavior.
ss
ss
23.5.8 GranularTemperature
The granular temperature for the s
th
solids phase is proportional to the kinetic energy of
the random motion of the particles. The trans port equation derived from kinetic theory
takes the form [79]
3
"
∂
#
(ρ
s
α
s
Θ
s
) + ∇ · (ρ
s
α
s
~v
s
Θ
s
)
= (−p
s
I +τ
s
) : ∇~v
s
+∇·(k
Θ
s
∇Θ
s
)−γ
Θ
s
+φ
ls
(23.5
75)
2 ∂t
where
(−p
s
I + τ
s
) : ∇~v
s
= the generation of energy by the solid stress tensor
k
Θ
s
∇Θ
s
= the diffusion of energy (k
Θs
is the diffusion coefficient)
γ
Θ
s
= the collisional dissipation of energy
φ
ls
= the energy exchange between the l
th
fluid or solid phase and the s
th
solid phase
Equation 23.575 contains the term k
Θ
s
∇Θ
s
describing the diffusive flux of granular
energy. When the default Syamlal et al. model [363] is used, the diffusion coefficient for
granular energy, k
Θ
s
is given by
15d
s
ρ
s
α
s
√
Θ
s
π
2
2
16
k
Θ
s
=
4(41 − 33η)
1 + η
5
(4η − 3)α
s
g
0,ss
+
15π
(41 − 33η)ηα
s
g
0,ss
)
(23.576)
where
1
η =
2
(1 + e
ss
)
FLUENT uses the following expression if the optional model of Gidaspow et al. [119] is
enabled:
150ρ
s
d
s
q
(Θπ)
6
2
2
s
Θ
s
k
Θ
s
=
384(1 + e )g
0,ss
1 + α
s
g
0,ss
(1 + e
s
)
5
+ 2ρ
s
α
s
d
s
(1 + e
ss
)g
0,ss
(23.577)
π
The collisional dissipation of energy, γ
Θ
s
, represents the rate of energy dissipation within
the s
th
solids phase due to collisions between particles. This term is represented by the
expression derived by Lun et al. [221]
γ
Θm
=
12(1 − e
2
)g
0,ss
ρ
s
α
2
Θ
3/2
(23.578)
d
s
√
π
s s
s
6
~
2
−
The transfer of the kinetic energy of random fluctuations in particle velocity from the s
th
solids phase to the l
th
fluid or solid phase is represented by φ
ls
[119]:
φ
ls
= −3K
ls
Θ
s
(23.579)
FLUENTallows the user to solve for the granular temperature with the following options:
• algebraic formulation (the default)
It is obtained by neglecting convection and diffusion in the transport equation,
Equation 23.575 [363].
• partial differential equation
This is given by Equation 23.575 and it is allowed to choose different options for
it properties.
• constant granular temperature
This is useful in very dense situations where the random fluctuations are small.
• UDF for granular temperature
For a granular phase s, we may write the shear force at the wall in the following form:
τ~ =
π
√
3φ
α
s
α
ρ
s
g
0
q
Θ
s
U
s,
(23.580)
s,max
Here U
~
s,
is the particle slip velocity parallel to the wall, φ is the specularity
coefficient between the particle and the wall, α
s,max
is the volume fraction for the
particles at maximum packing, and g
0
is the radial distribution function that is model
dependent.
The general boundary condition for granular temperature at the wall takes the form
[165]
π
√
α
s
q
π
√
α
s 2
3
q
s
=
6
3φ
α
s,max
ρ
s
g
0
Θ
s
)U
~
s,
· U
~
s,
−
4
3
α
s,max
(1 − e
sw
)ρ
s
g
0
Θ
s
(23.581)
23.5.9 Description of HeatTransfer
The internal energy balance for phase q is written in terms of the phase enthalpy, Equa
tion 23.511, defined by
Z
H
q
= c
p,q
dT
q
(23.582)
where c
p,q
is the specific heat at constant pressure of phase q. The thermal boundary
conditions used with multiphase flows are the same as those for a singlephase flow. See
Chapter 7: Boundary Conditions for details.
p
The HeatExchangeCoefficient
The rate of energy transfer between phases is assumed to be a function of the temperature
difference
Q
pq
= h
pq
(T
p
− T
q
) (23.583)
where h
pq
(= h
qp
) is the heat transfer coefficient between the p
th
phase and the q
th
phase.
The heat transfer coefficient is related to the p
th
phase Nusselt number, Nu
p
, by
h
pq
=
6κ
q
α
p
α
q
Nu
p
d
p
2
(23.584)
Here κ
q
is the thermal conductivity of the q
th
phase. The Nusselt number is typically
determined from one of the many correlations reported in the literature. In the case of
fluidfluid multiphase, FLUENTuses the correlation of Ranz and Marshall [295, 296]:
Nu
p
= 2.0 + 0.6Re
1/2
Pr
1/3
(23.585)
where Re
p
is the relative Reynolds number based on the diameter of the p
th
phase and
the relative velocity u~
p
− u~
q
, and Pr is the Prandtl number of the q
th
phase:
c
p
q
µ
q
Pr =
κ
q
(23.586)
In the case of granular flows (where p = s), FLUENT uses a Nusselt number correlation
by Gunn [128], applicable to a porosity range of 0.35–1.0 and a Reynolds number of up
to 10
5
:
Nu
s
= (7 − 10α
f
+ 5α
2
)(1 + 0.7Re
0.2
Pr
1/3
) + (1.33 − 2.4α
f
+ 1.2α
2
)Re
0.7
Pr
1/3
(23.5
87)
f s f s
The Prandtl number is defined as above with q = f . For all these situations, h
pq
should tend to zero whenever one of the phases is not present within the domain. To
enforce this, h
pq
is always multiplied by the volume fraction of the primary phase q, as
reflected in Equation 23.584.
23.5.10 TurbulenceModels
To describe the effects of turbule nt fluctuations of velocities and scalar quantities in
a single phase, FLUENT uses various types of closure models, as described in Chap
ter 12: Modeling Turbulence. In comparison to singlephase flows, the number of terms
to be modeled in the momentum equations in multiphase flows is large, and this makes
the modeling of turbulence in multiphase simulations extremely complex.
FLUENTprovides three methods for modeling turbulence in multiphase flows within the
context of the k models. In addition, FLUENT provides two turbulence options within
the context of the Reynolds stress models (RSM).
The k turbulence model options are:
• mixture turbulence model (the default)
• dispersed turbulence model
• turbulence model for each phase
i
Note that the descriptions of each method below are presented based
on the standard k model. The multiphase modifications to the RNG
and
realizable k models are similar, and are therefore not presented explicitl y.
The RSM turbulence model options are:
• mixture turbulence model (the default)
• dispersed turbulence model
For either category, the choice of model depends on the importance of the secondary
phase turbulence in your application.
k TurbulenceModels
FLUENT provides three turbulence model options in the context of the k models: the
mixture turbulence model (the default), the dispersed turbulence model, or a perphase
turbulence model.
N
k MixtureTurbulenceModel
The mixture turbulence model is the default multiphase turbulence model. It represents
the first extension of the singlephase k model, and it is applicable when phases separate,
for stratified (or nearly stratified) multiphase flows, and when the density ratio between
phases is close to 1. In these cases, using mixture properties and mixture velocities is
sufficient to capture important features of the turbule nt flow.
The k and equations describing this model are as follows:
∂
∂t
(ρ
m
k) + ∇ · (ρ
m
~v
m
k) = ∇
·
µ
t,m
k
σk
∇
+ G
k,m
− ρ
m
(23.588)
and
∂
∂t
(ρ
m
) + ∇ · (ρ
m
~v
m
) = ∇
·
µ
t,m
σ
∇
+
k
(C
1
G
k,m
− C
2
ρ
m
) (23.589)
where the mixture density and velocity, ρ
m
and ~v
m
, are computed from
N
ρ
m
=
X
α
i
ρ
i
(23.590)
i=1
and
N
X
α
i
ρ
i
~v
i
~v
m
=
i=1
X
α
i
ρ
i
i=1
(23.591)
the turbule nt viscosity, µ
t,m
, is computed from
k
2
µ
t,m
= ρ
m
C
µ
(23.592)
and the production of turbulence kinetic energy, G
k,m
, is computed from
G
k,m
= µ
t,m
(∇~v
m
+
(∇~v
m
)
T
) : ∇~v
m
(23.593)
The constants in these equations are the same as those described in Section 12.4.1: Stan
dard k Model for the singlephase k model.
k Dispersed TurbulenceModel
The dispersed turbulence model is the appropriate model when the concentrations of the
secondary phases are dilute. In this case, interparticle collisions are negligible and the
dominant process in the random motion of the secondary phases is the influence of the
primaryphase turbulence. Fluctuating quantities of the secondary phases can therefore
be given in terms of the mean characteristics of the primary phase and the ratio of the
particle relaxation time and eddyparticle interaction time.
The model is applicable when there is clearly one primary continuous phase and the rest
are dispersed dilute secondary phases.
Assumptions
The dispersed method for modeling turbulence in FLUENTassumes the following:
• a modified k model for the continuous phase
Turbulent predictions for the continuous phase are obtained using the standard
k model suppleme nted with extra terms that include the interphase turbule nt
momentum transfer.
• Tchentheory correlations for the dispersed phases
Predictions for turbulence quantities for the dispersed phases are obtained using the
Tchen theory of dispersion of discrete particles by homogeneous turbulence [141].
• interphase turbule nt momentum transfer
In turbule nt multiphase flows, the momentum exchange terms contain the cor
relation between the instantaneous distribution of the dispersed phases and the
turbule nt fluid motion. It is possible to take into account the dispersion of the
dispersed phases trans ported by the turbule nt fluid motion.
• a phaseweighted averaging process
The choice of averaging process has an impact on the modeling of dispersion in
turbule nt multiphase flows. A twostep averaging process leads to the appearance
of fluctuations in the phase volume fractions. When the twostep averaging pro
cess is used with a phaseweighted average for the turbulence, however, turbule nt
fluctuations in the volume fractions do not appear. FLUENT uses phaseweighted
averaging, so no volume fraction fluctuations are introduced into the continuity
equations.
k
k
q
3
∂
∂
k
Turbulence in the Continuous Phase
The eddy viscosity model is used to calculate averaged fluctuating quantities. The
Reynolds stress tensor for continuous phase q takes the following form:
τ
0 0
2 T
q
= −
3
(ρ
q
k
q
+ ρ
q
µ
t,q
∇ · U
~
q
)I + ρ
q
µ
t,q
(∇U
~
q
+ ∇U
~
q
where U
~
q
is the phaseweighted velocity.
) (23.594)
The turbule nt viscosity µ
t,q
is written in terms of the turbule nt kinetic energy of phase
q:
2
µ
t,q
= ρ
q
C
µ
q
q
(23.595)
and a characteristic time of the energetic turbule nt eddies is defined as
3 k
q
τ
t,q
=
2
C
µ
(23.596)
where
q
is the dissipation rate and C
µ
= 0.09.
The length scale of the turbule nt eddies is
L
t,q
=
s
3/2
C
µ
q
2
q
(23.597)
Turbulent predictions are obtained from the modified k model:
∂t
(α
q
ρ
q
k
q
) + ∇ · (α
q
ρ
q
U
~
q
k
q
) = ∇ ·
(α
q
µ
t,q
σ
k
∇k
q
) + α
q
G
k,q
− α
q
ρ
q
q
+ α
q
ρ
q
Π
k
q
(23.598)
and
∂t
(α
q
ρ
q
q
) + ∇ · (α
q
ρ
q
U
~
q
q
) = ∇ ·
(α
q
µ
t,q
σ
∇
q
) + α
q
q
(C
1
G
k,q
− C
2
ρ
q
q
) + α
q
ρ
q
Π
q
q
(23.599)
Here Π
k
q
and Π
q
represent the influence of the dispersed phases on the continuous phase
q, and G
k,q
is the production of turbule nt kinetic energy, as defined in Section 12.4.4:
Mod eling Turbulent Production in the k Models. All other terms have the same
meaning as in the singlephase k model.
k
Π = C
X
q p
X
The term Π
k
q
can be derived from the instantaneous equation of the continuous phase
and takes the following form, where M represents the number of secondary phases:
M
Π
kq
=
K
pq
(< ~v
0 0
· ~v
0 0
> +( U
~
p
− U
~
q
) · ~v
dr
) (23.5100)
p=1
α
q
ρ
q
which can be simplified to
M
Π
k
q
=
K
pq
(k
pq
− 2k
q
+ ~v
pq
· ~v
dr
) (23.5101)
p=1
α
q
ρ
q
where k
lq
is the covariance of the velocities of the continuous phase q and the dispersed
phase l (calculated from Equation 23.5109 below), ~v
pq
is the relative velocity, and ~v
dr
is the drift velocity (defined by Equation 23.5114 below).
Π
q
is modeled according to Elgobashi et al. [95]:
where C
3
= 1.2.
q
q 3
q
Π
k
q
(23.5102)
Turbulence in the Dis persed Phase
Time and length scales that characterize the motion are used to evaluate dispersion
coefficients, correlation functions, and the turbule nt kinetic energy of each dispersed
phase.
The characteristic particle relaxation time connected with inertial effects acting on a
dispersed phase p is defined as
ρ
p
!
τ
F,pq
= α
p
ρ
q
K
−1
+ C
V
(23.5103)
pq
ρ
q
The Lagrangian integral time scale calculated along particle trajectories, mainly affected
by the crossingtr ajectory effect [69], is defined as
τ
t,pq
= q
τ
t,q
(1 + C
β
ξ
2
)
(23.5104)
where
ξ =
~v
pq

τ
t,q
L
t,q
(23.5105)
τ
η =
3 3
σ α
ρ
and
C
β
= 1.8 − 1.35 cos
2
θ (23.5106)
where θ is the angle between the mean particle velocity and the mean relative velocity.
The ratio between these two characteristic times is written as
τ
t,pq
pq
F,pq
(23.5107)
Following Simonin [333], FLUENTwrites the turbulence quantities for dispersed phase p
as follows:
k
p
= k
q
b
2
+ η
pq
!
1 + η
pq
b + η
pq
!
(23.5108)
k
pq
= 2k
q
1
1 + η
pq
(23.5109)
D
t,pq
= k
pq
τ
t,pq
(23.5110)
3
2 1
D
p
= D
t,pq
+ k
p
− b k
pq
τ
F,pq
(23.5111)
1
ρ
p
!
−
b = (1 + C
V
) + C
V
q
(23.5112)
and C
V
= 0.5 is the addedmass coefficient.
Interphase Turbule nt Mome ntum Transfer
The turbule nt drag term for multiphase flows (K
pq
(~v
p
−~v
q
) in Equation 23.57) is
modeled as follows, for dispersed phase p and continuous phase q:
K
pq
(~v
p
− ~v
q
) = K
pq
(U
~
p
− U
~
q
) − K
pq
~v
dr
(23.5113)
The second term on the righthand side of Equation 23.5113 contains the drift velocity:
D
p
D
q
!
~v
dr
= −
σ
pq
α
p
∇α
p
−
pq q
∇α
q
(23.5114)
Here D
p
and D
q
are diffusivities, and σ
pq
is a dispersion Prandtl number. When using
Tchen theory in multiphase flows, FLUENT assumes D
p
= D
q
= D
t,pq
and the default
value for σ
pq
is 0.75.
multiphase
options
yes
yes
∂
The drift velocity results from turbule nt fluctuations in the volume fraction. When
multiplied by the exchange coefficient K
pq
, it serves as a correction to the momentum
exchange term for turbule nt flows. This correction is not included, by default, but you
can enable it during the problem setup.
You can enable the effect of drift velocity by performing the
following:
1. If it is not already done, set the kepsilon Multiphase Model to Dispersed in the
Viscous panel.
2. Enter the multiphaseoptions text command in the console window.
define −→ models −→ viscous −→ multiphaseturbulence −→
multiphaseoptions
/define/models/viscous/multiphaseturbulence>
Enable dispersion force in momentum?
[no] Enable interphase kepsilon source?
[no]
The effect of the drift velocity is influenced both by the momentum equation and,
to a lesser extent, the turbulence equation. Therefore, you should answer yes to
both questions to take into account the effect of drift velocity.
k TurbulenceModelfor EachPhase
The most general multiphase turbulence model solves a set of k and trans port equations
for each phase. This turbulence model is the appropriate choice when the turbulence
transfer among the phases plays a dominant role.
Note that, since FLUENT is solving two additional trans port equations for each sec
ondary phase, the perphase turbulence model is more computationally intensive than
the dispersed turbulence model.
Trans port Equations
The Reynolds stress tensor and turbule nt viscosity are computed using Equations 23.594
and 23.595. Turbulence predictions are obtained from
∂t
(α
q
ρ
q
k
q
) + ∇ · (α
q
ρ
q
U
~
q
k
q
) = ∇ ·
(α
q
µ
t,q
σ
k
∇k
q
) + (α
q
G
k,q
− α
q
ρ
q
q
) +
N N
µ
t,l
N
µ
t,q
X
K
lq
(C
lq
k
l
− C
ql
k
q
) −
X
K
lq
(U
~
l
− U
~
q
) ·
∇α
l
+
X
K
lq
(U
~
l
−
U
~
q
) ·
∇α
q
(23.5115)
l=1 l=1
α
l
σ
l
l=1
α
q
σ
q
∂
and
∂t
(α
q
ρ
q
q
) + ∇ · (α
q
ρ
q
U
~
q
q
) = ∇ ·
(α
q
µ
t,q
σ
∇
q
) +
q
"
k
q
C
1
α
q
G
k,q
− C
2
α
q
ρ
q
q
+
N N
µ
t,l
N
µ
t,q
!#
C
3
X
K
lq
(C
lq
k
l
− C
ql
k
q
) −
X
K
lq
(U
~
l
−
U
~
q
) ·
∇α
l
+
X
K
lq
(U
~
l
− U
~
q
) ·
∇α
q
l=1 l=1
α
l
σ
l
l=1
α
q
σ
q
(23.5116)
The terms C
lq
and C
ql
can be approximated as
C
lq
= 2, C
ql
= 2
η
lq
!
1 + η
lq
(23.5117)
where η
lq
is defined by Equation 23.5107.
Interphase Turbule nt Mome ntum Transfer
The turbule nt drag term (K
pq
(~v
p
− ~v
q
) in Equation 23.57) is modeled as follows,
where
l is the dispersed phase (replacing p in Equation 23.57) and q is the continuous phase:
N N N
X
K
lq
(~v
l
− ~v
q
) =
X
K
lq
(U
~
l
− U
~
q
) −
X
K
lq
~v
dr,lq
(23.5118)
l=1 l=1 l=1
Here
U
~
l
and U
~
q
are phaseweighted velocities, and ~v
dr,lq
is the drift velocity for
phase
l (computed using Equation 23.5114, substituting l for p). Note that FLUENT will
compute the diffusivities D
l
and D
q
directly from the trans port equations, rather than
using Tchen theory (as it does for the dispersed turbulence model).
As noted above, the drift velocity results from turbule nt fluctuations in the volume
fraction. When multiplied by the exchange coefficient K
lq
, it serves as a correction to
the momentum exchange term for turbule nt flows. This correction is not included, by
default, but you can enable it during the problem setup.
The turbulence model for each phase in FLUENTaccounts for the effect of the turbulence
field of one phase on the other(s). If you want to modify or enhance the interaction of the
multiple turbulence fields and interphase turbule nt momentum transfer, you can supply
these terms using userdefined functions.
˜
∂
c
=
RSMTurbulence Models
Multiphase turbulence modeling typically involves two equation models that are
based on singlephase models and often cannot accurately capture the underlying flow
physics. Additional turbulence modeling for multiphase flows is diminished even more
when the basic underlying singlephase model cannot capture the complex physics of the
flow. In such situations, the logical next step is to combine the Reynolds stress model
with the multiphase algorithm in order to handle challenging situations in which both
factors, RSM for turbulence and the Eulerian multiphase formulation, are a
precondition for accurate predictions [65].
The phaseaveraged continuity and momentum equations for a continuous phase are:
∂
(α
c
ρ
c
) + ∇ · (α
c
ρ
c
U
c
) = 0 (23.5119)
∂t
(α
c
ρ
rmc
U
˜
c
) + ∇ · (α
c
ρ
rmc
U
˜
c
O
U
˜
c
) = −α
c
∇p˜ + ∇ · τ˜
t
+ F
Dc
(23.5120)
∂t
c
For simplicity, the laminar stressstrain tensor and other body forces such as gravity
have been omitted from Equations 23.511923.5120. The tilde denotes phaseaveraged
variables while an overbar (e.g., α
c
) reflects timeaveraged values. In general, any variable
Φ can have a phaseaverage value defined as
Φ
˜
α
c
Φ
c
α
c
(23.5121)
Considering only two phases for simplicity, the drag force between the continuous
and the dispersed phases can be defined as:
"
α
d
u
0
c
u
0
!#
F
Dc
= K
dc
(U
˜
d
− U
˜
c
)
−
d c
− (23.5122)
α
d
α
c
where K
dc
is the drag coefficient. Several terms in the Equation 23.5122 need to be
modeled in order to close the phaseaveraged momentum equations. Full descriptions of
all modeling assumptions can be found in [64]. This section only describes the different
modeling definition of the turbule nt stresses τ
˜t
that appears in Equation 23.5120.
The turbule nt stress that appears in the momentum equations need to be defined on a
perphase basis and can be calculated as:
τ
˜t
k
= −α
k
ρ
k
R
˜
k,ij
(23.5123)
˜
where the subscript k is replaced by c for the primary (i.e., continuous) phase or by d for
any secondary (i.e., dispersed) phases. As is the case for singlephase flows, the current
multiphase Reynolds stress model (RSM) also solves the trans port equations for Reynolds
stresses R
ij
. FLUENTincludes two methods for modeling turbulence in multiphase flows
within the context of the RSM model: the dispersed turbulence model, and the mixture
turbulence model.
RSMDispersed TurbulenceModel
The dispersed turbulence model is used when the concentrations of the secondary phase
are dilute and the primary phase turbulence is regarded as the dominant process. Conse
quently, the trans port equations for turbulence quantities are only solved for the primary
(continuous) phase, while the predictions of turbulence quantities for dispersed phases
are obtained using the Tchen theory. The trans port equation for the primary phase
Reynolds stresses in the case of the dispersed model are:
∂
(αρR
ij
) +
∂t
∂
∂x
k
(αρU
˜
k
R
˜
ij
) =
−αρ
R
˜
i
k
∂U
˜
j
∂x
k
+
R
˜
jk
∂U
˜
i
!
+
∂x
k
∂
∂x
k
"
∂
αµ
∂x
k
#
(R
˜
ij
)
∂ ∂u
0
∂u
0
[αρu
0
u
0
u
0
] + αp(
i
+
j
)
−
∂x
k
i j k
∂x
j
∂x
i
− αρ ˜
ij
+ Π
R,ij
(23.5124)
The variables in Equation 23.5124 are per continuous phase c and the subscript is omitted
for clarity. The last term of Equation 23.5124, Π
R,ij
, takes into account the interaction
between the continuous and the dispersed phase turbulence. A general model for this
term can be of the form:
Π
R,ij
= K
dc
C
1,dc
(R
dc,ij
− R
c,ij
) + K
dc
C
2,dc
a
dc,i
b
dc,j
(23.5125)
where C
1
and C
2
are unknown coefficients, a
dc,i
is the relative velocity, b
dc,j
represents the drift or the relative velocity, and R
dc,ij
is the unknown particulate
fluid velocity correlation. To simplify this unknown term, the following assumption has
been made:
2
Π
R,ij
=
3
δ
ij
Π
k
(23.5126)
where δ
ij
is the Kronecker delta, and Π
k
represents the modified version of the original
Simonin model [333].
Π
kc
= K
dc
(k
˜
dc
− 2k
˜
c
+ V
˜
rel
· V
˜
drift
) (23.5127)
1
where
K
˜
c
represents the turbule nt kinetic energy of the continuous
phase,
k
˜
dc
is
the
continuousdispersed phase velocity covariance and finally, V
˜
rel
and V
˜
drift
stand for
the
relative and the drift velocities, respectively. In order to achieve full closure, the trans port
equation for the turbule nt kinetic energy dissipation rate ( ˜) is required. The
modeling of ˜ together with all other unknown terms in Equation 23.5127 are modeled
in the same way as in [64].
RSMMixtureTurbulence Model
The main assumption for the mixture model is that all phases share the same turbu
lence field which consequently means that the term Π
R
in the Reynolds stress trans port
equations (Equation 23.5124) is neglected. Apart from that, the equations maintain the
same form but with phase properties and phase velocities being replaced with mixture
properties and mixture velocities. The mixture density, for example, can be expressed as
N
ρ
m
=
X
α
i
ρ
i
(23.5128)
i=1
while mixture velocities can be expressed as
P
N
α
i
ρ
i
U
˜
i
U
˜
m
=
i=1
(23.5129) P
N
i=1
α
i
ρ
i
where N is the number of species.
23.5.11 SolutionMethodin FLUENT
For Eulerian multiphase calculations, FLUENT uses the phase coupled SIMPLE (PC
SIMPLE) algorithm [379] for the pressurevelocity coupling. PCSIMPLE is an extension
of the SIMPLE algorithm [276] to multiphase flows. The velocities are solved coupled by
phases, but in a segregated fashion. The block algebraic multigrid scheme used by the
densitybased solver described in [394] is used to solve a vector equation formed by the
velocity components of all phases simultaneously. Then, a pressure correction equation
is built based on total volume continuity rather than mass continuity. Pressure and
velocities are then corrected so as to satisfy the continuity constrai nt.
The PressureCorrectionEquation
For incompressible multiphase flow, the pressurecorrection equation takes the form
n
(
∂
X
α ρ + ∇ · α ρ
~v
0
+ ∇ · α
n
ρ ~v
∗
−
(
X
(m˙
−
m˙
)
)) = 0 (23.5130)
k=1
ρ
rk ∂t
k k k k k k k k lk kl
l=1
ModelingMultiphaseFlows
k k
where ρ
rk
is the phase reference density for the k
th
phase (defined as the total volume
average density of phase k),
~v
0
is the velocity correction for the k
th
phase, and ~v
∗
is
the
value of ~v
k
at the current iteration. The velocity corrections are themselves expressed
as functions of the pressure corrections.
VolumeFractions
The volume fractions are obtained from the phase continuity equations. In discretized
form, the equation of the k
th
volume fraction is
a
p,k
α
k
=
X
(a
nb,k
α
nb,k
) + b
k
= R
k
(23.5131)
nb
In order to satisfy the condition that all the volume fractions sum to one,
n
X
α
k
= 1 (23.5132)
k=1
23.6 Wet SteamModelTheory
23.6.1 Overview and Limitations of the Wet Steam
Model
Overview
During the rapid expansion of steam, a condensation process will take place shortly
after the state path crosses the vaporsaturation line. The expansion process causes the
superheated dry steam to first subcool and then nucleate to form a twophase mixture
of saturated vapor and fine liquid droplets known as wet steam.
Modeling wet steam is very important in the analysis and design of steam turbines. The
increase in steam turbine exit wetness can cause severe erosion to the turbine blades at
the lowpressure stages, and a reduction in aerodynamic efficiency of the turbine stages
operating in the wet steam region [250].
FLUENT has adopted the EulerianEulerian approach for modeling wet steam flow. The
flow mixture is modeled using the compressible NavierStokes equations, in addition to
two trans port equations for the liquidphase massfraction (β), and the number of liquid
droplets per unit volume (η). The phase change model, which involves the formation of
liquiddroplets in a homogeneous nonequilibrium condensation process, is based on the
classical nonisothermal nucleation theor y.
This section describes the theoretical aspects of the wet steam model. Information about
enabling the model and using your own property functions and data with the wet steam
model is provided in Section 23.13: Setting Up the Wet Steam Model. Solution settings
and strategies for the wet steam model can be found in Section 23.14.5: Wet Steam Model.
Postprocessing variables are described in Section 23.15.1: ModelSpecific Variables.
Limitations
The following restrictions and limitations currently apply to the wet steam model in
FLUENT:
• The wet steam model is available for the densitybased solvers only.
• Pressure inlet, massflow inlet, and pressure outlet are the only inflow and
outflow boundary conditions available.
• When the wet steam model is active, the access to the Materials panel is
restricted because the fluid mixture properties are determined from the built in
steam prop erty functions or from the userdefined wet steam property functions.
Therefore, if solid properties need to be set and adjusted, then it must be done in
the Materials panel before activating the wet steam model.
23.6.2 Wet Steam Flow Equations
The wet steam is a mixture of twophases. The primary phase is the gaseousphase
consisting of watervapor (denoted by the subscript v) while the secondary phase is the
liquidphase consisting of condensedwater droplets (denoted by the subscript l).
The following assumptions are made in this model:
• The velocity slip between the droplets and gaseousphase is negligible.
• The interactions between droplets are neglected.
• The mass fraction of the condensed phase, β (also known as wetness factor),
is small (β < 0.2).
• Since droplet sizes are typically very small (from approximately 0.1 microns
to approximately 100 microns), it is assumed that the volume of the condensed
liquid phase is negligible.
From the preceding assumptions, it follows that the mixture density (ρ) can be related
to the vapor density (ρ
v
) by the following equation:
ρ =
ρ
v
(1 − β)
(23.61)
V
In addition, the temperature and the pressure of the mixture will be equivalent to the
temperature and pressure of the vaporphase.
The mixture flow is governed by the compressible NavierStokes equations given in vector
form by Equation 25.54:
∂W ∂
Z
∂Q ∂t V
I
Q dV +
Z
[F − G] · dA = H dV (23.62)
where Q=( P,u,v,w,T) are mixture quantities. The flow equations are solved using the
same densitybased solver algorithms employed for general compressible flows.
To model wet steam, two additional trans port equations are needed [152]. The first
trans port equation governs the mass fraction of the condensed liquid phase (β):
∂ρβ
∂t
+ ∇ · (ρ
−→
v β) = Γ (23.63)
where Γ is the mass generation rate due to condensation and evaporation (kg per unit
volume per second). The second trans port equation models the evolution of the number
density of the droplets per unit volume:
∂ρη
∂t
+ ∇ · (ρ
−→
v η) = ρI (23.64)
where I is the nucleation rate (number of new droplets per unit volume per second).
To determine the number of droplets per unit volume, Equation 23.61 and the average
droplet volume V
d
are combined in the following expression:
β
η =
(1 − β)V
d
(ρ
l
/ρ
v
)
(23.65)
where ρ
l
is the liquid density and the average droplet volume is defined as
3
where r
d
is the droplet radius.
V
d
=
3
πr
d
(23.66)
Together, Equation 23.62, Equation 23.63, and Equation 23.64 form a closed system
of equations which, along with Equation 23.61, permit the calculation of the wet steam
flow field.
l
23.6.3 Phase ChangeModel
The following is assumed in the phase change model:
• The condensation is homogeneous (i.e., no impurities present to form nuclei).
• The droplet growth is based on average represe ntative mean radii.
• The droplet is assumed to be spherical.
• The droplet is surrounded by infinite vapor space.
• The heat capacity of the fine droplet is negligible compared with the latent
heat released in condensation.
The mass generation rate Γ in the classical nucleation theory during the nonequilibrium
condensation process is given by the sum of mass increase due to nucleation (the formation
of critically sized droplets) and also due to growth/demise of these droplets [152].
Therefore, Γ is written as:
4
3
Γ =
3
πρ
l
I r
∗ + 4πρ
l
ηr
2
∂r
∂t
(23.67)
where r is the average radius of the droplet, and r
∗
is the KelvinHelmholtz critical
droplet radius, above which the droplet will grow and below which the droplet will
evaporate. An expression for r
∗
is given by [410].
2σ
r
∗
=
ρ RT ln S
(23.68)
where σ is the liquid surface tension evaluated at temperature T , ρ
l
is the condensed
liquid density (also evaluated at temperature T ), and S is the super saturation ratio
defined as the ratio of vapor pressure to the equilibrium saturation pressure:
P
s =
P
sat
(T )
(23.69)
The expansion process is usually very rapid. Therefore, when the state path crosses the
saturated vapor line, the process will depart from equilibrium, and the supersaturation
ratio S can take on values greater than one.
v
e
−
3
∗
θ =
The condensation process involves two mechanisms, the transfer of mass from the vapor
to the droplets and the transfer of heat from the droplets to the vapor in the form of
latent heat. This energy transfer relation was presented in [408] and used in [152] and
can be written as:
∂r
=
∂t
h
lv
ρ
l
P
√
2πRT
γ + 1
2γ
C
p
(T
0
− T ) (23.610)
where T
0
is the droplet temperature.
The classical homogeneous nucleation theory describes the formation of a liquidphase
in the form of droplets from a supersaturated phase in the absence of impurities or
foreign particles. The nucleation rate described by the steadystate classical homogeneous
nucleation theory [410] and corrected for nonisothermal effects, is given by:
I =
q
c
ρ
2
!s
2σ
4 π r
2
σ
K
b
T
(23.611)
(1 + θ) ρ
l
M
m
3
π
where q
c
is evaporation coefficient, k
b
is the Boltzmann consta nt, M
m
is mass of one
molecule, σ is the liquid surface tension, and ρ
l
is the liquid density at temperature T .
A nonisothermal correction factor, θ, is given by:
2(γ − 1)
h
lv
!
h
lv
!
− 0.5 (23.612)
(γ + 1) RT RT
where h
lv
is the specific enthalpy of evaporation at pressure p and γ is the ratio of
specific heat capacities.
23.6.4 BuiltinThermodynamicWet SteamProperties
There are many equations that describe the thermodynamic state and properties of steam.
While some of these equations are accurate in generating property tables, they are not
suitable for fast CFD computations. Therefore, FLUENT uses a simpler form of the
thermodynamic state equations [409] for efficient CFD calculations that are accurate
over a wide range of temperatures and pressures. These equations are described below.
T
647.286
ρ
#
Equationof State
The steam equation of state used in the solver, which relates the pressure to the vapor
density and the temperature, is given by [409]:
P = ρ
v
RT (1 + Bρ
v
+ C ρ
v
2
) (23.613)
where B, and C are the second and the third virial coefficients given by the following
empirical functions:
B = a (1 +
τ
)
−1
1
α
+ a
2
e
τ
(1 −
e
−τ
5
)
2
+ a
3
τ (23.614)
where B is given in m
3
/kg, τ =
1500
a
2
= 0.000942, and a
3
= 0.0004882.
with T given in Kelvin, α = 10000.0, a
1
= 0.0015,
C = a(τ − τ
0
)e
−ατ
+ b (23.615)
where C is given in m
6
/kg
2
, τ =
T
with T given in Kelvin, τ
o
= 0.8978, α=11.16,
a= 1.772, and b= 1.5E06.
The two empirical functions that define the virial coefficients B and C cover the temper
ature range from 273 K to 1073 K.
The vapor isobaric specific heat capacity C p
v
is given by:
C p
v
= C p
0
(T ) +
R
[(1 − α
v
T )(B − B
1
) − B
2
] ρ
v
+
(1 − 2α
v
T )C + α
v
T C
1
−
C
2
2
2
v
The vapor specific enthalpy, h
v
is given by:
(23.616)
h
v
= h
0
(T ) +
RT
(B − B
1
)ρ
v
+ (C
−
1
)ρ
2
2
v (23.617)
The vapor specific entropy, s
v
is given by:
s
v
= s
0
(T ) − R
"
ln ρ
v
+ (B + B
1
)ρ
v
+
(
C + C
1
)
ρ
2
2
v
(23.618)
The isobaric specific heat at zero pressure is defined by the following empirical equation:
6
C p
0
(T ) =
X
a
i
T
i−2
(23.619)
i=1
2 2
where C p
0
is in KJ/kg K, a
1
= 46.0, a
2
= 1.47276, a
3
= 8.38930E04, a
4
= 2.19989E07,
a
5
= 2.46619E10, and a
6
= 9.70466E14.
and
B
1
= T
dB
, C
1
= T
dC
, B
2
= T
2 dB
, and C = T
2 dC
.
dT dT
dT
2 2
dT
2
Both h
0
(T ) and s
0
(T ) are functions of temperature and they are defined by:
Z
h
0
(T ) = C p
0
dT + h
c
(23.620)
s
0
(T ) =
where h
c
and s
c
are arbitrary constants.
Z
C p
0
dT + s (23.621)
T
c
The vapor dynamic viscosity µ
v
and thermal conductivity K t
v
are also functions of
temperature and were obtained from [408].
SaturatedVaporLine
The saturation pressure equation as a function of temperature was obtained from [301].
The example provided in Section 23.13.5: UDWSPF Example contains a function called
wetst satP() that represents the formulation for the saturation pressure.
SaturatedLiquidLine
At the saturated liquidline, the liquid density, surface tension, specific heat C p,
dynamic viscosity, and thermal conductivity must be defined. The equation for liquid
density, ρ
l
, was obtained from [301]. The liquid surface tension equation was obtained
from [408]. While the values of C p
l
, µ
l
and K t
l
were curve fit using published data
from [91] and then written in polynomial forms. The example provided in Section
23.13.5: UDWSPF
Example contains functions called wetst cpl(), wetst mul(), and wetst ktl() that
represent formulations for C p
l
, µ
l
and K t
l
.
MixtureProperties
The mixture properties are related to vapor and liquid properties via the wetness factor
using the following mixing law:
φ
m
= φ
l
β + (1 − β)φ
v
(23.622)
where φ represents any of the following thermodynamic properties: h, s, C p, C v, µ
or
K t.
23.7 ModelingMass Transferin MultiphaseFlows
This section describes the modeling of mass transfer in the framework of FLUENT’sgen
eral multiphase models (i.e., Eulerian multiphase, mixture multiphase, VOF multiphase).
There are numerous kinds of mass transfer processes that can be modeled in FLUENT. You
can use models available in FLUENT (e.g. FLUENT’s cavitation model), or define your
own mass transfer model via userdefined functions. See Section 23.7.3: UDFPrescri bed
Mass Transfer and the separate UDF Manual for more information about the modeling
of mass transfer via userdefined functions.
Information about mass transfer is presented in the following subsections:
• Section 23.7.1: Source Terms due to Mass Transfer
• Section 23.7.2: Unidirectional Consta nt Rate Mass Transfer
• Section 23.7.3: UDFPrescri bed Mass Transfer
• Section 23.7.4: Cavitation Models
i
Note that FLUENT’s current cavitation model can only be used in
the framework of the mixture multiphase model.
23.7.1 Source Terms due to Mass
Transfer
FLUENT adds contributions due to mass transfer only to the momentum, species, and
energy equations. No source term is added for other scalars such as turbulence or user
defined scalars.
Let m
p
i
q
j be the mass transfer rate per unit volume from the i
th
species of phase p to
the j
th
species of phase q. In case a particular phase does not have a mixture material
associated with it, the mass transfer will be with the bulk phase.
Mass Equation
The contribution to the mass source for phase p in a cell is
m
p
= −m
p
i
q
j (23.71)
and for phase q is
m
q
= m
p
i
q
j (23.72)
23.7ModelingMassTransferin Multiphase
Flows
p
p
p q
p
m
i
m
j
i
MomentumEquation
For VOF or mixture models, there is no momentum source.
For the Eulerian model, the momentum source in a cell for phase p is
m
p
~u
p
= −m
p
i
q
j ~u
p
(23.73)
and for phase q is
m
q
~u
q
= m
p
i
q
j ~u
p
(23.74)
Energy Equation
For all multiphase models, the following energy sources are added.
The energy source in a cell for phase p is
H
p
= −m
p
i
q
j (h
i
) (23.75)
and for phase q is
H
q
= m
p
i
q
j (h
i
+ h
f
p
− h
fj
q
) (23.76)
where h
f
i
and h
f
j
are the formation enthalpies of species i of phase p and species j of
phase q respectively and h
i
is the enthalpy of species i of phase p (with reference to the
formation enthal py).
Species Equation
The species source in a cell for species i of phase p is
p
= −m
p
i
q
j (23.77)
and for species j of phase q is
q
= m
p
i
q
j (23.78)
OtherScalarEquations
No source/sink terms are added for turbulence quantities and other scalars. The transfer
of these scalar quantities due to mass transfer could be modeled using userdefined source
terms.
23.7.2 Unidirectional ConstantRate Mass Transfer
The unidirectional mass transfer model defines a positive mass flow rate per unit volume
from phase p to phase q:
m˙
pq
= max[0, λ
pq
] − max[0, −λ
pq
] (23.79)
where
λ
pq
= r˙α
p
ρ
q
(23.710)
and r˙ is a constant rate of particle shrinking or swelling, such as the rate of burning
of a liquid droplet. This is not available for the VOF model.
If phase p is a mixture material and a mass transfer mechanism is defined for species i
of phase p, then
λ
pq
= r˙α
p
y
p,i
ρ
q
(23.711)
where y
p,i
is the mass fraction of species i in phase p.
23.7.3 UDFPrescribedMass Transfer
Because there is no universal model for mass transfer, FLUENTprovides a UDF that you
can use to input models for different types of mass transfer, e.g. evaporation, condensa
tion, boiling, etc. Note that when using this UDF, FLUENT will automatically add the
source contribution to all relevant momentum and scalar equations. This
contribution is based on the assumption that the mass “created” or “destroyed” will
have the same momentum and energy of the phase from which it was created or
destroyed. If you would like to input your source terms directly into momentum, energy,
or scalar equations, then the appropriate path is to use UDFs for userdefined sources for
all equations, rather than the UDF for mass transfer. See the separate UDF Manual
for more information about UDFbased mass transfer in multiphase.
23.7.4 CavitationModels
This section provides information about the cavitation model used in FLUENT. You can
use FLUENT’s current cavitation model to include cavitation effects in twophase flows
when the mixture model is used.
Overviewof the CavitationModel
A liquid at constant temperature can be subjected to a decreasing pressure, which may
fall below the saturated vapor pressure. The process of rupturing the liquid by a decrease
of pressure at constant temperature is called cavitation. The liquid also contains the
microbubbles of noncondensable (dissolved or ingested) gases, or nuclei, which under
decreasing pressure may grow and form cavities. In such processes, very large and steep
density variations happen in the lowpressure/c avitating regions.
The cavitation model implemented here is based on the socalled “full cavitation model”,
developed by Singhal et al. [334]. It accounts for all firstorder effects (i.e., phase change,
bubble dynamics, turbule nt pressure fluctuations, and noncondensable gases). However,
unlike the original approach [334] assuming singlephase, isothermal, variable fluid densi ty
flows, the cavitation model in FLUENT is under the framework of multiphase flows. It
has the capability to account for multiphase (Nphase) flows or flows with multiphase
species trans port, the effects of slip velocities between the liquid and gaseous phases, and
the thermal effects and compressibility of both liquid and gas phases. The cavitation
model can be used with the mixture multiphase model (with or without slip velocities).
The complete cavitation model capability in FLUENTcan be presented in two parts:
• the basic cavitation model
This includes a description of the fundame ntal modeling approach and the standard
twophase cavitation model.
• the extended cavitation model capabili ty
This includes a description of the extension of the cavitation model for multiphase
(Nphase) flows, or flows with multiphase species trans port applications.
Basic CavitationModel
In the standard twophase cavitation model, the following assumptions are made:
• The system under investigation involves only two phases (a liquid and its
vapor), and a certain fraction of separately modeled noncondensable gases.
• Both bubble formation (evaporation) and collapse (condensation) are taken
into account in the model.
• The mass fraction of noncondensable gases is known in
advance. The cavitation model offers the following capabilities:
• The cavitation model accounts for the mass transfer between a single liquid and
its vapor.
• It is compatible with all the available turbulence models in FLUENT.
• It can be solved with the mixture energy equation.
• It is fully compatible with dynamic mesh and nonconformal interfaces.
• Both liquid and vapor phases can be incompressible or compressible. The
noncon densible gases are assumed to always be compressible. For compressible
liquids, the density can be described using a userdefined function. See the separate
UDF Manual for more information on userdefined density functions.
• The parameters used in the mass transfer model for cavitation (vaporization
pres sure, liquid surface tension coefficient) can be either a constant or a
function of temperature.
The following limitations apply to the cavitation model in FLUENT:
• The cavitation model cannot be used with the VOF model, because the
surface tracking schemes for the VOF model are incompatible with the
interpenetrating continua assumption of the cavitation model.
• The cavitation model can be used only for multiphase simulations that use
the mixture model. It is always preferable to solve for cavitation using the
mixture model without slip velocity; slip velocities can be turned on if the problem
suggests that there is significant slip between phases.
• The cavitation model can only used for cavitating flow occurring in a single
liquid fluid;
• With the cavitation model, the primary phase must be liquid, the secondary
phase must be vapor.
VaporMassFractionandVaporTransport
The working fluid is assumed to be a mixture of liquid, vapor and noncondensable gases.
Standard governing equations in the mixture model and the mixture turbulence model
describe the flow and account for the effects of turbulence. A vapor trans port equation
governs the vapor mass fraction, f , given by:
∂
∂t
(ρf ) + ∇(ρv~
v
f ) = ∇(γ∇f ) + R
e
− R
c
(23.712)
where ρ is the mixture densi ty, v~
v
is the velocity vector of the vapor phase, γ is
the
effective exchange coefficient, and R
e
and R
c
are the vapor generation and condensation
rate terms (or phase change rates). The rate expressions are derived from the Rayleigh
Plesset equations, and limiting bubble size considerations (interface surface area per
unit volume of vapor) [334]. These rates are functions of the instantaneous, local static
pressure and are given by:
when p < p
sat
when p > p
sat
R
e
= C
e
V
ch
ρ ρ
σ
l v
s
2(p
sat
− p)
3ρ
l
(1 − f ) (23.713)
R
c
= C
c
V
ch
ρ ρ
σ
l v
s
2(p − p
sat
)
3ρ
l
f (23.714)
where the suffixes l and v denote the liquid and vapor phases, V
ch
is a characteristic
velocity, which is approximated by the local turbulence intensity, (i.e. V
ch
=
√
k), σ is
the surface tension coefficient of the liquid, p
sat
is the liquid saturation vapor pressure at
the given temperature, and C
e
and C
c
are empirical consta nts. The default values are
C
e
= 0.02 and C
c
= 0.01.
ρ
TurbulenceInducedPressure Fluctuations
Significant effect of turbulence on cavitating flows has been reported [311]. FLUENT’s
cavitation model accounts for the turbulenceinduced pressure fluctuations by simply
raising the phasechange threshold pressure from p
sat
to
where
1
p
v
=
2
(p
sat
+ p
turb
) (23.715)
p
turb
= 0.39ρk (23.716)
where k is the local turbulence kinetic energy.
Effectsof NoncondensableGases
The operating liquid usually contains small finite amounts of noncondensable gases (e.g.,
dissolved gases, aeration). Even a very small amount (e.g., 10 ppm) of noncondensable
gases can have significant effects on the cavitating flow field due to expansion at low
pressures (following the ideal gas law). In the present approach, the working fluid is
assumed to be a mixture of the liquid phase and the gaseous phase, with the gaseous
phase comprising of the liquid vapor and the noncondensable gases. The density of the
mixture, ρ, is calculated as
ρ = α
v
ρ
v
+ α
g
ρ
g
+ (1 − α
v
− α
g
)ρ
l
(23.717)
where ρ
l
, ρ
v
, and ρ
g
are the densities of the liquid, the vapor, and the noncondensable
gases, respectively, and α
l
, α
v
, and α
g
are the respective volume fractions. The relation
ship between the mass fraction (f
i
) in Equations 23.712–23.714 and the volume fraction
(α
i
) in Equation 23.717 is
ρ
α
i
= f
i
i
(23.718)
The combined volume fraction of vapor and gas (i.e., α
v
+ α
g
) is commonly referred to
as the void fraction (α).
It may be noted that the noncondensable gas is not defined as a phase or a material.
When using the ideal gas law to compute the noncondensable gas density, the molecular
weight and temperature are required. By default, the gas is assumed to be air and the
molecular weight is set to 29.0. However, if the noncondensable gas is not air, then
the molecular weight can be changed by using a text command. For more information,
contact your FLUENTsupport engineer.
As for the temperature, the default value is set to 300 K when the energy equation is not
acti vated. If the temperature is different, but still a constant (i.e., isothermal flow), you
can change the temperature in FLUENTin the following way:
• Activate the energy equation.
Define −→ Models
−→Energy...
• Open the Solution Initialization panel.
Solve −→ Initialize−→Initialize...
• In the Solution Initialization, set the initial value as a desired temperature.
• Open the Solution Controls panel.
Solve −→ Controls
−→Solution...
• In the Solution Controls panel, under Equations, turn off the energy equation
by deselecting Energy in the list.. By so doing, FLUENT uses the initial values for
the temperature.
Phase ChangeRates
After accounting for the effects of turbulenceinduced pressure fluctuations and noncon
densable gases, the final phase rate expressions are written as:
when p < p
v
when p > p
v
R
e
= C
e
√
k
ρ
l
ρ
v
σ
s
2(p
v
− p)
3ρ
l
(1 − f
v
− f
g
) (23.719)
√
k
s
2(p − p
v
)
R
c
= C
c
ρ
l
ρ
l
σ 3ρ
l
f
v
(23.720)
Additional Guidelinesfor the CavitationModel
In practical applications of the cavitation model, several factors greatly influence nu
merical stability. For instance, high pressure difference between the inlet and exit, large
ratio of liquid to vapor density, and near zero saturation pressure all cause unfavorable
effects on solution convergence. In addition, poor initial conditions very often lead to an
unrealistic pressure field and unexpected cavitating zones, which, once present, are then
usually very difficult for the model to correct. The following is a list of factors that must
be considered when using the cavitation model, along with tips to help address potential
numerical problems:
• relaxation factors
In general, small relaxation factors are advised for momentum equations, usually,
between 0.05 – 0.4; The relaxation factor for the pressurecorrection equation should
usually be larger than those for momentum equations, say in the range 0.2 – 0.7.
The density and the vaporization mass (source term in the vapor equation) can also
be relaxed to improve convergence, Typically, the relaxation factor for density is set
between the values of 0.3 and 1.0, while for the vaporization mass values between
0.1 and 1.0 may be appropriate. For some extreme cases, even smaller relaxation
factors may be required for all the equations.
• initial conditions
Though no special initial condition settings are required, it is suggested that the
vapor fraction is always set to inlet values. The pressure is set close to the high
est pressure among the inlets and outlets to avoid unexpected low pressure and
cavitating spots. Also, in complicated cases, it may be beneficial to obtain a real
istic pressure field before substa ntial cavities are formed. This can be achieved by
performing the following steps:
1. Set near zero relaxation factors for the vaporization mass and for density, and
increase them to reasonable values after a sufficient number of iterations.
2. Obtain a converged / nearconverged solution for a single phase liquid
flow, and then switch on the cavitation model.
• noncondensable gases
Noncondensable gases are usually present in liquids. Even a small amount
(e.g.,
15 ppm) of noncondensable gases can have significant effects on both the physical
realism and the convergence characteristics of the solution. A value of zero for the
mass fraction of noncondensable gases should generally be avoided. In some cases, if
the liquid is purified of noncondensable gases, a much smaller value (e.g., 10
−8
)
may be used to replace the default value of 1.5×10
−5
. In fact, higher mass
fractions of the noncondensable gases may, in many cases, enhance numerical
stability and lead to more realistic results. In particular, when the saturation
pressure of a liquid at a
certain temperature is zero or very small, noncondensable gases will play a crucial
role both numerically and physically.
• limits for dependent variables
In many cases, setting the pressure upper limit to a reasonable value can help
convergence greatly at the early stage of the solution. It is advised to always limit
the maximum pressure when it is possible. By default, FLUENTsets the maximum
pressure limit to 5.0×10
10
Pascal.
• the relaxation factor for the pressure correction equation
For cavitating flows, a special relaxation factor is introduced for the pressure cor
rection equation. By default, this factor is set to 0.7, which should work well for
most of the cases. For some very complicated cases, however, you may
experience the divergence of the AMG solver. Under those circumstances, this
value may be reduced to no less than 0.4. You can set the value of this relaxation
factor by typing a text command. For more information, contact your FLUENT
support engineer.
• pressure discretization schemes
As for many multiphase flows, it is more desirable to use the following pressure
discretization schemes in cavitation applications:
– body force weighted
– second order
– PRESTO!
The standard and linear schemes generally are not very effective in complex cavi
tating flows.
ExtendedCavitationModelCapability
In many practical applications, when cavitation occurs, there exist other gaseous species
in the systems investigated. For instance, in a ventilated supercavitating vehicle, air is
injected into a liquid to stabilize or increase the cavitation along the vehicle surfaces. Also
in some cases, the incoming flow is a mixture of a liquid and some gaseous species. In
order to be able to predict those type of cavitating flows, the basic twophase cavitation
model needs to be extended to a multiphase (Nphase) flows, or a multiphase species
trans port cavitation model.
MultiphaseCavitationModel
The multiphase cavitation model is an extension of the basic twophase cavitation model
to multiphase flows. In addition to the primary liquid and secondary vapor phase, more
secondary gaseous phases can be included into the computational system under the fol
lowing assumptions/limitations:
• Mass transfer (cavitation) only occurs between the first and the second phases.
• The basic cavitation model is still used to model the phase changes between
the liquid and vapor.
• Only one secondary phase can be defined as compressible gas phase, while a
user defined density may be applied to all the phases.
• The predescribed noncondensable gases can still be included in the system.
To exclude noncondensable gases from the system, the mass fraction needs to be
set
to 0, and the noncondensable gas needs to be modeled by a separate compressible
gas phase.
• For an noncavitating phase i, the general trans port equation governing the
mass fraction f
i
given by:
∂
∂t
(ρf
i
) + ∇(ρv~
v
f
i
) = ∇(γ
f
i
∇f
i
) + S
f
i
(23.721)
where S
f
i
is a (userdefined) source term. By default, S
f
i
= 0.
MultiphaseSpecies Transport CavitationModel
In some cases, there are several gas phase components in a system. It is desirable to
consider them all compressible. Since only one compressible gas phase is allowed in the
general multiphase approach, the multiphase species trans port approach offers an option
to handle these type of applications by assuming that there is one compressible gas phase
with multiple species.
The detailed description of the multiphase species trans port approach can be found in
Section 23.8: Modeling Species Transport in Multiphase Flows. The multiphase species
trans port cavitation model can be summarized as follows:
• All the assumptions/limitations for the multiphase cavitation model also
apply here.
• The primary phase can only be a single liquid.
• All the secondary phases allow more than one species.
23.8ModelingSpecies Transport in Multiphase
Flows
∂
∂t
q
• The vapor, either as a phase or a species, must be in the second phase.
• The mass transfer between a liquid and a vapor phase/s pecies is modeled by
the basic cavitation model.
• The mass transfer between other phases or species are modeled with the
standard mass transfer approach. In the standard model, the zero constant rate
should be chosen.
• For the phases with multiple species, the phase shares the same pressure as
the other phases, but each species has its own pressure (i.e., partial pressure).
As a result, the vapor density and the pressure used in Equation 23.719 are the
partial density and pressure of the vapor.
23.8 ModelingSpecies Transport in MultiphaseFlows
Species trans port, as described in Chapter 14: Modeling Species Transport and Finite
Rate Chemistry, can also be applied to multiphase flows. You can choose to solve the
conservation equations for chemical species in multiphase flows by having FLUENT, for
each phase k, predict the local mass fraction of each species, Y
i
k
, through the solution
of a convectiondiffusion equation for the i
th
species. The generalized chemical species
conservation equation (Equation 14.11), when applied to a multiphase mixture can be
represented in the following form:
(ρ
q
α
q
Y
i
q
)+∇·(ρ
q
α
q
~v
q
Y
i
q
) = −∇·α
q
J
~
i
n
+α
q
R
i
q
+α
q
S
i
q
+
X
(m˙
p
i
q
j − m˙
q
j
p
i )+R
(23.81)
p=1
where R
i
q
is the net rate of production of homogeneous species i by chemical reaction for
phase q, m˙
q
j
p
i is the mass transfer source between species i and j from phase q to p,
and R is the heterogeneous reaction rate. In addition, α
q
is the volume fraction for
phase q and S
i
q
is the rate of creation by addition from the dispersed phase plus any
userdefined sources.
FLUENT treats homogeneous gas phase chemical reactions the same as a singlephase
chemical reaction. The reacta nts and the products belong to the same mixture material
(set in the Species Model panel), and hence the same phase. The reaction rate is scaled
by the volume fraction of the particular phase in the cell.
The setup of a homogeneous gas phase chemical reaction in FLUENTis the same as it is
for a single phase. For more information, see Chapter 14: Modeling Species Transport
and FiniteRate Chemistry. For most multiphase species trans port problems, boundary
con ditions for a particular species are set in the associated phase boundary condition
panel (see Chapter 23.9.8: Defining Multiphase Boundary Conditions), and
postprocessing and
reporting of results is performed on a perphase basis (see Section 23.15: Postprocessing
for Multiphase Modeling).
For multiphase species trans port simulations, the Species Model panel allows you to in
clude Volumetric, Wall Surface, and Particle Surface reactions. FLUENTtreats multiphase
surface reactions as it would a singlephase reaction. The reaction rate is scaled with
the volume fraction of the particular phase in the cell. For more information, see Chap
ter 14: Modeling Species Transport and FiniteRate Chemistry .
i
To turn off reactions for a particular phase, while keeping the
reactions active for other phases. turn on Volumetric under Reactions in
the Species
Model panel. Then, in the Materials panel, select none from the Reactions
dropdown list.
The species of different phases is entirely independent. There is no implicit relationship
between them even if they share the same name. Explicit relationships between species of
different phases can be specified through mass transfer and heterogeneous reactions. For
more information on mass transfer and heterogeneous reactions, see Section 23.9.7: In
cluding Mass Transfer Effects and Section 23.9.6: Specifying Heterogeneous Reactions ,
respectively.
Some phases may have a fluid material associated with them instead of a mixture ma
terial. The species equations are solved in those phases that are assigned a mixture
material. The species equation above is solved for the mass fraction of the species in a
particular phase. The mass transfer and heterogeneous reactions will be associated with
the bulk fluid for phases with a single fluid material.
23.8.1 Limitations
The following limitations exist for the modeling of species trans port for multiphase
flows:
• The nonpremixed, premixed, partiallypremixed combustion, or the
composition PDF trans port species trans port models are not available for
multiphase species reactions.
• The stiff chemistry solver is not available for multiphase species reactions.
• Only the laminar finiterate, finiterate/eddydissipation and eddydissipation
turbulence chemistry models of homogeneous reactions are available for multiphase
species trans port.
• The discrete phase model (DPM) is not compatible with multiphase species
trans port.
23.8.2 Mass and MomentumTransferwith MultiphaseSpecies Transport
The FLUENT multiphase mass transfer model accommodates mass transfer between
species belonging to different phases. Instead of a matrix type input, multiple mass
transfer mechanisms need to be input. Each mass transfer mechanism defines the mass
transfer phenomenon from one entity to another entity. An entity is either a particular
species in a phase, or the bulk phase itself if the phase does not have a mixture mate
rial associated with it. The mass transfer phenomenon could be specified either through
the inbuilt unidirectional “consta ntrate” mass transfer (Section 23.7.2: Unidirectional
Consta nt Rate Mass Transfer) or through userdefined functions.
FLUENT loops through all the mass transfer mechanisms to compute the net mass
source/sink of each species in each phase. The net mass source/sink of a species is
used to compute species and mass source terms. FLUENT will also automatically add
the source contribution to all relevant momentum and energy equations based on that
assumption that the momentum and energy carried along with the transferred mass. For
other equations, the trans port due to mass transfer needs to be explicitly modeled by
the user.
Source Terms due to Heterogeneous Reactions
Consider the following reaction:
aA + bB → cC + dD (23.82)
Let as assume that A and C belong to phase 1 and B and D to phase 2.
Mass Transfer
Mass source for the phases are given by:
S
1
= R(cM
c
− aM
a
) (23.83)
S
2
= R(dM
d
− bM
b
) (23.84)
where S is the mass source, M is the molecular weight, and R is the reaction rate.
The general expression for the mass source for the i
th
phase is
X
r r
S
r
i
= −R
ri
γ
j
M
j
(23.85)
X
p p
S
p
i
= R
pi
γ
j
M
j
(23.86)
S
i
= S
p
i
+ S
r
i
(23.87)
where γ is the stoichiometric coefficient, p represents the product, and r represents the
reacta nt.
=
R(cM
c
~u
net
− aM
a
~u
1
)
(23.810)
= R(dM
d
~u
net
− bM
b
~u
2
) (23.811)
P
S
S
r j
r
M
p
p
i
k k
j
i
MomentumTransfer
Momentum transfer is more complicated, but we can assume that the reactants mix
(conserving momentum) and the products take momentum in the ratio of the rate of
their formation.
The net velocity, ~u
net
, of the reacta nts is given
by:
aM
a
~u
1
+
bM
b
~u
2
~u
net
=
aM
a
+ bM
b
(23.88)
The general expression for the net velocity of the reacta nts is given by:
r
γ
r
M
r
~u
r
~u
net
=
j j j
P
γ
r r
(23.89)
r j
M
j
where j represents the j
th
item (either a reacta nt or a product).
Momentum transfer for the phases is then given by:
~
u
1
~
u
2
The general expression is
S
~
u
X
r r
i
= S
pi
~u
net
− R
ri
γ
j
M
j
~u
i
(23.812)
If we assume that there is no momentum transfer, then the above term will be zero.
Species Transfer
The general expression for source for k
th
species in the j
th
phase is
S k = −R
X
γ
r r
k
i
(23.813)
p
k
p
k
S k = R
X
γ
j
M
j
(23.814)
k
i
S
k
i
= S
p
k + S
r
k (23.815)
i i
P
S
H
S
H
c
d
P
i
r
r
P
HeatTransfer
For heat transfer, we need to consider the formation enthalpies of the reacta nts and
products as well:
The net enthalpy of the reacta nts is given by:
aM
a
(H
a
+ h
f
) + bM
b
(H
b
+ h
f
)
H
net
=
a b
aM
a
+ bM
b
(23.816)
where h
f
represents the formation enthalpy, and H represents the enthal py.
The general expression for H
net
is:
r
γ
r
M
r
(H
r
+ h
f
)
H
net
=
j j j j
P (23.817)
r
γ
r r
j
M
j
If we assume that this enthalpy gets distributed to the products in the ratio of their mass
production rates, heat transfer for the phases are given by:
1
= R(cM
c
H
net
− aM
a
H
a
2
= R(dM
d
H
net
− bM
b
H
b
− cM
c
h
f
) (23.818)
− dM
d
h
f
) (23.819)
The last term in the above equations appears because our enthalpy is with reference to
the formation enthal py.
The general expression for the heat source is:
S
H
X
r r r
!
X
p p
f
p
i
= S
pi
H
net
−
R
γ
j
M
j
H
j
+
ri
γ
j
M
j
h
j
pi
(23.820)
If we assume that there is no heat transfer, we can assume that the different species only
carry their formation enthalpies with them. Thus the expression for H
net
will be:
r
γ
r
M
r
h
f
H
net
=
j j j
(23.821)
r
γ
r r
j
M
j
The expression S
H
will be
S
H
X
p p f
p
i
= S
p
i
H
net
− R
pi
γ
j
M
j
h
j
(23.822)
23.9 Steps for Using a MultiphaseModel
The procedure for setting up and solving a general multiphase problem is outlined below,
and described in detail in the subsections that follow. Remember that only the steps that
are pertinent to general multiphase calculations are shown here. For information about
inputs related to other models that you are using in conjunction with the multiphase
model, see the appropriate sections for those models.
See also Section 23.12.1: Additional Guidelines for Eulerian Multiphase Simulations for
guidelines on simplifying Eulerian multiphase simulations.
1. Enable the multiphase model you want to use (VOF, mixture, or Eulerian) and
specify the number of phases. For the VOF model, specify the VOF formulation
as well.
Define −→ Models
−→Multiphase...
See Sections 23.9.1 and 23.10.1 for details.
2. Copy the material representing each phase from the materials database.
Define −→Materials...
If the material you want to use is not in the database, create a new material.
See Section 8.1.2: Using the Materials Panel for details about copying from the
database and creating new materials. See Sections 23.10.5 and 23.11.3 for additional
information about specifying material properties for a compressible phase (VOF
and mixture models only). It is possible to turn off reactions in some materials
by selecting none in the Reactions dropdown list under Properties in the Materials
panel.
i
If your model includes a particulate (granular) phase, you will need
to create a new material for it in the fluid materials category (not the
solid
materials category).
3. Define the phases, and specify any interaction between them (e.g., surface tension
if you are using the VOF model, slip velocity functions if you are using the mixture
model, or drag functions if you are using the Eulerian model).
Define −→Phases...
See Sections 23.9.3–23.12.2 for details.
4. (Eulerian model only) If the flow is turbule nt, define the multiphase turbulence
model.
Define −→ Models
−→Viscous...
See Section 23.12.3: Modeling Turbulence for details.
23.9 Steps for Usinga MultiphaseModel
5. If body forces are present, turn on gravity and specify the gravitational acceleration.
Define −→Operating Conditions...
See Section 23.9.4: Including Body Forces for details.
6. Specify the boundary conditions, including the secondaryphase volume fractions
at flow boundaries and (if you are modeling wall adhesion in a VOF simulation)
the contact angles at walls.
Define −→Boundary Conditions...
See Section 23.9.8: Defining Multiphase Boundary Conditions for details.
7. Set any modelspecific solution parameters.
Solve −→ Controls −→Solution...
See Sections 23.10.4 and 23.14 for details.
8. Initialize the solution and set the initial volume fractions for the secondary phases.
Solve −→ Initialize−→Patch...
See Section 23.14.1: Setting Initial Volume Fractions for details.
9. Calculate a solution and examine the results. Postprocessing and reporting of
results is available for each phase that is selected.
See Sections 23.14 and 23.15 for details.
This section provides instructions and guidelines for using the VOF, mixture, and Eule
rian multiphase models.
Information is presented in the following subsections:
• Section 23.9.1: Enabling the Multiphase Model
• Section 23.9.2: Solving a Homogeneous Multiphase Flow
• Section 23.9.3: Defining the Phases
• Section 23.9.4: Including Body Forces
• Section 23.9.5: Modeling Multiphase Species Transport
• Section 23.9.6: Specifying Heterogeneous Reactions
• Section 23.9.7: Including Mass Transfer Effects
• Section 23.9.8: Defining Multiphase Boundary Conditions
ModelingMultiphaseFlows
23.9.1 Enablingthe MultiphaseModel
To enable the VOF, mixture, or Eulerian multiphase model, select Volume of Fluid,
Mix ture, or Eulerian as the Model in the Multiphase Model panel (Figure 23.9.1).
Define −→ Models −→Multiphase...
Figure 23.9.1: The Multiphase Model Panel
The panel will expand to show the relevant inputs for the selected multiphase model.
23.9 Steps for Usinga MultiphaseModel
If you selected the VOF model, the inputs are as follows:
• number of phases
• VOF formulation (see Section 23.10.1: Choosing a VOF Formulation)
• (optional) implicit body force formulation (see Section 23.9.4: Including
Body
Forces)
If you selected the mixture model, the inputs are as follows:
• number of phases
• whether or not to compute the slip velocities (see Section 23.9.2: Solving a
Homo geneous Multiphase Flow)
• (optional) implicit body force formulation (see Section 23.9.4: Including
Body
Forces)
If you selected the Eulerian model, the input is the following:
• number of phases
To specify the number of phases for the multiphase calculation, enter the appropriate
value in the Number of Phases field. You can specify up to 20 phases.
23.9.2 Solving a Homogeneous MultiphaseFlow
If you are using the mixture model, you have the option to disable the calculation of
slip velocities and solve a homogeneous multiphase flow (i.e., one in which the phases all
move at the same velocity). By default, FLUENTwill compute the slip velocities for the
secondary phases, as described in Section 23.4.5: Relative (Slip) Velocity and the Drift
Velocity. If you want to solve a homogeneous multiphase flow, turn off Slip Velocity
under Mixture Parameters.
ModelingMultiphaseFlows
23.9.3 Definingthe Phases
To define the phases (including their material properties) and any interphase interaction
(e.g., surface tension and wall adhesion for the VOF model, slip velocity for the mixture
model, drag functions for the mixture and the Eulerian models), you will use the Phases
panel (Figure 23.9.2).
Define −→Phases...
Figure 23.9.2: The Phases Panel
Each item in the Phase list in this panel is one of two types, as indicated in the Type list:
primaryphase indicates that the selected item is the primary phase, and secondaryphase
indicates that the selected item is a secondary phase. To specify any interaction between
the phases, click the Interaction... button.
Instructions for defining the phases and interaction are provided in Sections 23.10.3,
23.11.1, and 23.12.2 for the VOF, mixture, and Eulerian models, respectively.
23.9 Steps for Usinga MultiphaseModel
23.9.4 IncludingBody Forces
When large body forces (e.g., gravity or surface tension forces) exist in multiphase flows,
the body force and pressure gradient terms in the momentum equation are almost in
equilibrium, with the contributions of convective and viscous terms small in comparison.
Segregated algorithms converge poorly unless partial equilibrium of pressure gradient and
body forces is taken into account. FLUENT provides an optional “implicit body force”
treatme nt that can account for this effect, making the solution more robust.
The basic procedure involves augmenting the correction equation for the face flow rate,
Equation 25.413, with an additional term involving corrections to the body force. This
results in extra body force correction terms in Equation 25.411, and allows the flow to
achieve a realistic pressure field very early in the iterative process.
To include this body force, turn on Gravity in the Operating Conditions panel and
specify the Gravitational Acceleration.
Define −→Operating Conditions...
For VOF calculations, you should also turn on the Specified Operating Density option
in the Operating Conditions panel, and set the Operating Density to be the density of
the lightest phase. (This excludes the buildup of hydrostatic pressure within the lightest
phase, improving the roundoff accuracy for the momentum balance.) If any of the phases
is compressible, set the Operating Density to zero.
i
For VOF and mixture calculations involving body forces, it is
recom mended that you also turn on the Implicit Body Force treatme nt
for the
Body Force Formulation in the Multiphase Model panel. This treatme nt
im proves solution convergence by accounting for the partial
equilibrium of the pressure gradient and body forces in the momentum
equations. See Section 23.9.4: Including Body Forces for details.
ModelingMultiphaseFlows
23.9.5 ModelingMultiphaseSpecies Transport
FLUENT lets you describe a multiphase species trans port and volumetric reaction (Sec
tion 23.8: Modeling Species Transport in Multiphase Flows) in a fashion that is similar
to setting up a singlephase chemical reaction using the Species Model panel (e.g., Fig
ure 23.9.3).
Define −→ Models −→ Species −→Transport &
Reaction...
Figure 23.9.3: The Species Model Panel with a Multiphase Model Enabled
1. Select Species Transport under Model.
2. Turn on Volumetric under Reactions.
3. Select a specific phase using the Phase dropdown list under Phase Properties.
4. Click the Set... button to display the Phase Properties panel (Figure 23.9.4).
23.9 Steps for Usinga MultiphaseModel
Figure 23.9.4: The Phase Properties Panel
In the Phase Properties panel, the material for each phase is listed in the Material
dropdown list. From this list, you can choose the material that you want to use
for a specific phase. The dropdown list contains all of the materials that have been
defined for your simulation. If you want to inspect or edit any of the properties
of any of the materials, then you need to open the Materials panel by clicking the
Edit... button.
5. In the Species Model panel, choose the TurbulenceChemistry Interaction model.
Three models are available:
Laminar FiniteRate computes only the Arrhenius rate (see Equation 14.18) and
neglects turbulence chemistry interaction.
EddyDissipation (for turbule nt flows) computes only the mixing rate (see Equa
tions 14.126 and 14.127).
FiniteRate/EddyDissipation (for turbule nt flows) computes both the Arrhenius
rate and the mixing rate and uses the smaller of the two.
When modeling multiphase species trans port, additional inputs may also be required
depending on your modeling needs. See, for example, Section 23.9.6: Specifying Het
erogeneous Reactions for more information defining heterogeneous reactions, or Sec
tion 23.9.7: Including Mass Transfer Effects for more information on mass transfer effects.
ModelingMultiphaseFlows
23.9.6 SpecifyingHeterogeneous Reactions
You can use FLUENTto define multiple heterogeneous reactions and stoichiometry using
the Phase Interaction panel (e.g., Figure 23.9.5).
Define −→Phases...
1. In the Phases panel (Figure 23.9.2), click the Interaction... button to open the Phase
Interaction panel.
Figure 23.9.5: The Phase Interaction Panel for Heterogeneous Reactions
2. Click the Reactions tab in the Phase Interaction panel.
3. Set the total number of reactions (volumetric reactions, wall surface reactions, and
particle surface reactions) in the Total Number of Heterogeneous Reactions field.
(Use the arrows to change the value, or type in the value and press
<Enter>.)
4. Specify the Reaction Name of each reaction that you want to define.
5. Set the ID of each reaction you want to define. (Again, if you type in the value be
sure to press <Enter>.)
i,r
or ν
23.9 Steps for Usinga MultiphaseModel
6. For each reaction, specify how many reacta nts and products are involved in the
reaction by increasing the value of the Number of Reactants and the Number of
Products. Select each reacta nt or product in the Reaction tab and then set its stoi
chiometric coefficient in the Stoich. Coefficient field. (The stoichiometric coefficient
is the constant ν
0 0 0
i,r
in Equation 14.16.)
7. For each reaction, indicate the Phase and Species and the stoichiometric coefficient
for each of your reacta nts and products.
8. For each reaction, indicate an applicable userdefined function using the Reaction
Rate Function dropdown list.
i
The heterogeneous reaction rates can only be specified using a user
defined function. A UDF is available for an Arrhenius type reaction
with rate
exponents that are equivalent to the stoichiometric coefficients.
For more information, see the separate UDF Manual.
i
FLUENT assumes that the reacta nts are mixed thoroughly before
reacting together, thus the heat and momentum transfer is based on this
assump
tion. This assumption can be deacti vated using a text command. For more
information, contact your FLUENTsupport engineer.
23.9.7 Including Mass Transfer
Effects
As discussed in Section 23.7: Modeling Mass Transfer in Multiphase Flows, mass transfer
effects in the framework of FLUENT’s general multiphase models (i.e., Eulerian multi
phase, mixture multiphase, or VOF multiphase) can be modeled in one of three ways:
• unidirectional constant rate mass transfer (not available for VOF calculations)
• UDFprescribed mass transfer
• mass transfer through cavitation (only valid for the mixture multiphase model)
Because of the different procedures and limitations involved, defining mass transfer
through cavitation is described separately in Section 23.11.2: Including Cavitation Effects.
To define mass transfer in a multiphase simulation, either as unidirectional constant or
using a UDF, you will need to use the Phase Interaction panel (e.g., Figure 23.9.6).
Define −→Phases...
1. In the Phases panel (Figure 23.9.2), click the Interaction... button to open the Phase
Interaction panel.
Figure 23.9.6: The Phase Interaction Panel for Mass Transfer
2. Click the Mass tab in the Phase Interaction panel.
3. Specify the Number of Mass Transfer Mechanisms. You can include any number
of mass transfer mechanisms in your simulation. Note also that the same pair of
phases can have multiple mass transfer mechanisms and you have the ability to
activate and deacti vate the mechanisms of your choice.
4. For each mechanism, specify the phase of the source material under From Phase.
5. If species trans port is part of the simulation, and the source phase is composed of
a mixture material, then specify the species of the source phase mixture material
in the corresponding Species dropdown list.
6. For each mechanism, specify the phase of the destination material phase under To
Phase.
7. If species trans port is part of the simulation, and the destination phase is composed
of a mixture material, then specify the species of the destination phase mixture
material in the corresponding Species dropdown list.
8. For each mass transfer mechanism, select the desired mass transfer correlation
under Mechanism. The following choices are available:
constantrate enables a consta nt, unidirectional mass transfer.
userdefined allows you to implement a correlation reflecting a model of your
choice, through a userdefined function.
FLUENTwill automatically include the terms needed to model mass transfer in all
relevant conservation equations. Another option to model mass transfer between
phases is through the use of userdefined sources and their inclusion in the rele
vant conservation equations. This approach is a more involved but more powerful,
allowing you to split the source terms according to a model of your choice.
i
Note that momentum, energy, and turbulence are also trans ported with
the mass that is transferred. FLUENT assumes that the reacta nts are
mixed
thoroughly before reacting together, thus the heat and momentum transfer
is based on this assumption. This assumption can be deactivated using
a text command. For more information, contact your FLUENT support
engineer.
When your model involves the trans port of multiphase species, you can define a mass
transfer mechanism between species from different phases. If a particular phase does not
have a species associated with it, then the mass transfer throughout the system will be
performed by the bulk fluid material.
i
Note that including species trans port effects in the mass trans port of
multi phase simulation requires that Species Transport be turned on in the
Species
Model panel.
Define −→ Models −→ Species −→Transport &
Reaction...
23.9.8 Defining Multiphase Boundary
Conditions
Multiphase boundary conditions are set in the Boundary Conditions panel (Figure 23.9.7),
but the procedure for setting multiphase boundary conditions is slightly different than
for singlephase models. You will need to set some conditions separately for individual
phases, while other conditions are shared by all phases (i.e., the mixture), as described
in detail below.
Define −→Boundary
Conditions...
Figure 23.9.7: The Boundary Conditions Panel
BoundaryConditionsfor the Mixtureand the Individual Phases
The conditions you need to specify for the mixture and those you need to specify for the
individual phases will depend on which of the three multiphase models you are using.
Details for each model are provided below.
VOFModel
If you are using the VOF model, the conditions you need to specify for each type of zone
are listed below and summarized in Table 23.9.1.
• For an exhaust fan, inlet vent, intake fan, outlet vent, pressure inlet, pressure
outlet, or velocity inlet, there are no conditions to be specified for the primary
phase. For each secondary phase, you will need to set the volume fraction as a
consta nt, a profile (see Section 7.26: Boundary Profiles), or a userdefined
function (see the separate UDF Manual). All other conditions are specified for the
mixture.
• For a mass flow inlet, you will need to set the mass flow rate or mass flux for
each individual phase. All other conditions are specified for the mixture.
i
Note that if you read a VOF case that was set up in a version of FLUENT
prior to 6.1, you will need to redefine the conditions at the mass flow inlets.
• For an axis, fan, outflow, periodic, porous jump, radiator, solid, symmetry, or
wall zone, all conditions are specified for the mixture. There are no conditions to
be set for the individual phases.
• For a wall zone, you can specify the contact angle if wall adhesion option is
enabled.
• For a fluid zone, mass sources are specified for the individual phases, and all
other sources are specified for the mixture.
If the fluid zone is not porous, all other conditions are specified for the mixture.
If the fluid zone is porous, you will enable the Porous Zone option in the Fluid panel
for the mixture. The porosity inputs (if relevant) are also specified for the mixture.
The resistance coefficients and direction vectors, however, are specified separately
for each phase. See Section 7.19.6: User Inputs for Porous Media for details about
these inputs. All other conditions are specified for the mixture.
See Chapter 7: Boundary Conditions for details about the relevant conditions for each
type of boundary. Note that the pressure farfield boundary is not available with the
VOF model.
Table 23.9.1: PhaseSpecific and Mixture Conditions for the VOF Model
Type Primary Phase Secondary Phase Mixture
exhaust fan;
inlet vent;
intake fan;
outlet vent;
pressure inlet;
pressure outlet;
velocity inlet
nothing volume fraction all others
mass flow inlet mass flow/flux mass flow/flux all others
axis;
fan; outflow;
periodic;
porous jump;
radiator;
solid;
symmetry;
wall
nothing nothing all others
pressure farfield not available not available not available
fluid mass source;
other porous inputs
mass source;
other porous inputs
porous zone;
porosity;
all others
MixtureModel
If you are using the mixture model, the conditions you need to specify for each type of
zone are listed below and summarized in Table 23.9.2.
• For an exhaust fan, outlet vent, or pressure outlet, there are no conditions to
be specified for the primary phase. For each secondary phase, you will need to set
the volume fraction as a consta nt, a profile (see Section 7.26: Boundary
Profiles), or a userdefined function (see the separate UDF Manual). All other
conditions are specified for the mixture.
• For an inlet vent, intake fan, or pressure inlet, you will specify for the mixture
which direction specification method will be used at this boundary (Normal to
Boundary or Direction Vector). If you select the Direction Vector specification
method, you will specify the coordinate system (3D only) and flowdirection
components for the individual phases. For each secondary phase, you will need to
set the volume fraction (as described above). All other conditions are specified for
the mixture.
• For a mass flow inlet, you will need to set the mass flow rate or mass flux for
each individual phase. All other conditions are specified for the mixture.
i
Note that if you read a mixture multiphase case that was set up in a
version of FLUENTprevious to 6.1, you will need to redefine the conditions
at the
mass flow inlets.
• For a velocity inlet, you will specify the velocity for the individual phases. For
each secondary phase, you will need to set the volume fraction (as described
above). All other conditions are specified for the mixture.
• For an axis, fan, outflow, periodic, porous jump, radiator, solid, symmetry, or
wall zone, all conditions are specified for the mixture. There are no conditions to
be set for the individual phases. Outflow boundary conditions are not available
for the cavitation model.
• For a fluid zone, mass sources are specified for the individual phases, and all
other sources are specified for the mixture.
If the fluid zone is not porous, all other conditions are specified for the mixture.
If the fluid zone is porous, you will enable the Porous Zone option in the Fluid panel
for the mixture. The porosity inputs (if relevant) are also specified for the mixture.
The resistance coefficients and direction vectors, however, are specified separately
for each phase. See Section 7.19.6: User Inputs for Porous Media for details about
these inputs. All other conditions are specified for the mixture.
See Chapter 7: Boundary Conditions for details about the relevant conditions for each
type of boundary. Note that the pressure farfield boundary is not available with the
mixture model.
Table 23.9.2: PhaseSpecific and Mixture Conditions for the Mixture Model
Type Primary Phase Secondary Phase Mixture
exhaust fan;
outlet vent;
pressure outlet
nothing volume fraction all others
inlet vent;
intake fan;
pressure inlet
coord. system;
flow direction
coord. system;
flow direction;
volume fraction
dir. spec.
method;
all others
mass flow inlet mass flow/flux mass flow/flux all others
velocity inlet velocity velocity;
volume fraction
all others
axis;
fan;
outflow (n/a for
cavitation
model);
periodic;
porous jump;
radiator;
solid;
symmetry;
wall
nothing nothing all others
pressure farfield not available not available not available
fluid mass source;
other porous inputs
mass source;
other porous inputs
porous zone;
porosity;
all others
EulerianModel
If you are using the Eulerian model, the conditions you need to specify for each type
of zone are listed below and summarized in Tables 23.9.3, 23.9.4, 23.9.5, and 23.9.6.
Note that the specification of turbulence parameters will depend on which of the three
multiphase turbulence models you are using, as indicated in Tables 23.9.4–23.9.6. See
Sections 23.5.10 and 23.12.3 for more information about multiphase turbulence models.
• For an exhaust fan, outlet vent, or pressure outlet, there are no conditions to
be specified for the primary phase if you are modeling laminar flow or using the
mixture turbulence model (the default multiphase turbulence model), except for
backflow total temperature if heat transfer is on.
For each secondary phase, you will need to set the volume fraction as a constant,
a profile (see Section 7.26: Boundary Profiles), or a userdefined function (see the
separate UDF Manual). If the phase is granular, you will also need to set its
granular temperature. If heat transfer is on, you will also need to set the backflow
total temperature.
If you are using the mixture turbulence model, you will need to specify the turbu
lence boundary conditions for the mixture. If you are using the dispersed turbulence
model, you will need to specify them for the primary phase. If you are using the
perphase turbulence model, you will need to specify them for the primary phase
and for each secondary phase.
All other conditions are specified for the mixture.
• For an inlet vent, intake fan, or pressure inlet, you will specify for the mixture
which direction specification method will be used at this boundary (Normal to
Boundary or Direction Vector). If you select the Direction Vector specification
method, you will specify the coordinate system (3D only) and flowdirection
components for the individual phases. If heat transfer is on, you will also need
to set the total temperature for the individual phases.
For each secondary phase, you will need to set the volume fraction (as described
above). If the phase is granular, you will also need to set its granular temperature.
If you are using the mixture turbulence model, you will need to specify the turbu
lence boundary conditions for the mixture. If you are using the dispersed turbulence
model, you will need to specify them for the primary phase. If you are using the
perphase turbulence model, you will need to specify them for the primary phase
and for each secondary phase.
All other conditions are specified for the mixture.
• For a mass flow inlet, you will need to set the mass flow rate or mass flux for
each individual phase. You will also need to specify the temperature of each phase,
since the energy equations are solved for each phase.
For mass flow inlet boundary conditions, you can specify the slip velocity between
phases. When you select a mass flow inlet boundary for the secondary phase, two
options will be available for the Slip Velocity Specification Method, as shown in
Figure 23.9.8:
– Velocity Ratio
The value for the phase velocity ratio is the secondary phase to primary phase
velocity ratio. By default, it is 1.0, which means velocities are the same (no
slip). By entering a ratio that is greater than 1.0, you are indicating a larger
secondary phase velocity. Otherwise, you can enter a ratio that is less than
1.0 to indicate a smaller secondary phase velocity.
– Volume Fraction
If you specify the volume fraction at an inlet, FLUENTwill calculate the phase
velocities.
i
If a secondary phase has zero mass flux (i.e., the Eulerian model is used
to run a single phase case), neither Phase Velocity Ratio nor Volume
Fraction
will affect the solution.
Figure 23.9.8: MassFlow Inlet Boundary Condition Panel
• For a velocity inlet, you will specify the velocity for the individual phases. If
heat transfer is on, you will also need to set the total temperature for the
individual phases.
For each secondary phase, you will need to set the volume fraction (as described
above). If the phase is granular, you will also need to set its granular temperature.
If you are using the mixture turbulence model, you will need to specify the turbu
lence boundary conditions for the mixture. If you are using the dispersed turbulence
model, you will need to specify them for the primary phase. If you are using the
perphase turbulence model, you will need to specify them for the primary phase
and for each secondary phase.
All other conditions are specified for the mixture.
• For an axis, outflow, periodic, solid, or symmetry zone, all conditions are
specified for the mixture. There are no conditions to be set for the individual
phases.
• For a wall zone, shear conditions are specified for the individual phases. All
other conditions are specified for the mixture, including thermal boundary
conditions, if heat transfer is on.
• For a fluid zone, all source terms and fixed values are specified for the
individual phases, unless you are using the mixture turbulence model or the
dispersed tur bulence model. If you are using the mixture turbulence model,
source terms and fixed values for turbulence are specified instead for the mixture.
If you are using the dispersed turbulence model, they are specified only for the
primary phase.
If the fluid zone is not porous, all other conditions are specified for the mixture.
If the fluid zone is porous, you will enable the Porous Zone option in the Fluid panel
for the mixture. The porosity inputs (if relevant) are also specified for the mixture.
The resistance coefficients and direction vectors, however, are specified separately
for each phase. See Section 7.19.6: User Inputs for Porous Media for details about
these inputs. All other conditions are specified for the mixture.
See Chapter 7: Boundary Conditions for details about the relevant conditions for each
type of boundary. Note that the pressure farfield, fan, porous jump, radiator, and mass
flow inlet boundaries are not available with the Eulerian model.
Table 23.9.3: PhaseSpecific and Mixture Conditions for the Eulerian Model
(for Laminar Flow)
Type Primary Phase Secondary Phase Mixture
exhaust fan;
outlet vent;
pressure outlet
(tot. temperature) volume fraction;
gran. temperature
(tot. temperature)
all others
inlet vent;
intake fan;
pressure inlet
coord. system;
flow direction
(tot. temperature)
coord. system;
flow direction;
volume fraction;
gran. temperature
(tot. temperature)
dir. spec.
method;
all others
velocity inlet velocity
(tot. temperature)
velocity;
volume fraction;
gran. temperature
(tot. temperature)
all others
axis;
outflow;
periodic;
solid;
symmetry
nothing nothing all others
wall shear condition shear condition all others
pressure
farfield;
fan;
porous jump;
radiator;
mass flow inlet
not available not available not available
fluid all source terms;
all fixed values;
other porous inputs
all source terms;
all fixed values;
other porous inputs
porous zone;
porosity;
all others
Table 23.9.4: PhaseSpecific and Mixture Conditions for the Eulerian Model
(with the Mixture Turbulence Model)
Type Primary Phase Secondary Phase Mixture
exhaust fan;
outlet vent;
pressure outlet
(tot. temperature) volume fraction;
gran. temperature
(tot. temperature)
all others
inlet vent;
intake fan;
pressure inlet
coord. system;
flow direction
(tot. temperature)
coord. system;
flow direction;
volume fraction;
gran. temperature
(tot. temperature)
dir. spec.
method;
all others
velocity inlet velocity
(tot. temperature)
velocity;
volume fraction;
gran. temperature
(tot. temperature)
all others
axis;
outflow;
periodic;
solid;
symmetry
nothing nothing all others
wall shear condition shear condition all others
pressure
farfield;
fan;
porous jump;
radiator;
mass flow inlet
not available not available not available
fluid other source terms;
other fixed values;
other porous inputs
other source terms;
other fixed values;
other porous inputs
source terms for
turbulence;
fixed values for
turbulence;
porous zone;
porosity;
all others
Table 23.9.5: PhaseSpecific and Mixture Conditions for the Eulerian Model
(with the Dispersed Turbulence Model)
Type Primary Phase Secondary Phase Mixture
exhaust fan;
outlet vent;
pressure outlet
turb. parameters
(tot. temperature)
volume fraction;
gran. temperature
(tot. temperature)
all others
inlet vent;
intake fan;
pressure inlet
coord. system;
flow direction;
turb. parameters;
(tot. temperature)
coord. system;
flow direction;
volume fraction;
gran. temperature
(tot. temperature)
dir. spec.
method;
all others
velocity inlet velocity;
turb. parameters
(tot. temperature)
velocity;
volume fraction;
gran. temperature
(tot. temperature)
all others
axis;
outflow;
periodic;
solid;
symmetry
nothing nothing all others
wall shear condition shear condition all others
pressure
farfield;
fan;
porous jump;
radiator;
mass flow inlet
not available not available not available
fluid momentum, mass,
turb. sources;
momentum, mass,
turb. fixed values;
other porous inputs
momentum and mass
sources;
momentum and mass
fixed values;
other porous inputs
porous zone;
porosity;
all others
Table 23.9.6: PhaseSpecific and Mixture Conditions for the Eulerian Model
(with the PerPhase Turbulence Model)
Type Primary Phase Secondary Phase Mixture
exhaust fan;
outlet vent;
pressure outlet
turb. parameters
(tot. temperature)
volume fraction;
turb. parameters;
gran. temperature
(tot. temperature)
all others
inlet vent;
intake fan;
pressure inlet
coord. system;
flow direction;
turb. parameters
(tot. temperature)
coord. system;
flow direction;
volume fraction;
turb. parameters;
gran. temperature
(tot. temperature)
dir. spec.
method;
all others
velocity inlet velocity;
turb. parameters
(tot. temperature)
velocity;
volume fraction;
turb. parameters;
gran. temperature
(tot. temperature)
all others
axis;
outflow;
periodic;
solid;
symmetry
nothing nothing all others
wall shear condition shear condition all others
pressure
farfield;
fan;
porous jump;
radiator;
mass flow inlet
not available not available not available
fluid momentum, mass,
turb. sources;
momentum, mass,
turb. fixed values;
other porous inputs
momentum, mass,
turb. sources;
momentum, mass,
turb. fixed values;
other porous inputs
porous zone;
porosity;
all others
Steps for SettingBoundaryConditions
The steps you need to perform for each boundary are as follows:
1. Select the boundary in the Zone list in the Boundary Conditions panel.
2. Set the conditions for the mixture at this boundary, if necessary. (See above for
information about which conditions need to be set for the mixture.)
(a) In the Phase dropdown list, select mixture.
(b) If the current Type for this zone is correct, click Set... to open the
corresponding panel (e.g., the Pressure Inlet panel); otherwise, choose the
correct zone type in the Type list, confirm the change (when prompted), and
the corresponding panel will open automaticall y.
(c) In the corresponding panel for the zone type you have selected (e.g., the Pres
sure Inlet panel, shown in Figure 23.9.9), specify the mixture boundary condi
tions.
Figure 23.9.9: The Pressure Inlet Panel for a Mixture
Note that only those conditions that apply to all phases, as described above,
will appear in this panel.
i
For a VOF calculation, if you enabled the Wall Adhesion option in
the
Phase Interaction panel, you can specify the contact angle at the wall for
each pair of phases as a constant (as shown in Figure 23.9.10) or a UDF
(see the UDF manual for more information).
The contact angle (θ
w
in Figure 23.3.3) is the angle between the wall and the
tangent to the interface at the wall, measured inside the phase listed in the
left column under Wall Adhesion in the Momentum tab of the Wall panel. For
example, if you are setting the contact angle between the oil and air phases in
the Wall panel shown in Figure 23.9.10, θ
w
is measured inside the oil phase.
Figure 23.9.10: The Wall Panel for a Mixture in a VOF Calculation with
Wall Adhesion
The default value for all pairs is 90 degrees, which is equivalent to no wall
adhesion effects (i.e., the interface is normal to the adjacent wall). A contact
angle of 45
◦
, for example, corresponds to water creeping up the side of
a container, as is common with water in a glass.
(d) Click OK when you are done setting the mixture boundary conditions.
3. Set the conditions for each phase at this boundary, if necessary. (See above for
information about which conditions need to be set for the individual phases.)
(a) In the Phase dropdown list, select the phase (e.g., water).
i
Note that, when you select one of the individual phases (rather than
the mixture), only one type of zone appears in the Type list. It is not
possible
to assign phasespecific zone types at a given boundary; the zone type is
specified for the mixture, and it applies to all of the individual phases.
(b) Click Set... to open the panel for this phase’s conditions (e.g., the Pressure
Inlet panel, shown in Figure 23.9.11).
Figure 23.9.11: The Pressure Inlet Panel for a Phase
(c) Specify the conditions for the phase. Note that only those conditions that
apply to the individual phase, as described above, will appear in this panel.
(d) Click OK when you are done setting the phasespecific boundary conditions.
23.10Setting Up the VOFModel
Steps for Copying BoundaryConditions
The steps for copying boundary conditions for a multiphase flow are slightly different
from those described in Section 7.1.5: Copying Boundary Conditions for a singlephase
flow. The modified steps are listed below:
1. In the Boundary Conditions panel, click the Copy... button. This will open the Copy
BCs panel.
2. In the From Zone list, select the zone that has the conditions you want to copy.
3. In the To Zones list, select the zone or zones to which you want to copy the condi
tions.
4. In the Phase dropdown list, select the phase for which you want to copy the
conditions (either mixture or one of the individual phases).
i
Note that copying the boundary conditions for one phase does not
auto matically result in the boundary conditions for the other phases
and the
mixture being copied as well. You need to copy the conditions for each
phase on each boundary of interest.
5. Click Copy. FLUENT will set all of the selected phase’s (or mixture’s) boundary
conditions on the zones selected in the To Zones list to be the same as that phase’s
conditions on the zone selected in the From Zone list. (You cannot copy a subset
of the conditions, such as only the thermal conditions.)
See Section 7.1.5: Copying Boundary Conditions for additional information about copying
boundary conditions, including limitations.
23.10 SettingUp the VOFModel
23.10.1 Choosing a VOFFormulation
To specify the VOF formulation to be used, select the appropriate VOF Scheme under
VOF Parameters in the Multiphase Model panel.
The VOF formulations that are available in FLUENTare the Explicit and Implicit schemes.
ModelingMultiphaseFlows
ExplicitSchemes
• Timedependent with the explicit interpolation scheme: Since the donor
acceptor scheme is available only for quadrilateral and hexahedral meshes, it cannot
be used for a hybrid mesh containing twisted hexahedral cells. For such cases,
you should use the timedependent explicit scheme. This formulation can also
be used for other cases in which the geometric reconstruction scheme does not give
satisfactory results, or the flow calculation becomes unstable. Note that the
CICSAM scheme or the modified HRIC scheme can be computationally
inexpensive when compared to the geometric reconstruction scheme and improves
the robustness and stability of the calculations. The Volume Fraction
discretizations, Modified HRIC and CICSAM, are available in the Solution Controls
panel when the explicit VOF scheme is selected.
Note that FLUENTwill automatically turn on the unsteady formulation with first
order discretization for time in the Solver panel.
• Timedependent with the geometric reconstruction interpolation scheme: This
for mulation should be used whenever you are interested in the timeaccurate
transie nt behavior of the VOF solution.
To use this formulation, make sure Explicit is selected as the VOF Scheme in the
Multiphase panel, then select GeoReconstruct as the Volume Fraction Discretization
scheme in the Solution Controls panel.
• Timedependent with the donoracceptor interpolation scheme: This
formulation should be used instead of the timedependent formulation with the
geometric re construction scheme if your mesh contains highly twisted hexahedral
cells. For such cases, the donoracceptor scheme may provide more accurate
results.
The DonorAcceptor scheme is used when Explicit is selected as the VOF Scheme
in the Multiphase panel. Initially, this formulation is not available in the GUI. To
make it available, use the following text command:
solve −→ set
−→expert
You will be asked a series of questions, one of which is
Allow selection of all applicable discretization schemes? [no]
If your response is yes, then many more discretization schemes will be available
for your selection. You can now use this formulation by selecting DonorAcceptor
as the Volume Fraction Discretization in the Solution Controls panel.
• The CICSAM scheme gives interface sharpness of the same level as the
geometric reconstruction scheme and is particularly suitable for flows with high
viscosity ratios between the phases.
23.10Setting Up the VOFModel
To use this formulation, select Explicit as the VOF Scheme in the Multiphase panel,
then select CICSAMas the Volume Fraction Discretization in the Solution Controls
panel.
While the explicit timedependent formulation is less computationally expensive
than the geometric reconstruction scheme, the interface between phases will not be
as sharp as that predicted with the geometric reconstruction scheme. To reduce this
diffusivity, it is recommended that you use the secondorder discretization scheme
for the volume fraction equations. In addition, you may want to consider turning
the geometric reconstruction scheme back on after calculating a solution with the
implicit scheme, in order to obtain a sharper interface.
i
For the geometric reconstruction and donoracceptor schemes, if you
are using a conformal grid (i.e., if the grid node locations are identical at
the
boundaries where two subdomains meet), you must ensure that there are
no twosided (zerothickness) walls within the domain. If there are, you
will need to slit them, as described in Section 6.8.6: Slitting Face Zones.
i
The issues discussed above for the explicit timedependent
formulation also apply to the implicit steadystate and timedependent
formulations,
described below. You should take the precautions described above to im
prove the sharpness of the interface.
ImplicitSchemes
• Timedependent with the implicit interpolation scheme: This formulation can
be used if you are looking for a steadystate solution and you are not interested in
the intermediate transie nt flow behavior, but the final steadystate solution is
dependent on the initial flow conditions and/or you do not have a distinct
inflow boundary for each phase.
To use this formulation, select Implicit as the VOF Scheme, and enable an Unsteady
calculation in the Solver panel (opened with the Define/Models/Solver... menu
item).
• Steadystate with the implicit interpolation scheme: This formulation can be
used if you are looking for a steadystate solution, you are not interested in the
intermediate transie nt flow behavior, and the final steadystate solution is not
affected by the initial flow conditions and there is a distinct inflow boundary for
each phase. Note that the implicit modified HRIC scheme can be used as a robust
alternati ve to the explicit geometric reconstruction scheme.
To use this formulation, select Implicit as the VOF Scheme in the Multiphase panel,
then select Modified HRIC as the Volume Fraction Discretization in the Solution Con
trols panel.
ModelingMultiphaseFlows
Examples
To help you determine the best formulation to use for your problem, examples that use
different formulations are listed below:
• jet breakup
Use the explicit scheme (timede pendent with the geometric reconstruction scheme
or the donoracceptor) if problems occur with the geometric reconstruction scheme.
• shape of the liquid interface in a centrifuge
Use the timedependent with the implicit interpolation scheme.
• flow around a ship’s hull
Use the steadystate with the implicit interpolation scheme.
23.10.2 ModelingOpenChannelFlows
Using the VOF formulation, open channel flows can be modeled in FLUENT. To start
using the open channel flow boundary condition, perform the following:
1. Turn on gravity.
(a) Open the Operating Conditions panel.
Define −→Operating Conditions...
(b) Turn on Gravity and set the gravitational acceleration fields.
2. Enable the volume of fluid model.
(a) Open the Multiphase Model panel.
Define −→ Models
−→Multiphase...
(b) Under Model, turn on Volume of Fluid.
(c) Under VOF Scheme, select either Implicit, Explicit.
3. Under VOF Parameters, select Open Channel Flow.
23.10Setting Up the VOFModel
In order to set specific parameters for a particular boundary for open channel flows,
turn on the Open Channel Flow option in the corresponding boundary condition panel.
Table 23.10.1 summarizes the types of boundaries available to the open channel
flow boundary condition, and the additional parameters needed to model open channel
flow. For more information on setting boundary condition parameters, see Chapter 7:
Boundary Conditions .
Table 23.10.1: Open Channel Boundary Parameters for the VOF Model
Boundary Type Parameter
pressure inlet Inlet Group ID;
Secondary Phase for Inlet;
Flow Specification Method;
Free Surface Level, Bottom Level;
Velocity Magnitude
pressure outlet Outlet Group ID;
Pressure Specification Method;
Free Surface Level; Bottom Level
mass flow inlet Inlet Group ID;
Secondary Phase for Inlet;
Free Surface Level;
Bottom Level
outflow Flow Rate Weighting
DefiningInletGroups
Open channel systems involve the flowing fluid (the secondary phase) and the fluid above
it (the primary phase).
If both phases enter through the separate inlets (e.g., inletphase2 and inlet
phase1), these two inlets form an inlet group. This inlet group is recognized by the
parameter Inlet Group ID, which will be same for both the inlets that make up the inlet
group. On
the other hand, if both the phases enter through the same inlet (e.g., inlet
combined),
then the inlet itself represents the inlet group.
i
In threephase flows, only one secondary phase is allowed to pass
through one inlet group.
ModelingMultiphaseFlows
DefiningOutletGroups
Outletgroups can be defined in the same manner as the inlet groups.
i
In threephase flows, the outlet should represent the outlet group,
i.e., separate outlets for each phase are not recommended in threephase
flows.
Setting the InletGroup
For pressure inlets and mass flow inlets, the Inlet Group ID is used to identify the
different inlets that are part of the same inlet group. For instance, when both phases
enter through the same inlet (single face zone), then those phases are part of one inlet
group and you would set the Inlet Group ID to 1 for that inlet (or inlet group).
In the case where the same inlet group has separate inlets (different face zones) for each
phase, then the Inlet Group ID will be the same for each inlet of that group.
When specifying the inlet group, use the following guidelines:
• Since the Inlet Group ID is used to identify the inlets of the same inlet group,
general information such as Free Surface Level, Bottom Level, or the mass flow rate
for each phase should be the same for each inlet of the same inlet group.
• You should specify a different Inlet Group ID for each distinct inlet group.
For example, consider the case of two inlet groups for a particular problem. The
first inlet group consists of water and air entering through the same inlet (a single
face zone). In this case, you would specify an inlet group ID of 1 for that inlet (or
inlet group). The second inlet group consists of oil and air entering through the
same inlet group, but each uses a different inlet (oilinlet and airinlet) for
each phase. In this case, you would specify the same Inlet Group ID of 2 for both of
the inlets that belong to the inlet group.
Setting the OutletGroup
For pressure outlet boundaries, the Outlet Group ID is used to identify the different outlets
that are part of the same outlet group. For instance, when both phases enter through
the same outlet (single face zone), then those phases are part of one outlet group and
you would set the Outlet Group ID to 1 for that outlet (or outlet group).
In the case where the same outlet group has separate outlets (different face zones) for
each phase, then the Outlet Group ID will be the same for each outlet of that group.
23.10Setting Up the VOFModel
When specifying the outlet group, use the following guidelines:
• Since the Outlet Group ID is used to identify the outlets of the same outlet
group, general information such as Free Surface Level or Bottom Level should be
the same for each outlet of the same outlet group.
• You should specify a different Outlet Group ID for each distinct outlet group.
For example, consider the case of two outlet groups for a particular problem. The
first inlet group consists of water and air exiting from the same outlet (a single face
zone). In this case, you would specify an outlet number of 1 for that outlet (or
outlet group). The second outlet group consists of oil and air exiting through the
same outlet group, but each uses a different outlet (oiloutlet and airoutlet)
for each phase. In this case, you would specify the same Outlet Group ID of 2 for
both of the outlets that belong to the outlet group.
i
For threephase flows, when all the phases are leaving through the
same outlet, the outlet should consist only of a single face zone.
Determiningthe Free SurfaceLevel
For the appropriate boundary, you need to specify the Free Surface Level value. This pa
rameter is available for all relevant boundaries, including pressure outlet, mass flow inlet,
and pressure inlet. The Free Surface Level, is represented by y
local
in Equation 23.325.
y
local
= −(
−→
a · gˆ) (23.101)
where
−→
a is the position vector of any point on the free surface, and gˆ is the unit
vector in the direction of the force of gravity. Here we assume a horizontal free surface
that is normal to the direction of gravity.
We can simply calculate the free surface level in two steps:
1. Determine the absolute value of height from the free surface to the origin in the
direction of gravity.
2. Apply the correct sign based on whether the free surface level is above or below
the origin.
If the liquid’s free surface level lies above the origin, then the Free Surface Level is positive
(see Figure 23.10.1). Likewise, if the liquid’s free surface level lies below the origin, then
the Free Surface Level is negati ve.
ModelingMultiphaseFlows
Determiningthe BottomLevel
For the appropriate boundary, you need to specify the Bottom Level value. This
parameter is available for all relevant boundaries, including pressure outlet, mass flow
inlet, and pressure inlet. The Bottom Level, is represented by a relation similar to
Equation 23.325.
y
bottom
= −(
−→
b · gˆ) (23.102)
where
−→
b is the position vector of any point on the bottom of the channel, and gˆ
is the unit vector of gravity. Here we assume a horizontal free surface that is normal
to the direction of gravity.
We can simply calculate the bottom level in two steps:
1. Determine the absolute value of depth from the bottom level to the origin in the
direction of gravity.
2. Apply the correct sign based on whether the bottom level is above or below the
origin.
If the channel’s bottom lies above the origin, then the Bottom Level is positive (see
Figure 23.10.1). Likewise, if the channel’s bottom lies below the origin, then the Bottom
Level is negati ve.
Free Surface Level (positive)
g
Bottom Level (positive)
Origin
Reference Level
Figure 23.10.1: Determining the Free Surface Level and the Bottom Level
Specifyingthe TotalHeight
The total height, along with the velocity, is used as an option for describing the
flow. The total height is given as
V
2
y
tot
= y
local
+
2g
(23.103)
where V is the velocity magnitude and g is the gravity magnitude.
Determiningthe VelocityMagnitude
Pressure inlet boundaries require the Velocity Magnitude for calculating the dynamic
pressure at the boundary. This is to be specified as the magnitude of the upstream inlet
velocity in the flow.
Determiningthe SecondaryPhase for the Inlet
For pressure inlets and mass flow inlets, the Secondary Phase for Inlet field is significant
in cases of threephase flows.
i
Note that only one secondary phase is allowed to pass through one
inlet group.
Consider a problem involving a threephase flow consisting of air as the primary phase,
and oil and water as the secondary phases. Consider also that there are two inlet groups:
• water and air
• oil and air
For the former inlet group, you would choose water as the secondary phase. For the
latter inlet group, you would choose oil as the secondary phase.
Choosing the Pressure SpecificationMethod
For a pressure outlet boundary, the outlet pressure can be specified in one of two ways:
• by prescribing the local height (i.e., a hydrostatic pressure profile)
• by specifying the constant pressure
i
This option is not available in the case of threephase flows since the
pres sure on the boundary is taken from the neighboring cell.
Limitations
The following list summarizes some issues and limitations associated with the open chan
nel boundary condition.
• The conservation of the Bernoulli integral does not provide the conservation
of mass flow rate for the pressure boundary. In the case of a coarser mesh, there
can be a significant difference in mass flow rate from the actual mass flow rate.
For finer meshes, the mass flow rate comes closer to the actual value. So, for
problems having constant mass flow rate, the mass flow rate boundary condition
is a better option. The pressure boundary should be selected when steady and
nonoscillating drag is the main objective.
• Specifying the top boundary as the pressure outlet can sometimes lead to a
diver gent solution. This may be due to the corner singularity at the pressure
boundary in the air region or due to the inability to specify local flow direction
correctly if the air enters through the top locally.
• Only the heavier phase should be selected as the secondary phase.
• In the case of threephase flows, only one secondary phase is allowed to
enter through one inlet group (i.e., the mixed inflow of different secondary phases
is not allowed).
Recommendations for SettingUp an Open ChannelFlow Problem
The following list represents a list of recommendations for solving problems using the
open channel flow boundary condition:
• In the cases where the inlet group has a different inlet for each phase of fluid,
then the parameter values (such as Free Surface Level, Bottom Level, and Mass Flow
Rate) for each inlet should correspond to all other inlets that belong to the inlet
group.
• The solution begins with an estimated pressure profile at the outlet boundar y.
In general, you can start the solution by assuming that the level of liquid at the
outlet corresponds to the level of liquid at the inlet. The convergence and solution
time is very dependent on the initial conditions. When the flow is completely
subcritical (upstream and downstream), in marine applications for instance, the
above approach is recommended.
If the final conditions of the flow can be predicted by other means, the solution
time can be significantly reduced by using the proper boundary condition.
• The initialization procedure is very critical in the open channel analysis.
If you are interested in the final steady state solution, then perform the following
initialization procedure:
1. Initialize the domain by setting the volume fraction of the secondary phase to
0, and providing the inlet velocity.
2. Patch the domain using a volume fraction value of the secondary phase to 1,
up to the Free Surface Level specified at the inlet.
3. Patch the inlet velocity again in the full domain.
If the Free Surface Level values are different at the inlet and outlet, then pat ching
some regions with inlet Free Surface Level values and some regions with outlet Free
Surface Level values could be useful for some problems.
The same steps for initialization are also recommended for unsteady flows, but now
the initial conditions are dependent on the user.
• For the initial stability of the solution, a smaller time step is recommended.
You can increase the time step once the solution becomes more stable.
23.10.3 Definingthe Phases for the VOFModel
Instructions for specifying the necessary information for the primary and secondary
phases and their interaction in a VOF calculation are provided below.
i
In general, you can specify the primary and secondary phases
whichever way you prefer. It is a good idea, especially in more complicated
problems,
to consider how your choice will affect the ease of problem setup. For
example, if you are planning to patch an initial volume fraction of 1 for
one phase in a portion of the domain, it may be more convenient to make
that phase a secondary phase. Also, if one of the phases is a compressible
ideal gas, it is recommended that you specify it as the primary phase to
improve solution stabili ty.
i
Recall that only one of the phases can be a compressible ideal gas. Be
sure that you do not select a compressible ideal gas material (i.e., a
material
that uses the compressible ideal gas law for density) for more than one of
the phases. See Sections 23.10.5 and 23.11.3 for details.
Definingthe PrimaryPhase
To define the primary phase in a VOF calculation, perform the following steps:
1. Select phase1 in the Phase list.
2. Click Set... to open the Primary Phase panel (Figure 23.10.2).
Figure 23.10.2: The Primary Phase Panel
3. In the Primary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
5. Define the material properties for the Phase Material.
(a) Click Edit..., and the Material panel will open.
(b) In the Material panel, check the properties, and modify them if necessary. (See
Chapter 8: Physical Properties for general information about setting material
properties, Section 23.10.5: Modeling Compressible Flows for specific informa
tion related to compressible VOF calculations, and Section 23.10.6: Modeling
Solidification/Melting for specific information related to melting/solidification
VOF calculations.)
i
If you make changes to the properties, remember to click Change
before closing the Material panel.
6. Click OK in the Primary Phase panel.
Defininga SecondaryPhase
To define a secondary phase in a VOF calculation, perform the following steps:
1. Select the phase (e.g., phase2) in the Phase list.
2. Click Set... to open the Secondary Phase panel (Figure 23.10.3).
Figure 23.10.3: The Secondary Phase Panel for the VOF Model
3. In the Secondary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
5. Define the material properties for the Phase Material, following the procedure out
lined above for setting the material properties for the primary phase.
6. Click OK in the Secondary Phase panel.
IncludingSurfaceTension and Wall AdhesionEffects
As discussed in Section 23.3.8: When Surface Tension Effects Are Important, the impor
tance of surface tension effects depends on the value of the capillary number, Ca (defined
by Equation 23.316), or the Weber number, We (defined by Equation 23.317). Surface
tension effects can be neglected if Ca 1 or We 1.
i
Note that the calculation of surface tension effects will be more
accurate if you use a quadrilateral or hexahedral mesh in the area(s) of the
compu
tational domain where surface tension is significant. If you cannot use a
quadrilateral or hexahedral mesh for the entire domain, then you should
use a hybrid mesh, with quadrilaterals or hexahedra in the affected areas.
FLUENT also offers an option to use VOF gradients at the nodes for cur
vature calculations on meshes when more accuracy is desired. For more
information, see Section 23.3.8: Surface Tension and Wall Adhesion.
If you want to include the effects of surface tension along the interface between one or
more pairs of phases, as described in Section 23.3.8: Surface Tension and Wall Adhesion,
click Interaction... to open the Phase Interaction panel (Figure 23.10.4).
Figure 23.10.4: The Phase Interaction Panel for the VOF Model (Surface Ten
sion Tab)
Perform the following steps to include surface tension (and, if appropriate, wall adhesion)
effects along the interface between one or more pairs of phases:
1. Click the Surface Tension tab.
2. If you want to include wall adhesion, turn on the Wall Adhesion option. When
Wall Adhesion is enabled, you will need to specify the contact angle at each wall
as a boundary condition (as described in Section 23.9.8: Defining Multiphase
Boundary Conditions ).
3. For each pair of phases between which you want to include the effects of surface
tension, specify a constant surface tension coefficient. Alternati vely you can spec
ify a temperature dependent, polynomial, piecewise polynomial, piecewise linear,
or a userdefined surface tension coefficient. See Section 23.3.8: Surface Tension
and Wall Adhesion for more information on surface tension, and the separate UDF
Manual for more information on userdefined functions. All surface tension coeffi
cients are equal to 0 by default, representing no surface tension effects along the
interface between the two phases.
i
For calculations involving surface tension, it is recommended that you
also turn on the Implicit Body Force treatme nt for the Body Force
Formulation
in the Multiphase Model panel. This treatme nt improves solution conver
gence by accounting for the partial equilibrium of the pressure gradient and
surface tension forces in the momentum equations. See Section 23.9.4: In
cluding Body Forces for details.
23.10.4 SettingTimeDependent Parameters for the VOFModel
If you are using the timedependent VOF formulation in FLUENT, an explicit solution
for the volume fraction is obtained either once each time step or once each iteration,
depending upon your inputs to the model. You also have control over the time step used
for the volume fraction calculation.
To compute a timedependent VOF solution, you will need to enable the Unsteady option
in the Solver panel (and choose the appropriate Unsteady Formulation, as discussed in
Section 25.17.1: User Inputs for TimeDependent Problems). If you choose the
Explicit scheme, FLUENT will turn on the firstorder unsteady formulation for you
automaticall y, so you need not visit the Solver panel yourself.
Define −→ Models
−→Solver...
There are two inputs for the timedependent calculation for the VOF model:
• By default, FLUENT will solve the volume fraction equation(s) once for each
time step. This means that the convective flux coefficients appearing in the other
trans port equations will not be completely updated each iteration, since the
volume fraction fields will not change from iteration to iteration.
If you want FLUENT to solve the volume fraction equation(s) at every iteration
within a time step, use the text command:
define −→ models −→ multiphase
−→
and select vof as the model. When prompted to solve vof every iteration?,
enter yes. When FLUENT solves these equations every iteration, the convective
flux
coefficients in the other trans port equations will be updated based on the updated
volume fractions at each iteration. This choice is the less stable of the two, and
requires more computational effort per time step than the default choice.
i
If you are using sliding meshes, or dynamic meshes with layering
and/or remeshing, using the solve vof every iteration? option
will yield
more accurate results, although at a greater computational cost.
• When FLUENT performs a timedependent VOF calculation, the time step
used for the volume fraction calculation will not be the same as the time step
used for the rest of the trans port equations. FLUENT will refine the time step
for VOF automaticall y, based on your input for the maximum Courant Number
allowed near the free surface. The Courant number is a dimensionless number
that compares the time step in a calculation to the characteristic time of transit of
a fluid element across a control volume:
∆t
∆x
cell
/v
fluid
(23.104)
In the region near the fluid interface, FLUENT divides the volume of each cell by
the sum of the outgoing fluxes. The resulting time represents the time it would
take for the fluid to empty out of the cell. The smallest such time is used as
the characteristic time of transit for a fluid element across a control volume, as
described above. Based upon this time and your input for the maximum allowed
Courant Number in the Multiphase Models panel, a time step is computed for use
in the VOF calculation. For example, if the maximum allowed Courant
number is 0.25 (the default), the time step will be chosen to be at most one
fourth the minimum transit time for any cell near the interface.
Note that these inputs are not required when the implicit scheme is used.
23.10.5 ModelingCompressibleFlows
If you are using the VOF model for a compressible flow, note the following:
• Only one of the phases can be defined as a compressible ideal gas (i.e., you
can select the ideal gas law for the density of only one phase’s material). There is
no limitation on using compressible liquids using userdefined functions.
• When using the VOF model, for stability reasons, it is better (although not
re quired) if the primary phase is a compressible ideal gas.
• If you specify the total pressure at a boundary (e.g., for a pressure inlet or
intake fan) the specified value for temperature at that boundary will be used
as total temperature for the compressible phase, and as static temperature for
the other phases (which are incompressible).
• For each mass flow inlet, you will need to specify mass flow or mass flux for
each individual phase.
i
Note that if you read a case file that was set up in a version of FLUENT
previous to 6.1, you will need to redefine the conditions at the mass flow
inlets. See Section 23.9.8: Defining Multiphase Boundary Conditions for
more information on defining conditions for a mass flow inlet in VOF mul
tiphase calculations.
See Section 9.6: Compressible Flows for more information about compressible flows.
23.10.6 ModelingSolidification/Melting
If you are including melting or solidification in your VOF calculation, note the following:
• It is possible to model melting or solidification in a single phase or in
multiple phases.
• For phases that are not melting or solidifying, you must set the latent heat
(L), liquidus temperature (T
liquidus
), and solidus temperature (T
solidus
) to zero.
See Chapter 24: Modeling Solidification and Melting for more information about melting
and solidification.
23.11 SettingUp the MixtureModel
23.11.1 Definingthe Phases for the MixtureModel
Instructions for specifying the necessary information for the primary and secondary
phases and their interaction for a mixture model calculation are provided below.
i
Recall that only one of the phases can be a compressible ideal gas. Be
sure that you do not select a compressible ideal gas material (i.e., a
material
that uses the compressible ideal gas law for density) for more than one of
the phases. See Section 23.11.3: Modeling Compressible Flows for details.
Definingthe PrimaryPhase
The procedure for defining the primary phase in a mixture model calculation is the same
as for a VOF calculation. See Section 23.10.3: Defining the Primary Phase for details.
Defininga NongranularSecondary Phase
To define a nongranular (i.e., liquid or vapor) secondary phase in a mixture multiphase
calculation, perform the following steps:
1. Select the phase (e.g., phase2) in the Phase list.
2. Click Set... to open the Secondary Phase panel (Figure 23.11.1).
23.11Setting Up the MixtureModel
Figure 23.11.1: The Secondary Phase Panel for the Mixture Model
3. In the Secondary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
5. Define the material properties for the Phase Material, following the same pro
cedure you used to set the material properties for the primary phase (see Sec
tion 23.10.3: Defining the Primary Phase). For a particulate phase (which must
be placed in the fluid materials category, as mentioned in Section 23.9: Steps for
Using a Multiphase Model), you need to specify only the density; you can ignore
the values for the other properties, since they will not be used.
6. In the Secondary Phase panel, specify the Diameter of the bubbles, droplets, or
particles of this phase (d
p
in Equation 23.412). You can specify a constant value,
or use a userdefined function. See the separate UDF Manual for details about
userdefined functions. Note that when you are using the mixture model without
slip velocity, this input is not necessary, and it will not be available to you.
7. Click OK in the Secondary Phase panel.
ModelingMultiphaseFlows
Defininga GranularSecondaryPhase
To define a granular (i.e., particulate) secondary phase in a mixture model multiphase
calculation, perform the following steps:
1. Select the phase (e.g., phase2) in the Phase list.
2. Click Set... to open the Secondary Phase panel (Figure 23.11.2).
Figure 23.11.2: The Secondary Phase Panel for a Granular Phase Using the
Mixture Model
3. In the Secondary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
23.11Setting Up the MixtureModel
5. Define the material properties for the Phase Material, following the same pro
cedure you used to set the material properties for the primary phase (see Sec
tion 23.10.3: Defining the Primary Phase). For a granular phase (which must be
placed in the fluid materials category, as mentioned in Section 23.9: Steps for Us
ing a Multiphase Model), you need to specify only the density; you can ignore the
values for the other properties, since they will not be used.
i
Note that all properties for granular flows can utilize userdefined functions
(UDFs).
See the separate UDF Manual for details about userdefined functions.
6. Turn on the Granular option.
7. In the Secondary Phase panel, specify the following properties of the particles of
this phase:
Diameter specifies the diameter of the particles. You can select constant in the
dropdown list and specify a constant value, or select userdefined to use a user
defined function. See the separate UDF Manual for details about userdefined
functions.
Granular Viscosity specifies the kinetic part of the granular viscosity of the par
ticles (µ
s,kin
in Equation 23.418). You can select constant (the default) in the
dropdown list and specify a constant value, select syamlalobrien to compute
the value using Equation 23.420, select gidaspow to compute the value using
Equation 23.421, or select userdefined to use a userdefined function.
Granular Temperature specifies temperature for the solids phase and is propor
tional to the kinetic energy of the random motion of the particles. Choose
either the algebraic, the constant, or userdefined option.
Solids Pressure specifies the pressure gradient term, ∇p
s
, in the granularphase
momentum equation. Choose either the lunetal, the syamlalobrien, the ma
ahmadi, or the userdefined option.
Radial Distribution specifies a correction factor that modifies the probability of
collisions between grains when the solid granular phase becomes dense. Choose
either the lunetal, the syamlalobrien, the maahmadi, the arastoopour, or a
userdefined option.
Elasticity Modulus is defined as
with G ≥
0.
G =
∂P
s
∂α
s
(23.111)
Choose either the derived or userdefined options.
Packing Limit specifies the maximum volume fraction for the granular phase
(α
s, max
). For monodispersed spheres, the packing limit is about 0.63, which is
the default value in FLUENT. In polydispersed cases, however, smaller spheres
can fill the small gaps between larger spheres, so you may need to increase the
maximum packing limit.
8. Click OK in the Secondary Phase panel.
DefiningDrag BetweenPhases
For mixture multiphase flows with slip velocity, you can specify the drag function to
be used in the calculation. The functions available here are a subset of those discussed
in Section 23.12.2: Defining the Phases for the Eulerian Model. See Section 23.4.5:
Relati ve (Slip) Velocity and the Drift Velocity for more information.
To specify drag laws, click Interaction... to open the Phase Interaction panel (Figure 23.11.3),
and then click the Drag tab.
Figure 23.11.3: The Phase Interaction Panel for the Mixture Model (Drag
Tab)
Definingthe Slip Velocity
If you are solving for slip velocities during the mixture calculation, and you want to
modify the slip velocity definition, click Interaction... to open the Phase Interaction panel
(Figure 23.11.4), and then click the Slip tab.
Figure 23.11.4: The Phase Interaction Panel for the Mixture Model (Slip Tab)
Under Slip Velocity, you can specify the slip velocity function for each secondary phase
with respect to the primary phase by choosing the appropriate item in the adjacent
dropdown list.
• Select maninnenetal (the default) to use the algebraic slip method of Manninen
et al. [229], described in Section 23.4.5: Relative (Slip) Velocity and the Drift
Velocity.
• Select none if the secondary phase has the same velocity as the primary phase
(i.e., no slip velocity).
• Select userdefined to use a userdefined function for the slip velocity.
See the
separate UDF Manual for details.
23.11.2 IncludingCavitationEffects
For mixture model calculations, it is possible to include the effects of cavitation, using
FLUENT’scavitation model described in Section 23.7.4: Cavitation Models. To enable the
cavitation model, turn on the Cavitation option in the Mass tab of the Phase Interaction
panel.
Figure 23.11.5: The Phase Interaction Panel for Mass Transfer with Cavita
tion Enabled
When you are using FLUENT’scavitation model, you will specify three parameters to be
used in the calculation of mass transfer due to cavitation. Under Cavitation Parameters
in the Phase Interaction panel, set the Vaporization Pressure (p
sat
in Equation 23.715),
the Surface Tension Coefficient , and the NonCondensableGas Mass Fraction . The default
value of p
sat
is 2540 Pa, the vaporization pressure for water at ambient temperature.
Note that p
sat
and the surface tension are properties of the liquid, depending mainly on
temperature. NonCondensable Gas Mass Fraction is the mass fraction of dissolved gases,
which depends on the purity of the liquid.
When multiple species are included in one or more secondary phases, or the heat transfer
due to phase change needs to be taken into account, the mass transfer mechanism must
be defined before turning on the Cavitation option. This is defined in the same way as
described Section 23.9.7: Including Mass Transfer Effects. It may be noted, however,
that for cavitation problems, at least two mass transfer mechanisms are defined:
• mass transfer from liquid to vapor.
• mass transfer from vapor to liquid.
In the Phase Interaction panel (Figure 23.11.6), you should choose the constantrate op
tion in the Mechanism dropdown list and leave the value as 0 for both mass transfer
mechanisms before turning on the Cavitation option.
Figure 23.11.6: The Phase Interaction Panel for Mass Transfer with Cavita
tion Disabled
23.11.3 ModelingCompressibleFlows
If you are using the mixture model for a compressible flow, note the following:
• Only one of the phases can be defined as a compressible ideal gas (i.e., you
can select the ideal gas law for the density of only one phase’s material). There is
no limitation on using compressible liquids using userdefined functions.
• If you specify the total pressure at a boundary (e.g., for a pressure inlet or
intake fan) the specified value for temperature at that boundary will be used
as total temperature for the compressible phase, and as static temperature for
the other phases (which are incompressible).
• For each mass flow inlet, you will need to specify mass flow or mass flux for
each individual phase.
i
Note that if you read a case file that was set up in a version of FLUENT
previous to 6.1, you will need to redefine the conditions at the mass flow
inlets. See Section 23.9.8: Defining Multiphase Boundary Conditions for
more information on defining conditions for a mass flow inlet in mixture
multiphase calculations.
See Section 9.6: Compressible Flows for more information about compressible flows.
23.12SettingUp the EulerianModel
23.12 SettingUp the EulerianModel
23.12.1 Additional Guidelines for EulerianMultiphaseSimulations
Once you have determined that the Eulerian multiphase model is appropriate for your
problem (as described in Section 23.2: Choosing a General Multiphase Model), you should
consider the computational effort required to solve your multiphase problem. The re
quired computational effort depends strongly on the number of trans port equations be
ing solved and the degree of coupling. For the Eulerian multiphase model, which has a
large number of highly coupled trans port equations, computational expense will be high.
Before setting up your problem, try to reduce the problem stateme nt to the simplest
form possible.
Instead of trying to solve your multiphase flow in all of its complexity on your first
solution attempt, you can start with simple approximations and work your way up to
the final form of the problem definition. Some suggestions for simplifying a multiphase
flow problem are listed below:
• Use a hexahedral or quadrilateral mesh (instead of a tetrahedral or
triangular mesh).
• Reduce the number of phases.
You may find that even a very simple approximation will provide you with useful infor
mation about your problem.
See Section 23.14.4: Eulerian Model for more solution strategies for Eulerian multiphase
calculations.
23.12.2 Definingthe Phases for the EulerianModel
Instructions for specifying the necessary information for the primary and secondary
phases and their interaction for an Eulerian multiphase calculation are provided below.
Definingthe PrimaryPhase
The procedure for defining the primary phase in an Eulerian multiphase calculation is
the same as for a VOF calculation. See Section 23.10.3: Defining the Primary Phase for
details.
ModelingMultiphaseFlows
Defininga NongranularSecondary Phase
To define a nongranular (i.e., liquid or vapor) secondary phase in an Eulerian multiphase
calculation, perform the following steps:
1. Select the phase (e.g., phase2) in the Phase list.
2. Click Set... to open the Secondary Phase panel (Figure 23.12.1).
Figure 23.12.1: The Secondary Phase Panel for a Nongranular Phase
3. In the Secondary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
5. Define the material properties for the Phase Material, following the same pro
cedure you used to set the material properties for the primary phase (see Sec
tion 23.10.3: Defining the Primary Phase).
6. In the Secondary Phase panel, specify the Diameter of the bubbles or droplets of
this phase. You can specify a constant value, or use a userdefined function. See
the separate UDF Manual for details about userdefined functions.
7. Click OK in the Secondary Phase panel.
23.12SettingUp the EulerianModel
Defininga GranularSecondaryPhase
To define a granular (i.e., particulate) secondary phase in an Eulerian multiphase calcu
lation, perform the following steps:
1. Select the phase (e.g., phase2) in the Phase list.
2. Click Set... to open the Secondary Phase panel (Figure 23.12.2).
Figure 23.12.2: The Secondary Phase Panel for a Granular Phase
3. In the Secondary Phase panel, enter a Name for the phase.
4. Specify which material the phase contains by choosing the appropriate material in
the Phase Material dropdown list.
ModelingMultiphaseFlows
5. Define the material properties for the Phase Material, following the same pro
cedure you used to set the material properties for the primary phase (see Sec
tion 23.10.3: Defining the Primary Phase). For a granular phase (which must be
placed in the fluid materials category, as mentioned in Section 23.9: Steps for Us
ing a Multiphase Model), you need to specify only the density; you can ignore the
values for the other properties, since they will not be used.
i
Note that all properties for granular flows can utilize userdefined functions
(UDFs).
See the separate UDF Manual for details about userdefined functions.
6. Turn on the Granular option.
7. (optional) Turn on the Packed Bed option if you want to freeze the velocity field
for the granular phase. Note that when you select the packed bed option for a
phase, you should also use the fixed velocity option with a value of zero for all
velocity components for all interior cell zones for that phase.
8. Specify the Granular Temperature Model. Choose either the default Phase Prop
erty option or the Partial Differential Equation option. See Section 23.5.8: Granular
Temperature for details.
9. In the Secondary Phase panel, specify the following properties of the particles of
this phase:
Diameter specifies the diameter of the particles. You can select constant in the
dropdown list and specify a constant value, or select userdefined to use a user
defined function. See the separate UDF Manual for details about userdefined
functions.
Granular Viscosity specifies the kinetic part of the granular viscosity of the par
ticles (µ
s,kin
in Equation 23.562). You can select constant (the default) in the
dropdown list and specify a constant value, select syamlalobrien to compute
the value using Equation 23.564, select gidaspow to compute the value using
Equation 23.565, or select userdefined to use a userdefined function.
Granular Bulk Viscosity specifies the solids bulk viscosity (λ
q
in Equation 23.56).
You can select constant (the default) in the dropdown list and specify a con
stant value, select lunetal to compute the value using Equation 23.566, or
select userdefined to use a userdefined function.
Frictional Viscosity specifies a shear viscosity based on the viscousplastic flow
(µ
s,fr
in Equation 23.562). By default, the frictional viscosity is neglected,
as indicated by the default selection of none in the dropdown list. If you
want to include the frictional viscosity, you can select constant and specify a
constant value, select schaeffer to compute the value using Equation 23.567,
select johnsonetal to compute the value using Equation 23.572, or select
userdefined to use a userdefined function.
23.12SettingUp the EulerianModel
Angle of Internal Friction specifies a constant value for the angle φ used in Scha
effer’s expression for frictional viscosity (Equation 23.567). This parameter
is relevant only if you have selected schaeffer or userdefined for the
Frictional Viscosity.
Frictional Pressure specifies the pressure gradient term, ∇P
f rictio n
, in the granular
phase momentum equation. Choose none to exclude frictional pressure from
your calculation, johnsonetal to apply Equation 23.572, syamlalobrien to ap
ply Equation 23.530, basedktgf, where the frictional pressure is defined by the
kinetic theory [79]. The solids pressure tends to a large value near the packing
limit, depending on the model selected for the radial distribution function.
You must hook a userdefined function when selecting the userdefined option.
See the separate UDF manual for information on hooking a UDF.
Frictional Modulus is defined as
G =
∂P
f riction
∂α
f riction
(23.121)
with G ≥ 0, which is the derived option. You can also specify a user
defined
function for the frictional modulus.
Friction Packing Limit specifies the maximum volume fraction for the granular
phase (α
s, max
). For monodispersed spheres, the packing limit is about 0.63,
which is the default value in FLUENT. In polydispersed cases, however, smaller
spheres can fill the small gaps between larger spheres, so you may need to
increase the maximum packing limit.
Granular Conductivity specifies the solids granular conductivity (k
Θ
s
in
Equation 23.575). You can select syamlalobrien to compute the value using
Equation 23.576, select gidaspow to compute the value using Equation 23.577,
or select userdefined to use a userdefined function. Note, however, that
FLU ENT currently uses an algebraic relation for the granular temperature.
This has been obtained by neglecting convection and diffusion in the
trans port equation, Equation 23.575 [363].
Granular Temperature specifies temperature for the solids phase and is propor
tional to the kinetic energy of the random motion of the particles. Choose
either the algebraic, the constant, or userdefined option.
Solids Pressure specifies the pressure gradient term, ∇p
s
, in the granularphase
momentum equation. Choose either the lunetal, the syamlalobrien, the ma
ahmadi, none, or a userdefined option.
Radial Distribution specifies a correction factor that modifies the probability of
collisions between grains when the solid granular phase becomes dense. Choose
either the lunetal, the syamlalobrien, the maahmadi, the arastoopour, or a
userdefined option.
Elasticity Modulus is defined as
with G ≥
0.
G =
∂P
s
∂α
s
(23.122)
Packing Limit specifies the maximum volume fraction for the granular phase
(α
s, max
). For monodispersed spheres, the packing limit is about 0.63, which is
the default value in FLUENT. In polydispersed cases, however, smaller spheres
can fill the small gaps between larger spheres, so you may need to increase the
maximum packing limit.
10. Click OK in the Secondary Phase panel.
Definingthe InteractionBetweenPhases
For both granular and nongranular flows, you will need to specify the drag function to
be used in the calculation of the momentum exchange coefficients. For granular
flows, you will also need to specify the restitution coefficient(s) for particle collisions. It
is also possible to include an optional lift force and/or virtual mass force (described
below) for both granular and nongranular flows.
To specify these parameters, click Interaction... to open the Phase Interaction panel (Fig
ure 23.12.3).
Figure 23.12.3: The Phase Interaction Panel for the Eulerian Model
Specifying the DragFunction
FLUENT allows you to specify a drag function for each pair of phases. Perform the
following steps:
1. Click the Drag tab to display the Drag Function inputs.
2. For each pair of phases, select the appropriate drag function from the corresponding
dropdown list.
• Select schillernaumann to use the fluidfluid drag function described by
Equa tion 23.518. The Schiller and Naumann model is the default method,
and it is acceptable for general use in all fluidfluid multiphase calculations.
• Select morsialexander to use the fluidfluid drag function described by
Equa tion 23.522. The Morsi and Alexander model is the most complete,
adjusting the function definition frequently over a large range of Reynolds
numbers, but calculations with this model may be less stable than with the
other models.
• Select symmetric to use the fluidfluid drag function described by
Equation 23.527. The symmetric model is recommended for flows in which
the secondary (dispersed) phase in one region of the domain becomes the
primary (continuous) phase in another. For example, if air is injected into the
bottom of a container filled halfway with water, the air is the dispersed phase
in the bottom half of the container; in the top half of the container, the air is
the continuous phase.
• Select wenyu to use the fluidsolid drag function described by Equation 23.5
39.
The Wen and Yu model is applicable for dilute phase flows, in which the
total secondary phase volume fraction is significantly lower than that of the
primary phase.
• Select gidaspow to use the fluidsolid drag function described by
Equation 23.541. The Gidaspow model is recommended for dense fluidized
beds.
• Select syamlalobrien to use the fluidsolid drag function described by
Equa tion 23.531. The SyamlalO’Brien model is recommended for use in
conjunc tion with the SyamlalO’Brien model for granular viscosity.
• Select syamlalobriensymmetric to use the solidsolid drag function
described by Equation 23.543. The symmetric SyamlalO’Brien model is
appropriate for a pair of solid phases.
• Select constant to specify a constant value for the drag function, and
then specify the value in the text field.
• Select userdefined to use a userdefined function for the drag function (see
the separate UDF Manual for details).
• If you want to temporarily ignore the interaction between two phases,
select
none.
Specifyingthe RestitutionCoefficients(Granular Flow Only)
For granular flows, you need to specify the coefficients of restitution for collisions
between particles (e
ls
in Equation 23.543 and e
ss
in Equation 23.544). In addition to
specifying the restitution coefficient for collisions between each pair of granular phases,
you will also specify the restitution coefficient for collisions between particles of the
same phase.
Perform the following steps:
1. Click the Collisionstab to display the Restitution Coefficient inputs.
2. For each pair of phases, specify a constant restitution coefficient. All restitution
coefficients are equal to 0.9 by default.
Includingthe LiftForce
For both granular and nongranular flows, it is possible to include the effect of lift forces
(F
~
lift
in Equation 23.58) on the secondary phase particles, droplets, or bubbles.
These lift forces act on a particle, droplet, or bubble mainly due to velocity gradients
in the primaryphase flow field. In most cases, the lift force is insignificant compared
to the drag force, so there is no reason to include it. If the lift force is significant (e.g.,
if the phases separate quickly), you may want to include this effect.
i
Note that the lift force will be more significant for larger particles, but the
FLUENTmodel assumes that the particle diameter is much smaller than the
interparticle spacing. Thus, the inclusion of lift forces is not appropriate
for closely packed particles or for very small particles.
To include the effect of lift forces, perform the following steps:
1. Click the Lift tab to display the Lift Coefficient inputs.
2. For each pair of phases, select the appropriate specification method from the cor
responding dropdown list. Note that, since the lift forces for a particle, droplet, or
bubble are due mainly to velocity gradients in the primaryphase flow field, you will
not specify lift coefficients for pairs consisting of two secondary phases; lift coeffi
cients are specified only for pairs consisting of a secondary phase and the primary
phase.
• Select none (the default) to ignore the effect of lift forces.
• Select constant to specify a constant lift coefficient, and then specify the
value in the text field.
• Select userdefined to use a userdefined function for the lift coefficient (see
the separate UDF Manual for details).
Includingthe VirtualMassForce
For both granular and nongranular flows, it is possible to include the “virtual mass force”
(F
~
vm
in Equation 23.59) that is present when a secondary phase accelerates relative to
the primary phase. The virtual mass effect is significant when the secondary phase
densi ty is much smaller than the primary phase density (e.g., for a transie nt bubble
column).
To include the effect of the virtual mass force, turn on the Virtual Mass option in the
Phase Interaction panel. The virtual mass effect will be included for all secondary phases;
it is not possible to enable it just for a particular phase.
23.12.3 ModelingTurbulence
If you are using the Eulerian model to solve a turbule nt flow, you will need to choose
one of turbulence models described in Section 23.5.10: Turbulence Models in the Viscous
Model panel (Figure 23.12.4).
Figure 23.12.4: The Viscous Model Panel for an Eulerian Multiphase Calcu
lation
The procedure is as follows:
1. Select either kepsilon or Reynolds Stress under Model.
2. Select the desired kepsilon Model or RSM Multiphase Model and any other related
parameters, as described for singlephase calculations in Section 12.12: Steps in
Using a Turbulence Model.
3. Under kepsilon Multiphase Model or RSM Multiphase Model, indicate the desired
multiphase turbulence model (see Section 23.5.10: Turbulence Models for details
about each):
• Select Mixture to use the mixture turbulence model. This is the default model.
• Select Dispersed to use the dispersed turbulence model. This model is
ap plicable when there is clearly one primary continuous phase and the rest
are dispersed dilute secondary phases.
• Select Per Phase to use a k turbulence model for each phase. This model
is appropriate when the turbulence transfer among the phases plays a
dominant role.
IncludingSource Terms
By default, the interphase momentum, k, and sources are not included in the calcula
tion. If you want to include any of these source terms, you can enable them using the
multiphaseoptions command in the define/models/viscous/multiphase
turbulence/
text menu. Note that the inclusion of these terms can slow down convergence noticeabl y.
If you are looking for additional accuracy, you may want to compute a solution first
without these sources, and then continue the calculation with these terms included. In
most cases these terms can be neglected.
Customizing the k MultiphaseTurbulent Viscosity
If you are using the k multiphase turbulence model, a userdefined function can be
used to customize the turbule nt viscosity for each phase. This option will enable you to
modify µ
t
in the k model. For more information, see the separate UDF Manual.
In the Viscous Model panel, under UserDefined Functions, select the appropriate user
defined function in the Turbulent Viscosity dropdown list.
23.12.4 IncludingHeatTransferEffects
To define heat transfer in a multiphase Eulerian simulation, you will need to visit the
Phase Interaction panel, after you have turned on the energy equation in the Energy panel.
Define −→Phases...
1. Click the Interaction... button to open the Phase Interaction panel (e.g., Fig
ure 23.12.5).
Figure 23.12.5: The Phase Interaction Panel for Heat Transfer
2. Click on the Heat tab in the Phase Interaction panel.
3. Select the desired correlation for the Heat Transfer Coefficient. Note the following
regarding the available choices:
gunn is frequently used for Eulerian multiphase simulations involving a granular
phase.
ranzmarshall is frequently used for Eulerian multiphase simulations not involv
ing a granular phase.
none allows you to ignore the effects of heat transfer between the two phases
userdefined allows you to implement a correlation reflecting a model of your
choice, through a userdefined function.
4. Set the appropriate thermal boundary conditions. You will specify the thermal
boundary conditions for each individual phase on most boundaries, and for the
mixture on some boundaries. See Chapter 7: Boundary Conditions for more in
formation on boundary conditions, and Section 23.9.8: Eulerian Model for more
information on specifying boundary conditions for a Eulerian multiphase calcula
tion.
See Section 23.5.9: Description of Heat Transfer for more information on heat transfer in
the framework of a Eulerian multiphase simulation.
23.12.5 Modeling Compressible
Flows
You can model compressible multiphase flows, and can use it in conjunction with the
energy multiphase equations and available multiphase turbulence models. When using
the Eulerian multiphase model for a compressible flow, note the following:
• While you can specify both compressible gas phases and compressible liquid
phases, you can only define one of the phases as a compressible ideal gas (i.e., you
can select the idealgas for the density in the Materials panel of only one phase’s
material). There is no limitation on using compressible liquids using userdefined
functions.
• You can define only one compressible fluid phase.
• For each mass flow inlet, you will need to specify mass flow or mass flux for
each individual phase.
• If you specify the total pressure at a boundary (e.g., for a pressure inlet or
intake fan), FLUENT will use the specified value for temperature at that
boundary as total temperature for the compressible phase, and as static
temperature for the other phases (which are incompressible).
i
Note that if you read a case file that was set up in a version of FLUENT
previous to 6.1, you will need to redefine the conditions at the mass flow
inlets. See Section 23.9.8: Defining Multiphase Boundary Conditions for
more information on defining conditions for a mass flow inlet in Eulerian
multiphase calculations.
See Section 9.6: Compressible Flows for more information about compressible flows.
23.13 SettingUp the Wet SteamModel
Once you have enabled either of the densitybased solvers in FLUENT,you can activate the
wet steam model (see Section 23.6: Wet Steam Model Theory) by opening the Multiphase
panel and selecting the Wet Steam option.
Define −→ Models
−→Multiphase...
Figure 23.13.1: The Multiphase Model Panel with the Wet Steam Model Ac
tivated
This section includes information about using your own property functions and data with
the wet steam model. Solution settings and strategies for the wet steam model can be
found in Section 23.14.5: Wet Steam Model. Postprocessing variables are described in
Section 23.15.1: ModelSpecific Variables.
23.13.1 Using UserDefinedThermodynamicWet SteamProperties
FLUENT allows you to use your own property functions and data with the wet steam
model. This is achieved with userdefined wet steam property functions (UDWSPF).
These userdefined functions are written in the C programming language and there is a
certain programming format that must be used so that you can build a successful library
that can be loaded into the FLUENTcode.
23.13SettingUp the Wet SteamModel
The following is the procedure for using the userdefined wet steam property functions
(UDWSPF):
1. Define the wet steam equation of state and all related thermodynamic and trans port
property equations.
2. Create a C source code file that conforms to the format defined in this section.
3. Start FLUENTand set up your case file in the usual way.
4. Turn on the wet steam model.
5. Compile your UDWSPF C functions and build a shared library file using the text
user interface.
define −→ models −→ multiphase −→
wetsteam −→compileuserdefinedwe tsteamfunctions
6. Load your newly created UDWSPF library using the text user interface.
define −→ models −→ multiphase −→
wetsteam −→loadunloaduserdefine dwetsteamlibrary
7. Run your calculation.
i
Note that the UDWSPF can only be used when the wet steam model
is acti vated. Therefore, the UDWSPF are available for use with the
densi ty
based solver only.
23.13.2 Writingthe UserDefinedWet SteamProperty Functions(UDWSPF)
Creating a UDWSPF C function library is reasonably straightforward:
• The code must contain the udf.h file inclusion directive at the beginning of
the source code. This allows the definitions for DEFINE macros and other
FLUENT
functions to be accessible during the compilation process.
• The code must include at least one of the UDF’s DEFINE functions
(i.e. DEFINE ON DEMAND) to be able to use the compiled UDFs utili ty.
• Any values that are passed to the solver by the UDWSPF or returned by the
solver to the UDWSPF are assumed to be in SI units.
• You must use the principle set of userdefined wet steam property functions
in your UDWSPF library, as described in the list that follows. These functions
are the mechanism by which your thermodynamic property data is transferred to
the FLUENTsolver.
ModelingMultiphaseFlows
The following lists the userdefined wet steam property function names and arguments,
as well as a short description of their functions. Function inputs from the FLUENTsolver
consist of one or more of the following variables: T = temperature (K ), P =
pressure (P a), and ρ = vaporphase density (kg/m
3
).
• void wetst init(Domain *domain)
This will be called when you load the UDWSPF. You use it to initialize wet steam
model constants or your own model consta nts. It returns nothing.
• real wetst satP(real T)
This is the saturated pressure function, which takes on temperature in K and
returns saturation pressure in Pa.
• real wetst satT(real P, real T)
This is the saturated temperature function, which takes on pressure in Pa and a
starting guess temperature in K and returns saturation temperature in K.
• real wetst eosP(real rho, real T)
This is the equation of state, which takes on vapor density in kg/m
3
and Temper
ature in K and returns pressure in Pa.
• real wetst eosRHO(real P, real T)
This is the equation of state, which takes on pressure in Pa and temperature in K
and returns vapor density in kg/m
3
.
• real wetst cpv(real T, real rho)
This is the vapor specific heat at constant pressure, which takes on temperature
in K and vapor density in kg/m
3
and returns specific heat at constant pressure in
J/kg/K.
• real wetst cvv(real T, real rho)
This is the vapor specific heat at constant volume, which takes on temperature
in K and vapor density in kg/m
3
and returns specific heat at constant volume in
J/kg/K.
• real wetst hv(real T,real rho)
This is the vapor specific enthalpy, which takes on temperature in K and vapor
density in kg/m
3
and returns specific enthalpy in J/Kg.
• real wetst sv(real T, real rho)
This is the vapor specific entropy, which takes on temperature in K and vapor
density in kg/m
3
and returns specific entropy in J/Kg/K.
23.13SettingUp the Wet SteamModel
• real wetst muv(real T, real rho)
This is the vapor dynamic viscosity, which takes on temperature in K and
vapor density in kg/m
3
and returns viscosity in kg/m/s.
• real wetst ktv(real T, real rho)
This is the vapor thermal conductivi ty, which takes on temperature in K and
vapor density in kg/m
3
and returns thermal conductivity in W/m/K.
• real wetst rhol(real T)
This is the saturated liquid density, which takes on temperature in K and returns
liquid density in kg/m
3
.
• real wetst cpl(real T)
This is the saturated liquid specific heat at constant pressure, which takes on tem
perature in K and returns liquid specific heat in J/kg/K.
• real wetst mul(real T)
This is the liquid dynamic viscosity, which takes on Temperature in K and returns
dynamic viscosity in kg/m/s.
• real wetst ktl(real T)
This is the liquid thermal conductivi ty, which takes on temperature in K and returns
thermal conductivity in W/m/K.
• real wetst surft(real T)
This is the liquid surface tension, which takes on Temperature in K and returns
surface tension N/m.
ModelingMultiphaseFlows
At the end of the code you must define a structure of type WS Functions whose
mem
bers are pointers to the principle functions listed previously. The structure is of type
WS Functions and its name is WetSteamFunctionLis t.
UDF_EXPORT WS_Functions WetSteamFunctionList =
{
wetst_init, /*initialization function*/
wetst_satP, /*Saturation pressure*/
wetst_satT, /*Saturation temperature*/
wetst_eosP, /*equation of state*/
wetst_eosRHO, /*equation of state*/
wetst_hv, /*vapor enthalpy*/
wetst_sv, /*vapor entropy*/
wetst_cpv, /*vapor isobaric specific heat*/
wetst_cvv, /*vapor isochoric specific heat*/
wetst_muv, /*vapor dynamic viscosity*/
wetst_ktv, /*vapor thermal conductivity*/
wetst_rhol, /*sat. liquid density*/
wetst_cpl, /*sat. liquid specific heat*/
wetst_mul, /*sat. liquid viscosity*/
wetst_ktl, /*sat. liquid thermal conductivity*/
wetst_surft /*liquid surface tension*/
};
23.13.3 CompilingYour UDWSPFand Building a Shared Library
File
This section presents the steps you will need to follow to compile your UDWSPF C code
and build a shared library file. This process requires the use of a C compiler. Most UNIX
operating systems provide a C compiler as a standard feature. If you are using a PC,
you will need to ensure that a C ++ compiler is installed before you can proceed (e.g.,
Microsoft Visual C ++, v6.0 or higher).
i
To use the UDWSPF you will need to first build the UDWSPF
library by compiling your UDWSPF C code and then loading the library
into the
FLUENTcode.
The UDWSPF shared library is built in the same way that the FLUENTexecutable itself
is built. Internally, a script called Makefile is used to invoke the system C compiler
to build an object code library that contains the native machine language translation of
your higherlevel C source code. This shared library is then loaded into FLUENT(either
at runtime or automatically when a case file is read) by a process called dynamic loading.
The object libraries are specific to the computer architecture being used, as well as to the
particular version of the FLUENT executable being run. The libraries must, therefore,
23.13SettingUp the Wet SteamModel
be rebuilt any time FLUENT is upgraded, when the computer’s operating system level
changes, or when the job is run on a different type of computer.
The general procedure for compiling UDWSPF C code is as follows:
• Place the UDWSPF C code in your working directory (i.e., where your case
file resides).
• Launch FLUENT.
• Read your case file into FLUENT.
• You can now compile your UDWSPF C code and build a shared library file
using the commands provided in the text command interface (TUI):
– Select the define/models/multiphase/wetsteam menu item
define −→ models −→ multiphase −→wet
steam
– Select the compileuserdefinedwetsteamfunctions option.
– Enter the compiled UDWSPF library name.
The name given here is the name of the directory where the shared library
(e.g., libudf) will reside. For example, if you hit <Enter> then a
directory should exist with the name libudf, and this directory will contain
library file called libudf. If, however, you type a new library name such as
mywetsteam, then a directory called mywetsteam will be created and
it will contain the library libudf.
– Continue on with the procedure when prompted.
– Enter the C source file names.
i
Ideally you should place all of your functions into a single file.
However, you can split them into separate files if desired.
– Enter the header file names, if applicable. If you do not have an extra header
file, then press <Enter> when prompted.
FLUENT will then start compiling the UDWSPF C code and put it in the
appropriate architecture director y.
ModelingMultiphaseFlows
23.13.4 Loadingthe UDWSPFShared LibraryFile
To load the UDWSPF library, perform the following steps:
• Go to the define/models/multiphase/wetsteam menu item in the text
user interface.
define −→ models −→ multiphase −→wetsteam
• Select the loadunloaduserdefinedwetsteamlibrary option and
follow the procedure when prompted.
If the loading of the UDWSPF library is successful, you will see a message similar
to the following:
Opening userdefined wet steam library "libudf"...
Library "libudf/lnx86/2d/libudf.so" opened
Setting material properties to WetSteam...
Initializing user defined material properties...
23.13SettingUp the Wet SteamModel
23.13.5 UDWSPFExample
This section describe a simple UDWSPF. You can use this example as a the basis for your
own UDWSPF code. For approximate calculations at low pressure, the simple idealgas
equation of state and constant isobaric specific heat is assumed and used. The properties
at the saturated liquid line and the saturated vapor line used in this example are similar
to the one used by FLUENT.
/********************** ************************ ************************/
/* User Defined Wet Steam
Properties: EOS : Ideal Gas Eq.
Vapor Sat. Line : W.C.Reynolds tables (1979)
Liquid Sat. Line: E. Eckert & R. Drake book (1972)
Use idealgas EOS with Steam properties
to model wet steam condensation in low pressure nozzle
Author: L. Zori
Date : Jan. 29 2004
*/
/********************** ************************ ************************/
#include "udf.h"
#include "stdio.h"
#include "ctype.h"
#include "stdarg.h"
/*Global Constants for this model*/
real ws_TPP = 338.150 ;
real ws_aaa = 0.01 ;
real cpg = 1882.0 ;/* Cpvapor at lowpressure region*/
DEFINE_ON_DEMAND(I_do_n othing)
{
/* This is a dummy function to allow us to use */
/* the Compiled UDFs utility */
}
ModelingMultiphaseFlows
void
wetst_init(Domain *domain)
{
/*
You must initialize these material property constants..
they will be used in the wet steam model in fluent
*/
ws_Tc = 647.286 ;/*Critical Temp. */
ws_Pc = 22089000.00 ;/*Critical Pressure */
mw_f = 18.016 ;/*fluid droplet molecular weight (water) */
Rgas_v = 461.50 ;/*vapor Gas Const*/
}
real
wetst_satP(real T)
{
real psat ;
real
SUM=0.0;
real pratio;
real F ;
real a1 = 7.41924200 ;
real a2 = 2.97210000E
01
;
real a3 =1.15528600E
01
;
real a4 = 8.68563500E
03
;
real a5 = 1.09409899E
03
;
real a6 =4.39993000E
03
;
real a7 = 2.52065800E
03
;
real a8 =5.21868400E
04
;
if (T > ws_Tc) T = ws_Tc ;
F = ws_aaa*(T  ws_TPP) ;
SUM = a1 + F*(a2+ F*(a3+ F*(a4+ F*(a5+ F*(a6+ F*(a7+ F*a8)))))) ;
pratio = (ws_Tc/T  1.0)*SUM;
psat = ws_Pc *exp(pratio) ;
return psat; /*Pa */
}
23.13SettingUp the Wet SteamModel
real
wetst_satT(real P, real T)
{
real tsat ;
real dT, dTA,dTM,dP,p1,p2,dPdT;
int i ;
for (i=0; i<25; ++i)
{
if (T > ws_Tc) T = ws_Tc0.5;
p1= wetst_satP(T) ;
p2= wetst_satP(T+0.1) ;
dPdT = (p2p1)/0.1 ;
dP = P  p1 ;
dT = dP/dPdT ;
dTA =
fabs(dT);
dTM = 0.1*T
;
if (dTA > dTM)
dT=dT*dTM/dTA ; T = T + dT;
if (fabs(dT) < TEMP_eps*T)
break;
}
tsat = T;
return tsat; /*K */
}
real
wetst_eosP(real rho, real T)
{
real P ;
P = rho* Rgas_v * T ;
return P; /*Pa */
}
ModelingMultiphaseFlows
real
wetst_eosRHO(real P, real T)
{
real rho ;
rho = P/(Rgas_v * T) ;
return rho; /*kg/m3 */
}
real
wetst_cpv(real T, real rho)
{
real cp;
cp = cpg ;
return cp; /* (J/Kg/K) */
}
real
wetst_cvv(real T, real rho)
{
real cv;
cv = wetst_cpv(T,rho)  Rgas_v ;
return cv; /* (J/Kg/K) */
}
real
wetst_hv(real T,real rho)
{
real h;
h = T* wetst_cpv(T,rho) ;
return h; /* (J/Kg) */
}
23.13SettingUp the Wet SteamModel
real
wetst_sv(real T, real rho)
{
real s ;
real TDatum=288.15;
real PDatum=1.01325e5;
s=wetst_cpv(T,rho)*log( T/TDatum)+
Rgas_v*log(PDatum/
(Rgas_v*T*rho));
return s; /* (J/Kg/K) */
}
real
wetst_muv(real T, real rho)
{
real muv ;
muv=1.7894e05 ;
return muv; /* (Kg/m/s) */
}
real
wetst_ktv(real T, real rho)
{
real ktv ;
ktv=0.0242 ;
return ktv; /* W/m/K */
}
ModelingMultiphaseFlows
real
wetst_rhol(real T)
{
real rhol;
real SUM = 0.0 ;
int ii ;
int i ;
real rhoc = 317.0 ;
real D[8] ;
D[0] = 3.6711257 ;
D[1] = 
2.8512396E+01 ; D[2]
=
2.2265240E+02
; D[3] = 
8.8243852E+02 ; D[4]
=
2.0002765E+03
; D[5] = 
2.6122557E+03 ; D[6]
=
1.8297674E+03
; D[7] = 
5.3350520E+02 ;
if (T > ws_Tc) T = ws_Tc ;
for(ii=0;ii < 8;++ii)
{
i = ii+1 ;
SUM += D[ii] * pow((1.0  T/ws_Tc), i/3.0) ;
}
rhol = rhoc*(1.0+SUM);
return rhol; /* (Kg/m3) */
}
real
wetst_cpl(real T)
{
real cpl;
real a1= 36571.6 ;
real a2= 555.217 ;
real a3= 2.96724 ;
real a4=
23.13SettingUp the Wet SteamModel
0.00778551; real a5=
1.00561e05; real
a6= 5.14336E
09;
ModelingMultiphaseFlows
if (T > ws_Tc) T = ws_Tc ;
cpl = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*a6)))) ;
return cpl; /* (J/Kg/K) */
}
real
wetst_mul(real T)
{
real mul ;
real a1= 0.530784;
real a2= 0.00729561;
real a3= 4.16604E05
; real a4= 1.26258E
07; real a5=
2.13969E10;
real a6= 1.92145E13;
real a7= 7.14092E17;
if (T > ws_Tc) T = ws_Tc ;
mul = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*(a6+ T*a7))))) ;
return mul; /* (Kg/m/s) */
}
real
wetst_ktl(real T)
{
real ktl ;
real a1= 1.17633;
real a2=
0.00791645; real a3=
1.48603E
05; real a4=
1.31689E07; real
a5= 2.47590E
10; real a6=
1.55638E13;
if (T > ws_Tc) T = ws_Tc ;
ktl = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*a6)))) ;
return ktl; /* W/m/K */
}
23.13SettingUp the Wet SteamModel
real
wetst_surft(real T)
{
real sigma ;
real Tr ;
real a1= 82.27 ;
real a2= 75.612 ;
real a3= 256.889 ;
real a4= 95.928 ;
if (T > ws_Tc) T = ws_Tc ;
Tr = T/ws_Tc ;
sigma = 0.001*(a1 + Tr*(a2+ Tr*(a3+
Tr*a4))) ;
return sigma ;/* N/m */
}
/* do not change the order of the function list */
UDF_EXPORT WS_Functions WetSteamFunctionList
=
{
wetst_init, /*initialization function*/
wetst_satP, /*Saturation pressure*/
wetst_satT, /*Saturation temperature*/
wetst_eosP, /*equation of state*/
wetst_eosRHO, /*equation of state*/
wetst_hv, /*vapor enthalpy*/
wetst_sv, /*vapor entropy*/
wetst_cpv, /*vapor isobaric specific heat*/
wetst_cvv, /*vapor isochoric specific heat*/
wetst_muv, /*vapor dynamic viscosity*/
wetst_ktv, /*vapor thermal conductivity*/
wetst_rhol, /*sat. liquid density*/
wetst_cpl, /*sat. liquid specific heat*/
wetst_mul, /*sat. liquid viscosity*/
wetst_ktl, /*sat. liquid thermal conductivity*/
wetst_surft /*liquid surface tension*/
};
/********************** ************************ ************************/
ModelingMultiphaseFlows
23.14 SolutionStrategiesfor MultiphaseModeling
23.14.1 Setting InitialVolumeFractions
Once you have initialized the flow (as described in Section 25.14: Initializing the Solution),
you can define the initial distribution of the phases. For a transie nt simulation, this dis
tribution will serve as the initial condition at t = 0; for a steadystate simulation, setting
an initial distribution can provide added stability in the early stages of the calculation.
You can patch an initial volume fraction for each secondary phase using the Patch panel.
Solve −→ Initialize
−→Patch...
If the region in which you want to patch the volume fraction is defined as a separate
cell zone, you can simply patch the value there. Otherwise, you can create a cell “reg
ister” that contains the appropriate cells and patch the value in the register. See Sec
tion 25.14.2: Patching Values in Selected Cells for details.
Solution strategies for the VOF, mixture, and Eulerian models are provided in Sec
tions 23.14.2, 23.14.3, and 23.14.4, respectively.
23.14.2 VOFModel
Several recommendations for improving the accuracy and convergence of the VOF solu
tion are presented here.
Setting the ReferencePressure Location
The site of the reference pressure can be moved to a location that will result in less round
off in the pressure calculation. By default, the reference pressure location is the center
of the cell at or closest to the point (0,0,0). You can move this location by specifying a
new Reference Pressure Location in the Operating Conditions panel.
Define −→Operating Conditions...
The position that you choose should be in a region that will always contain the least
dense of the fluids (e.g., the gas phase, if you have a gas phase and one or more liquid
phases). This is because variations in the static pressure are larger in a more dense
fluid than in a less dense fluid, given the same velocity distribution. If the zero of the
relative pressure field is in a region where the pressure variations are small, less roundoff
will occur than if the variations occur in a field of large nonzero values. Thus in systems
containing air and water, for example, it is important that the reference pressure location
be in the portion of the domain filled with air rather than that filled with water.
23.14SolutionStrategiesfor MultiphaseModeling
Pressure InterpolationScheme
For all VOF calculations, you should use the bodyforceweighted pressure interpolation
scheme or the PRESTO! scheme.
Solve −→ Controls
−→Solution...
DiscretizationScheme Selection for the Implicitand ExplicitFormulations
When the implicit scheme is used, the available options for Volume Fraction Discretization
are
• First Order Upwind
• Second Order upwind
• Modified HRIC
• QUICK
When the explicit scheme is used, the available options for Volume Fraction Discretization
are
• GeoReconstruct
• CICSAM
• Modified HRIC
• QUICK
When using the explicit scheme, First Order Upwind, Second Order upwind, and Donor
Acceptor can be made available under Volume Fraction Discretization by using the
following text command:
solve −→ set
−→expert
You will be asked a series of questions, one of which is
Allow selection of all applicable discretization schemes? [no]
to which you will respond yes.
i
You are encouraged to use the CICSAM scheme, as it gives a
sharper interface than the modified HRIC scheme.
ModelingMultiphaseFlows
i
In VOF modeling, using a highorder discretization scheme for the
momen tum trans port equations may reduce the stability of the solution
compared
to cases using firstorder discretization. In such situations, it is recom
mended to use a loworder variant of RhieChow face flux interpolation,
which can be turned on using the text command:
solve −→ set
−→numerics
When asked to disable high order RhieChow flux?[no], enter
yes.
Pressure Velocity Couplingand UnderRelaxationfor the Timedependent
Formulations
Another change that you should make to the solver settings is in the pressurevelocity
coupling scheme and underrelaxation factors that you use. The PISO scheme is recom
mended for transie nt calculations in general. Using PISO allows for increased values on
all underrelaxation factors, without a loss of solution stability. You can generally in
crease the underrelaxation factors for all variables to 1 and expect stability and a rapid
rate of convergence (in the form of few iterations required per time step). For
calculations on tetrahedral or triangular meshes, an underrelaxation factor of 0.7–0.8 for
pressure is recommended for improved stability with the PISO scheme.
Solve −→ Controls
−→Solution...
As with any FLUENT simulation, the underrelaxation factors will need to be decreased
if the solution exhibits unstable, divergent behavior with the underrelaxation factors set
to 1. Reducing the time step is another way to improve the stabili ty.
UnderRelaxationfor the SteadyStateFormulation
If you are using the steadystate implicit VOF scheme, the underrelaxation factors for
all variables should be set to values between 0.2 and 0.5 for improved stabili ty.
23.14.3 MixtureModel
Setting the UnderRelaxationFactorfor the Slip Velocity
You should begin the mixture calculation with a low underrelaxation factor for the slip
velocity. A value of 0.2 or less is recommended. If the solution shows good convergence
behavior, you can increase this value graduall y.
23.14SolutionStrategiesfor MultiphaseModeling
Calculatingan InitialSolution
For some cases (e.g., cyclone separation), you may be able to obtain a solution more
quickly if you compute an initial solution without solving the volume fraction and slip
velocity equations. Once you have set up the mixture model, you can temporarily disable
these equations and compute an initial solution.
Solve −→ Controls
−→Solution...
In the Solution Controls panel, deselect Volume Fraction and Slip Velocity in the Equations
list. You can then compute the initial flow field. Once a converged flow field is
obtained, turn the Volume Fraction and Slip Velocity equations back on again, and
compute the mixture solution.
23.14.4 EulerianModel
Calculatingan InitialSolution
To improve convergence behavior, you may want to compute an initial solution before
solving the complete Eulerian multiphase model. There are three methods you can use
to obtain an initial solution for an Eulerian multiphase calculation:
• Set up and solve the problem using the mixture model (with slip velocities)
instead of the Eulerian model. You can then enable the Eulerian model,
complete the setup, and continue the calculation using the mixturemodel solution
as a starting point.
• Set up the Eulerian multiphase calculation as usual, but compute the flow for
only the primary phase. To do this, deselect Volume Fraction in the Equations list
in the Solution Controls panel. Once you have obtained an initial solution for the
primary phase, turn the volume fraction equations back on and continue the
calculation for all phases.
• Use the mass flow inlet boundary condition to initialize the flow conditions. It
is recommended that you set the value of the volume fraction close to the value
of the volume fraction at the inlet.
i
You should not try to use a singlephase solution obtained without
the mixture or Eulerian model as a starting point for an Eulerian
multiphase
calculation. Doing so will not improve convergence, and may make it even
more difficult for the flow to converge.
ModelingMultiphaseFlows
Temporari ly IgnoringLiftand VirtualMass Forces
If you are planning to include the effects of lift and/or virtual mass forces in a steadystate
Eulerian multiphase simulation, you can often reduce stability problems that sometimes
occur in the early stages of the calculation by temporarily ignoring the action of the lift
and the virtual mass forces. Once the solution without these forces starts to converge,
you can interrupt the calculation, define these forces appropriatel y, and continue the
calculation.
Using WCycle Multigrid
For problems involving a packedbed granular phase with very small particle sizes (on
the order of 10 µm), convergence can be obtained by using the Wcycle multigrid for the
pressure. Under Fixed Cycle Parameters in the Multigrid Controls panel, you may need to
use higher values for PreSweeps, PostSweeps, and Max Cycles. When you are choosing
the values for these parameters, you should also increase the Verbosity to 1 in order to
monitor the AMG performance; i.e., to make sure that the pressure equation is solved to
a desired level of convergence within the AMG solver during each global iteration. See
Section 23.12.2: Defining the Phases for the Eulerian Model for more information about
granular phases, and Sections 25.6.2 and 25.22.3 for details about multigrid cycles.
23.14.5 Wet SteamModel
BoundaryConditions, Initialization, and Patching
When you use the wet steam model (described in Section 23.6: Wet Steam Model Theory
and Section 23.13: Setting Up the Wet Steam Model), the following two field variables
will show up in the inflow, outflow boundary panels, and in the Solution Initialization and
Patch panels.
• Liquid Mass Fraction (or the wetness factor)
In general, for dry steam entering flow boundaries the wetness factor is zero.
• Log10(Droplets Per Unit Volume)
In general this value is set to zero, indicating zero droplets entering the domain.
23.14SolutionStrategiesfor MultiphaseModeling
SolutionLimitsfor the Wet SteamModel
When you activate the wet steam model for the first time, a message is displayed indicat
ing that the Minimum Static Temperature should be adjusted to 273 K since the accuracy
of the builtin steam data is not guaranteed below a value of 273 K. If you use your own
steam property functions, you can adjust this limit to whatever is permissible for your
data.
To adjust the temperature limits, go to the Solution Limits panel.
Solve −→ Controls
−→Limits...
The default maximum wetness factor or liquid mass fraction (β) is set to 0.1. In general,
during the convergence process, it is common that this limit will be reached, but even
tually the wetness factor will drop below the value of 0.1. However, in cases where the
limit must be adjusted, you can do so using the text user interface.
define −→ models −→ multiphase −→ wetsteam −→
set −→
maxliquidmass
fraction
i
Note that the maximum wetness factor should not be set beyond 0.2
since the present model assumes a low wetness factor. When the wetness
factor is
greater than 0.1, the solution tends to be less stable due to the large source
terms in the trans port equations. Thus, the maximum wetness factor has
been set to a default value of 0.1, which corresponds to the fact that most
nozzle and turbine flows will have a wetness factor less that 0.1.
SolutionStrategiesfor the Wet SteamModel
If you face convergence difficulties while solving wet steam flow, try to initially lower
the CFL value and use firstorder discretization schemes for the solution. If you are still
unable to obtain a converged solution, then try the following solver settings:
1. Lower the underrelaxation factor for the wet steam equation below the current set
value. The underrelaxation factor can be found in the Solution Controls panel.
Solve −→ Controls −→Solution...
2. Solve for an initial solution with no condensation. Once you have obtained a proper
initial solution, turn on the condensation.
To turn condensation on or off, go to the Solution Controls panel.
Solve −→ Controls
−→Solution...
In the Solution Controls panel, deselect Wet Steam in the Equations list. When
doing so, you are preventing condensation from taking place while still computing
the flow based on steam properties. Once a converged flow field is obtained, turn
the Wet Steam equation back on again and compute the mixture solution.
ModelingMultiphaseFlows
23.15 Postprocessing for MultiphaseModeling
Each of the three general multiphase models provides a number of additional field func
tions that you can plot or report. You can also report flow rates for individual phases
for all three models, and display velocity vectors for the individual phases in a mixture
or Eulerian calculation.
Information about these postprocessing topics is provided in the following subsections:
• Section 23.15.1: ModelSpecific Variables
• Section 23.15.2: Displaying Velocity Vectors
• Section 23.15.3: Reporting Fluxes
• Section 23.15.4: Reporting Forces on Walls
• Section 23.15.5: Reporting Flow Rates
23.15.1 ModelSpecificVariables
When you use one of the general multiphase models, some additional field functions
will be available for postprocessing, as listed in this section. Most field functions that
are available in single phase calculations will be available for either the mixture or each
individual phase, as appropriate for the general multiphase model and specific options
that you are using. See Chapter 30: Field Function Definitions for a complete list of field
functions and their definitions. Chapters 28 and 29 explain how to generate graphics
displays and reports of data.
VOFModel
For VOF calculations you can generate graphical plots or alphanumeric reports of the
following additional item:
• Volume fraction (in the Phases...
category) This item is available for each
phase.
The variables that are not phase specific are available (e.g., variables in the Pressure... and
Velocity... categories) represent mixture quantities. Thermal quantities will be
available only for calculations that include the energy equation.
23.15 Postprocessing for MultiphaseModeling
MixtureModel
For calculations with the mixture model, you can generate graphical plots or alphanu
meric reports of the following additional items:
• Diameter (in the Properties... category)
This item is available only for secondary phases.
• Volume fraction (in the Phases...
category) This item is available for each
phase.
The variables that are not phase specific are available (e.g., variables in the Pressure...
category) represent mixture quantities. Thermal quantities will be available only for
calculations that include the energy equation.
EulerianModel
For Eulerian multiphase calculations you can generate graphical plots or alphanumeric
reports of the following additional items:
• Diameter (in the Properties... category)
This item is available only for secondary phases.
• Granular Conductivity (in the Properties...
category) This item is available only for granular
phases.
• Granular Pressure (in the Granular Pressure...
category) This item is available only for granular
phases.
• Granular Temperature (in the Granular Temperature...
category) This item is available only for granular phases.
• Volume fraction (in the Phases...
category) This item is available for each
phase.
The availability of turbulence quantities will depend on which multiphase turbulence
model you used in the calculation. Thermal quantities will be available (on a perphase
basis) only for calculations that include the energy equation.
ModelingMultiphaseFlows
MultiphaseSpecies Transport
For calculations using species trans port with either of the multiphase models, you can
generate graphical plots or alphanumeric reports of the following additional items:
• Mass Fraction of speciesn (in the Species...
category) This item is available for each species.
• Mole Fraction of speciesn (in the Species...
category) This item is available for each species.
• Molar Concentration of speciesn (in the Species...
category) This item is available for each species.
• Lam. Diff Coeff of speciesn (in the Species...
category) This item is available for each species.
• Eff. Diff. Coeff. of speciesn (in the Species...
category) This item is available for each species.
• Enthalpy of speciesn (in the Species...
category) This item is available for each
species.
• Relative Humidity (in the Species... category).
• Turbulent Rate of Reactionn (in the Reactions...
category) This item is available for each species.
• Rate of Reaction (in the Reactions... category).
• Mass Transfer Rate n (in the Phase Interaction... category)
This item is available for each mass transfer mechanism that you defined.
Thermal quantities will be available only for calculations that include the energy equation.
23.15 Postprocessing for MultiphaseModeling
Wet SteamModel
FLUENT provides a wide range of postprocessing information related to the wet steam
model.
The wet steam related items can be found in Wet Steam.... category of the variable
selection dropdown list that appears in the postprocessing panels.
• Liquid Mass Fraction
• Liquid Mass Generation Rate
• Log10(Droplets Per Unit Volume)
• Log10(Droplets Nucleation Rate)
• Steam Density (GasPhase)
• Liquid Density (LiquidPhase)
• Mixture Density
• Saturation Ratio
• Saturation Pressure
• Saturation Temperature
• Subcooled Vapor Temperature
• Droplet Surface Tension
• Droplet Critical Radius (microns)
• Droplet Average Radius (microns)
• Droplet Growth Rate (microns/s)
ModelingMultiphaseFlows
23.15.2 Displaying VelocityVectors
For mixture and Eulerian calculations, it is possible to display velocity vectors for the
individual phases using the Vectors panel.
Display
−→Vectors...
To display the velocity of a particular phase, select Velocity in the Vectors Of drop
down list, and then select the desired phase in the Phase dropdown list. You can also
choose Relative Velocity to display the phase velocity relative to a moving reference frame.
To display the mixture velocity ~v
m
(relevant for mixture model calculations only),
select Velocity (or Relative Velocity for the mixture velocity relative to a moving reference
frame), and mixture as the Phase. Note that you can color vectors by values of any
available variable, for any phase you defined. To do so, make the appropriate
selections in the Color by and following Phase dropdown lists.
23.15.3 Reporting Fluxes
When you use the Flux Reports panel to compute fluxes through boundaries, you will be
able to specify whether the report is for the mixture or for an individual phase.
Report
−→Fluxes...
Select mixture in the Phase dropdown list at the bottom of the panel to report fluxes for
the mixture, or select the name of a phase to report fluxes just for that phase.
23.15.4 Reporting Forces on Walls
For Eulerian calculations, when you use the Force Reports panel to compute forces or
moments on wall boundaries, you will be able to specify the individual phase for which
you want to compute the forces.
Report
−→Forces...
Select the name of the desired phase in the Phase dropdown list on the left side of the
panel.
23.15 Postprocessing for MultiphaseModeling
23.15.5 Reporting Flow Rates
You can obtain a report of mass flow rate for each phase (and the mixture) through each
flow boundary using the report/massflow text command:
report −→mass
flow
When you specify the phase of interest (the mixture or an individual phase), FLUENT
will list each zone, followed by the mass flow rate through that zone for the specified
phase. An example is shown below.
/report> mf
(mixture water air)
domain id/name [mixture] air
zone 10 (spiralpressoutlet): 1.2330244
zone 3 (pressureoutlet): 9.7560663
zone 11 (spiralvelinlet): 0.6150589
zone 8 (spiralwall): 0
zone 1 (walls): 0
zone 4 (velocityinlet): 4.9132133
net massflow: 5.4608185
M odelingM ultiphase Flows
23.1 In t oduction r
A large number of flows encountered in nature and technology are a mixture of phases. Physical phases of matter are gas, liquid, and solid, but the concept of phase in a multiphase flow system is applied in a broader sense. In multiphase flow, a phase can be defined as an identifiable class of material that has a particular inertial response to and interaction with the flow and the potential field in which it is immersed. For example, differentsized solid particles of the same material can be treated as different phases because each collection of particles with the same size will have a similar dynamical response to the flow field.
23.1.1 M ultiphase Flow Regimes
Multiphase flow regimes can be grouped into four categories: gasliquid or liquidliquid flows; gassolid flows; liquidsolid flows; and threephase flows.
GasLiquid LiquidLiquid or Flows
The following regimes are gasliquid or liquidliquid flows: • Bubbly flow: This is the flow of discrete gaseous or fluid bubbles in a continuous fluid. • Droplet flow: This is the flow of discrete fluid droplets in a continuous gas. • Slug flow: This is the flow of large bubbles in a continuous fluid. • Stratified/freesurface flow: This is the flow of immiscible fluids separated by a clearlydefined interface. See Figure 23.1.1 for illustrations of these regimes.
GasSolidFlows
The following regimes are gassolid flows: • Particleladen flow: This is flow of discrete particles in a continuous gas. • Pneumatic trans port: This is a flow pattern that depends on factors such as solid loading, Reynolds numbers, and particle properties. Typical patterns are dune flow, slug flow, packed beds, and homogeneous flow. • Fluidized bed: This consists of a vertical cylinder containing particles, into which a gas is introduced through a distributor. The gas rising through the bed suspends the particles. Depending on the gas flow rate, bubbles appear and rise through the bed, intensifying the mixing within the bed.
232
c Flue nt Inc. Septe mber 29, 2006
23.1 In t oduction r See Figure 23.1.1 for illustrations of these regimes.
LiquidSolid Flows
The following regimes are liquidsolid flows: • Slurry flow: This flow is the trans port of particles in liquids. The fundame ntal behavior of liquidsolid flows varies with the properties of the solid particles relati ve to those of the liquid. In slurry flows, the Stokes number (see Equation 23.24) is normally less than 1. When the Stokes number is larger than 1, the characteristic of the flow is liquidsolid fluidization. • Hydrotrans port: continuous liquid This describes denselydistributed solid particles in a
• Sedime ntation: This describes a tall column initially containing a uniform dispersed mixture of particles. At the bottom, the particles will slow down and form a sludge layer. At the top, a clear interface will appear, and in the middle a consta nt settling zone will exist. See Figure 23.1.1 for illustrations of these regimes.
ThreePhaseFlows
Threephase flows are combinations of the other flow regimes listed in the previous sections.
c Flue nt Inc. Septe mber 29, 2006
233
1. or particleladen flow stratified/f reesurface flow pneumatic transport. or slurry flow sedimentation fluidized bed Figure 23. 2006 .1: Multiphase Flow Regimes 234 c Flue nt Inc.M odelingM ultiphase Flows slug flow bubbl y. droplet. Septe mber 29. hyd rotransport.
atomizers. and dustladen environme ntal flows. droplets. • Droplet flow examples include absorbers.23. and the appropriate model for different amounts of coupling.2 Choosing a GeneralM ultiphase Model The first step in solving any multiphase problem is to determine which of the regimes provides some broad guidelines for determining appropriate models for each regime. • Stratified/freesurface flow examples include sloshing in offshore separator devices and boiling and condensation in nuclear reactors. and metal pow. c Flue nt Inc. dryers. evap. and scrubbers.1.2 Choosinga G eneralM ultiphase odel M 23. grains.chemical fluid systems • Sedime ntation examples include mineral processing.ders. air classifiers. 2006 235 . and scrubbers. cavitation. • Fluidized bed examples include fluidized bed reactors and circulating fluidized beds. dust collec. air lift pumps. and how to determine the degree of interphase coupling for flows involving bubbles.ing. combustors.2 Examplesof Multiphase Systems Specific examples of each regime described in Section 23. cryogenic pump.1. Septe mber 29.1: Multiphase Flow Regimes are listed below: • Bubbly flow examples include absorbers. 23. flotation.tors. • Pneumatic trans port examples include trans port of cement. evaporation. • Slug flow examples include large bubble motion in pipes or tanks. • Particleladen flow examples include cyclone separators. aeration.orators. or particles. • Slurry flow examples include slurry trans port and mineral processing • Hydrotrans port examples include mineral processing and biomedical and physio. gas cooling.
in the case of granular flows.4: Mixture Model Theory) is designed for two or more phases (fluid or particulate). bubbly flows. 236 c Flue nt Inc. and the steady or transie nt tracking of any liquidgas interface. Septe mber 29.M odelingM ultiphase Flows 23. The mixture model solves for the mixture momentum equation and prescribes relative velocities to describe the dispersed phases. and the volume fraction of each of the fluids in each computational cell is tracked throughout the domain. which have similar structure for all phases. the different phases are treated mathematically as interpenetrating continua. The EulerEulerApproach In the EulerEuler approach. or. T he VO F Model The VOF model (described in Section 23. freesurface flows. Conservation equations for each phase are derived to obtain a set of equations. Applications of the mixture model include particleladen flows with low loading. These volume fractions are assumed to be continuous functions of space and time and their sum is equal to one. These equations are closed by providing constituti ve relations that are obtained from empirical information. the motion of liquid after a dam break. and the Eulerian model.1. sedimentation.2.1: Overview) and the EulerEuler approach (discussed in the following section). T he M ixture Model The mixture model (described in Section 23. As in the Eulerian model.3: Volume of Fluid (VOF) Model Theory) is a surfacetracking technique applied to a fixed Eulerian mesh. Applications of the VOF model include stratified flows. filling. the phases are treated as interpenetrating continua. The mixture model can also be used without relative velocities for the dispersed phases to model homogeneous multiphase flow.1 Approaches to MultiphaseModeling Advances in computational fluid mechanics have provided the basis for further insight into the dynamics of multiphase flows. the concept of phasic volume fraction is introduced. 2006 . Currently there are two approaches for the numerical calculation of multiphase flows: the EulerLagrange approach (discussed in Section 22. sloshing. It is designed for two or more immiscible fluids where the position of the interface between the fluids is of interest. three different EulerEuler multiphase models are available: the volume of fluid (VOF) model. In the VOF model. and cyclone separators. by application of kinetic theor y. the prediction of jet breakup (surface tension). the mixture model. the motion of large bubbles in a liquid. Since the volume of a phase cannot be occupied by the other phases. In FLUENT. a single set of momentum equations is shared by the fluids.
lerian Model Theory for more information about the Eulerian model. 2006 237 .4 and 23. granular (fluidsolid) flows are handled differently than nongra nular (fluidfluid) flows. since the model you use will be valid for only part of the flow you are modeling. use the Eulerian model for granular flow. particle suspension. use the VOF model.3: Volume of Fluid (VOF) Model Theory for more information about the VOF model. once you have determined the flow regime that best represents your multiphase system. droplet.4: Mixture Model Theory) or the Eulerian model for granular flow (described in Section 23. • For stratified/freesurface flows. Septe mber 29. use either the mixture model (de. It solves a set of n momentum and continuity equations for each phase. • For slurry flows and hydrotrans port. • For fluidized beds. use the VOF model. respectively. complex multiphase flows that involve multiple flow regimes.3: Volume of Fluid (VOF) Model Theory for more information about the VOF model. Note that the accuracy of results will not be as good as for flows that involve just one flow regime. • For sedime ntation. in Sections 23.4: Mixture Model Theory) or the Eulerian model (described in Section 23.5). use the Eulerian model.2 M odelComparisons In general. • For pneumatic trans port. The manner in which this coupling is handled depends upon the type of phases involved. Coupling is achieved through the pressure and interphase exchange coefficients. For granular flows. See Section 23.23. • For general.2 Choosinga G eneralM ultiphase odel M The Eulerian Model The Eulerian model (described in Section 23. Momentum exchange between the phases is also dependent upon the type of mixture being modeled. FLUENT’s userdefined functions allow you to customize the calculation of the momentum exchange. See Section 23. you can select the appropriate model based on the following guidelines: • For bubbly. and particleladen flows in which the phases mix and/or dispersedphase volume fractions exceed 10%. c Flue nt Inc. • For slug flows. 23. See Section 23. risers. Applications of the Eulerian multiphase model include bubble columns.5: Eulerian Model Theory ).5: Eu. the properties are obtained from application of kinetic theory. and choose the model that is most appropriate for that aspect of the flow.5: Eulerian Model Theory) is the most complex of the multiphase models in FLUENT.2.5: Eulerian Model Theory ).5: Eulerian Model Theory for more information about the Eulerian model. select the aspect of the flow that is of most interest. and fluidized beds. use the mixture or Eulerian model (described.scribed in Section 23. use the mixture model for homogeneous flow (described in Section 23. See Section 23.
For more information on the dynamic mesh feature. see Section 11: Modeling Flows Using Sliding and Deforming Meshes.) To choose between the mixture model and the Eulerian model. then a userdefined function can be used. As a general guide. the mixture model may be preferable (i. Choosing a model for the other types of flows is less straig htforward. and the mixture and Eulerian models are appropriate for flows in which the phases mix or separate and/or dispersedphase volume fractions exceed 10%. as you can for a nongranular Eulerian simulation. and the Stokes number. since it solves a smaller number of equations than the Eulerian model. the mixture model may be a better option. then the SchillerNaumann law is more than adequate. as indicated in Section 23.. is straig htforward. that the complexity of the Eulerian model can make it less computationally stable than the mixture model. the VOF model is appropriate for stratified or freesurface flows. less computationally expensive). Septe mber 29. 2006 .e.. the Eulerian model is a better choice. • If you want to solve a simpler problem. (Note that the word “particle” is used in this discussion to refer to a particle. β. droplet. If accuracy is more importa nt than computational effort.) 238 c Flue nt Inc.2.M odelingM ultiphase Flows As discussed in this section. For more information about how other FLUENT models are compatible with FLUENT’smultiphase models. Keep in mind. which requires less computational effort. St. For most cases with spherical particles. the mixture model may be a better choice. (Flows in which the dispersedphase volume fractions are less than or equal to 10% can be modeled using the discrete phase model described in Chapter 22: Modeling Discrete Phase . or bubble. FLUENT’s multiphase models are compatible with FLUENT’s dynamic mesh modeling feature. however. DetailedGuidelines For stratified and slug flows. there are some parameters that help to identify the appropriate multiphase model for these other flows: the particulate loading.2: Model Comparisons. For cases with nonspherical particles. see Appendix A: FLUENT Model Compatibili ty. the Eulerian model can usually provide more accurate results than the mixture model. you should consider the following guidelines: • If there is a wide distribution of the dispersed phases (i. you should use the Eulerian model instead. If the dispersed phases are conce ntrated just in portions of the domain. if the interphase drag laws are unknown or their applicability to your system is questionable.e. • If interphase drag laws that are applicable to your system are available (either within FLUENT or through a userdefined function). Even though you can apply the same drag laws to the mixture model. the choice of the VOF model. if the particles vary in size and the largest particles do not separate from the primary flow field).
e.21) is greater than 1000 for gassolid flows. the coupling is twoway (i. but the particles in turn influence the carrier fluid via reduction in mean momentum and turbulence). Information about these parameters is importa nt for determining how the dispersed phase should be treated.. but the particles have no influence on the fluid carrier). The discrete phase (Chapter 22: Modeling Discrete Phase). the particle can therefore be treated as isolated (i. the degree of interaction between the phases can be divided into the following three categories: • For very low loading. the coupling between the phases is oneway (i. The discrete phase(Chapter 22: Modeling Discrete Phase) .. For example. The particulate loading is defined as the mass density ratio of the dispersed phase (d) to that of the carrier phase (c): β= The material density ratio γ= ρd ρc (23. the fluid carrier influences the particulate phase via drag and turbulence. the interparticle space dd is about 8. for a gasparticle flow with a particulate L loading of 1.22) αd ρd αc ρc (23. the fluid carrier influences the particles via drag and turbulence. An estimate of this distance has been given by Crowe et al. 2006 239 . Septe mber 29. Since the Eulerian model is the most expensive. [68]: L π1+ κ = dd 6 κ 1/3 (23. and Eulerian models are all applicable in this case.e.e. c Flue nt Inc. but you need to take into account other factors in order to decide which model is more appropriate. about 1 for liquidsolid flows.2 Choosinga G eneralM ultiphase odel M The Effectof ParticulateLoading Particulate loading has a major impact on phase interactions. See below for information about using the Stokes number as a guide. and less than 0. mixture. the discrete phase or mixture model is recommended.23. very low particulate loading). Using these parameters it is possible to estimate the average distance between the individual particles of the particulate phase. and Eulerian models can all handle this type of problem correctl y.. • For intermediate loading. mixture.23) β where κ = γ .001 for gasliquid flows. Depending on the particulate loading.
For St 1.04 for particles with a diameter of 30 microns. the Stokes number is 0. (The volume fractions are too high for the discrete phase model (Chapter 22: Modeling Discrete Phase).005 for particles with a diameter of 300 microns. For St ≈ 1. Only the Eulerian model will handle this type of problem correctl y. the discrete phase model is the only multiphase model that allows you to specify the particle distribution or include combustion modeling in your simulation. there is twoway coupling plus particle pressure and viscous stresses due to particles (fourway coupling). For the case of mineral processing. Also.0. or the most appropriate considering other factors. Examples For a coal classifier with a characteristic length of 1 m and a characteristic velocity of 10 m/s. For St > 1. In this case.0.• For high loading. but 4. The Stokes number can be defined as the relation between the particle response time and the system response time: St = where τd = ρd d2 d 18µ c τd ts (23. the Stokes number is 0. the particles will move independently of the flow and either the discrete phase model (Chapter 22: Modeling Discrete Phase ) or the Eulerian model is applicable. you can choose between the mixture and Eulerian models. you can therefore choose the least expensive (the mixture model. in a system with a characteristic length of 0. you can choose the least expensive or the most appropriate considering other factors.0.) O therConsiderations Keep in mind that the use of the discrete phase model (Chapter 22: Modeling Discrete Phase) is limited to low volume fractions. estimating the value of the Stokes number can help you select the most appropriate model. mixture. The Significanceof the Stokes Number For systems with intermediate particulate loading. in most cases). Clearly the mixture model will not be applicable to the latter case.2 m and a characteristic velocity of 2 m/s. or Eulerian) is applicable. as noted below.24) and ts is based on the characteristic length (Ls ) and the characteristic velocity (Vs ) of the system under investigation: ts =VsLs . the particle will follow the flow closely and any of the three models (discrete phase(Chapter 22: Modeling Discrete Phase) . again any of the three models is applicable.0 for particles with a diameter of 300 microns. .
As a fully implicit scheme. In order to accurately model multiphase flow.23. including mixture phase momentum equations.315) may be written as ∂(αρφ) + ∇ · (αρV~ φ) = ∇ · τ + Sφ ∂t (23. i The secondorder time scheme cannot be used with the VOF Explicit Schemes. The secondorder time scheme has been adapted to all the trans port equations.25) Where φ is either a mixture (for the mixture model) or a phase variable. ρ is the mixture phase density.2. the process can vary spatially as well as temporally. a general trans port equation (similar to that of Equation 25.3 TimeSchemes in M ultiphase Flow In many multiphase applications. turbulence models.27) P Anb n+1 + S p n+1 + n+1 1. phase volume fraction equations. τ is the diffusion term. the pressure correction equation. V~ is the mixture or phase velocity (depending on the equations). Equation 23.317). The general trans port equation.5(αp ρp V ∆t n+ 1 n−1 Sφ = S U ol) + 2(αp ρp φp V n −0. and with the VOF Implicit Scheme. and the granular flow model. this secondorder timeaccurate scheme achieves its accuracy by using an Euler backward approximation in time (see Equation 25.26 can be written in simpler form: Ap φp = where Ap = ol) X An bφn b + Sφ (23. In multiphase flow. species trans port equations. the secondorder time scheme is available in the Mixture and Eulerian multiphase models. In addition to the firstorder time scheme in FLUENT. both higherorder spatial and time discretization schemes are necessary.25 is discretized as 3(αp ρp φp V ol)n+1 − 4(αp ρp φp V ol)n + (αp ρp φp )n−1 = 2∆t X (23.5(α p ρp φ p V ol) ∆t . energy equations.26) [Anb (φnb − φp )] n+1 + S U n+1 − S p n+1 φp n+1 Equation 23. and Sφ is the source term. α is the phase volume fraction (unity for the mixture equation).
the motion of liquid after a dam break. 23. the motion of large bubbles in a liquid. Stratified flows of immiscible fluids should be solved with the VOF model (see Section 23. the secondorder time scheme is. it is also recommended not to start with a patch of volume fraction equal to zero. in this version of FLUENT. and the steady or transie nt tracking of any liquidgas interface. It is unconditionally stable. Some problems involving small volume fractions can be solved more efficiently with the Lagrangian discrete phase model (see Chapter 22: Modeling Discrete Phase ). and patched fields are used for the initial conditions. a bounded secondorder time scheme has been impleme nted for compressible liquid flows only.This scheme is easily implemented based on FLUENT’s existing firstorder Euler scheme. may produce oscillatory solutions if the time steps are large. the bounded scheme. If a timede pendent problem is being solved. and you may encounter some stability or convergence problems. by default. Many stability and convergence problems can be minimized if care is taken during the setup and solution processes (see Section 23. Another option is to start with a mixture multiphase calculation.3: Volume of Fluid (VOF) Model Theory). Therefore. oscillating solutions are most likely seen in compressible liquid flows. 23. For single phase and multiphase compressible liquid flows. However. of the threetime level method.3 Volumeof F luid (VO F )M odelTheory 23. however.2. and then switch to the Eulerian multiphase model. For steady solutions it is recommended that you start with a small underrelaxation factor for the volume fraction.14.3. Typical applications include the prediction of jet breakup. at least an order of magnitude smaller than the characteristic time of the flow. This problem can be eliminated if a bounded secondorder scheme is introduced. . You can increase the size of the time step after performing a few time steps. it is recommended that you perform a few iterations with a small time step.4: Eulerian Model).1 Overviewand Limitationsof the VO F Model Overview The VOF model can model two or more immiscible fluids by solving a single set of momentum equations and tracking the volume fraction of each of the fluids throughout the domain. the negative coefficient at the time level tn−1 .4 Stabilityand Convergence The process of solving a multiphase system is inherently difficult.
There is no limitation on using compressible liquids using userdefined functions. • The secondorder implicit timestepping formulation cannot be used with the VOF explicit scheme. such a problem must be solved using the timede pendent formulation. Thus the variables and properties in any given cell are either purely represe ntative of one of the phases. as long as the volume fraction of each of the phases is known at each location. a variable is introduced: the volume fraction of the phase in the computational cell. The VOF formulation relies on the fact that two or more fluids (or phases) are not interpenetrating. the DPM model cannot be used with the VOF model if the shared memory option is enabled (Section 22. the flow of water in a channel with a region of air on top and a separate air inlet can be solved with the steadystate formulation. • When tracking particles in parallel.9: Parallel Processing for the Discrete Phase Model). (Note that using the message passing option. or . the volume fractions of all phases sum to unity. with either of the densitybased solvers. enables the compatibili ty of all multiphase flow models with the DPM model. when running in parallel. For each additional phase that you add to your model.3 Volum eof F lu id(VO F )M odelTheory Limitations The following restrictions FLUENT: apply to the VOF model in • You must use the pressurebased solver.23. since the shape of the free surface inside a rotating cup depends on the initial level of the fluid. The VOF model is not available • All control volumes must be filled with either a single fluid phase or a combination of phases.) SteadyState and TransientVO F Calculations The VOF formulation in FLUENT is generally used to compute a timede pendent solution.11. but for problems in which you are concerned only with a steadystate solution. • Streamwise periodic flow (either specified mass flow rate or specified pressure drop) cannot be modeled when the VOF model is used. it is possible to perform a steadystate calculation. The VOF model does not allow for void regions where no fluid of any type is present. On the other hand. The fields for all variables and properties are shared by the phases and represent volumeaveraged values. A steadystate VOF calculation is sensible only when your solution is independent of the initial conditions and there are distinct inflow boundaries for the individual phases. For example. • Only one of the phases can be defined as a compressible ideal gas. In each control volume.
31. In other words. See Section 23. the primaryphase volume fraction will be computed based on the following constrai nt: n X q=1 αq = 1 (23. then the following three conditions are possible: • αq = 0: The cell is empty (of the q th fluid). the source term on the righthand side of Equation 23. depending upon the volume fraction values.31) p=1 where m˙ qp is the mass transfer from phase q to phase p and m˙ pq is the mass transfer from phase p to phase q.2 VolumeFractionEquation The tracking of the interface(s) between the phases is accomplished by the solution of a continuity equation for the volume fraction of one (or more) of the phases.3. For the q th phase. 23. By default.32) The volume fraction equation may be solved either through implicit or explicit time discretization. • 0 < αq < 1: The cell contains the interface between the q th fluid and one or more other fluids. the appropriate properties and variables will be assigned to each control volume within the domain.M odelingM ultiphase Flows represe ntative of a mixture of the phases. . is zero. but you can specify a consta nt or userdefined mass source for each phase. • αq = 1: The cell is full (of the q th fluid). The volume fraction equation will not be solved for the primary phase. Sαq .7: Modeling Mass Transfer in Multiphase Flows for more information on the modeling of mass transfer in FLUENT’s general multiphase models. this equation has the following form: 1 ∂ ρq ∂t (αq ρq ) + ∇ · (αq ρq~vq ) = Sαq + n X (m˙ pq − m˙ qp ) (23. Based on the local value of αq . if the q th fluid’s volume fraction in the cell is denoted as αq .
10.34) f where n + 1 = n = = αq. and QUICK. αn+1 ρ n+1 − αn ρn n X X q q q q n+1 n+1 n+1 Uf αq. Second Order Upwind.f V Uf = = index for new (curre nt) time step index for previous time step face value of the q th volume fraction. a timede pendent solution must be com. or CICSAM scheme volume of cell volume flux through the face. as is needed for the implicit scheme. as for the explicit scheme). modified HRIC.f n X ∆t m˙ p=1 ( pq − m˙ ) + S qp αq V (23. and the Modified HRIC schemes.2: Interpolation near the Interface). CICSAM.f ) = Sαq + (m˙ pq − m˙ qp ) V V + (ρ q ∆t p=1 f (23.puted. FLUENT’s standard finitedifference interpolation schemes. The ExplicitScheme In the explicit approach. Second Order Upwind and First Order Upwind.3. based on normal velocity This formulation does not require iterative solution of the trans port equation during each time step.1: Choosing a VOF Formulation for details. The reconstruction based schemes available in FLUENT are GeoReconstruct and DonorAcceptor. FLUENT’s standard finitedifference interpolation schemes are applied to the volume fraction values that were computed at the previous time step.The Im plicitScheme When the implicit scheme is used for time discretization. i When the explicit scheme is used. The discretization schemes available with explicit scheme for VOF are First Order Upwind. Modified HRIC. . The implicit scheme can be used for both timede pendent and steadystate calculations. QUICK. including those near the interface. a standard scalar trans port equation is solved iteratively for each of the secondaryphase volume fractions at each time step. See Section 23. are used to obtain the face fluxes for all cells. n αn+1 ρn+1 − αn ρ q q q q = V + X (ρ U n αn ) q f q.33) Since this equation requires the volume fraction values at the curre nt time step (rather than at the previous step. QUICK. the face fluxes can be interpolated either using interface reconstruction or using a finite volume discretization scheme (Section 23. computed from the firstor secondorder upwind. When the explicit scheme is used for time discretization.
Interpolation near the Interface FLUENT’scontrolvolume formulation requires that convection and diffusion fluxes through the control volume faces be computed and balanced with source terms within the control volume itself. In the geometric reconstruction and donoracceptor schemes.3.3.1: Interface Calculations . FLUENT applies a special interpolation treatme nt to the cells that lie near the interface between two phases.1 shows an actual interface shape along with the interfaces assumed during computation by these two methods. Figure 23. actual interface shape interface shape represented by the geometric reconstruction (piecewiselinear) scheme interface shape represented by the dono racceptor scheme Figure 23.
1: FirstOrder Upwind Scheme). you must ensure that there are no twosided (zerothickness) walls within the domain. rather than applying a special treatme nt.3. When the cell is near the interface between two phases. If there are.3.8. and uses this linear shape for calculation of the advection of fluid through the cell faces.) The first step in this reconstruction scheme is calculating the position of the linear interface relative to the center of each partiallyfilled cell. . modified HRIC (Section 25.3.1.3. The geometric reconstruction scheme is generalized for unstructured meshes from the work of Youngs [411]. a timede pendent solu. the geometric reconstruction scheme is used. In FLUENT this scheme is the most accurate and is applicable for general unstructured meshes.3. if the grid node locations are identical at the boundaries where two subdomains meet). if you are using a conformal grid (i. It assumes that the interface between two fluids has a linear slope within each cell.e.1: QUICK Scheme. The geometric reconstruction scheme represents the interface between fluids using a piecewiselinear approach. you will need to slit them. the standard interpolation schemes that are used in FLUENT are used to obtain the face fluxes whenever a cell is completely filled with one phase or another. QUICK (Section 25.1: SecondOrder Upwind Scheme).. using the standard upwind (Section 25. i When the geometric reconstruction scheme is used.e. based on information about the volume fraction and its derivatives in the cell. The third step is calculating the volume fraction in each cell using the balance of fluxes calculated during the previous step.tion must be computed. (See Figure 23. Also. The second step is calculating the advecting amount of fluid through each face using the computed linear interface representation and information about the normal and tange ntial velocity distribution on the face.. secondorder (Section 25.1: Modified HRIC Scheme). or CICSAM scheme).6: Slitting Face Zones. The Geometric Reconstruction Scheme In the geometric reconstruction approach. as described in Section 6.The explicit scheme and the implicit scheme treat these cells with the same interpolation as the cells that are completely filled with one phase or the other (i.
The CICSAM scheme is particularly suitable for flows with high ratios of viscosities between the phases. If there are. the donoracceptor scheme can be used only with quadrilateral or hexahedral meshes.The Dono rAcceptorScheme In the donoracceptor approach.6: Slitting Face Zones. a “donoracceptor” scheme is used to determine the amount of fluid advected through the face [144]. and that of the neighbor cell that shares the face in question. if the grid node locations are identical at the boundaries where two subdomains meet). or some combination of the two. Also.8. you will need to slit them. The orientation of the interface is also used in determining the face fluxes. pure downwinding. The CompressiveInterface Capturing Schem e for A rb itrary eshes (CICSAM ) M The compressive interface capturing scheme for arbitrary meshes (CICSAM).e. as described in Section 6. i When the donoracceptor scheme is used. . you must ensure that there are no twosided (zerothickness) walls within the domain. is a high resolution differencing scheme. The amount of fluid from one phase that can be convected across a cell boundary is limited by the minimum of two values: the filled volume in the donor cell or the free volume in the acceptor cell. a timede pendent solution must be computed. In addition. The interface orientation is either horizontal or vertical. and is used to prevent numerical diffusion at the interface. CICSAM is implemented in FLUENT as an explicit scheme and offers the advantage of producing an interface that is almost as sharp as the geometric reconstruction scheme. the standard interpolation schemes that are used in FLUENT are used to obtain the face fluxes whenever a cell is completely filled with one phase or another. flux values are obtained by pure upwinding. based on the Ubbink’s work [376]. depending on the direction of the volume fraction gradient of the q th phase within the cell. When the cell is near the interface between two phases. Depending on the interface’s orientation as well as its motion.. This scheme identifies one cell as a donor of an amount of fluid from one phase and another (neighbor) cell as the acceptor of that same amount of fluid. if you are using a conformal grid (i.
the accuracy of the velocities computed near the interface can be adversely affected.23. viscosity) are computed in this manner. Note that if the viscosity ratio is more than 1x103 .2: The Compressive Interface Capturing Scheme for Arbitrary Meshes (CICSAM)) is suitable for flows with high ratios of viscosities between the phases. .3 M aterial Properties The properties appearing in the trans port equations are determined by the presence of the component phases in each control volume.averaged density takes on the following form: ρ= X α q ρq (23. the volumefraction. and if the volume fraction of the second of these is being tracked.35) In general.37) One limitation of the sharedfields approximation is that in cases where large velocity differences exist between the phases. for an nphase system. shown below. The compressive interface capturing scheme for arbitrary meshes (CICSAM) (Section 23. for example. thus solving the problem of poor convergence. if the phases are represented by the subscripts 1 and 2. The momentum equation.4 M omentum Equation A single momentum equation is solved throughout the domain.g. In a twophase system.3. the density in each cell is given by ρ = α2 ρ2 + (1 − α2 )ρ1 (23.36) All other properties (e.3. h i ∂ (ρ~v) + ∇ · (ρ~v~v) = −∇p + ∇ · µ ∇~v + ∇~v T + ∂t ρ~g + F~ (23. is dependent on the volume fractions of all phases through the properties ρ and µ. 23.3.. this may lead to convergence difficulties. and the resulting velocity field is shared among the phases.
also shared among the phases.3. T . or once for each iteration within each time step.6 Additional Scalar Equations Depending upon your problem definition.3. 23. if a model includes liquid metal in combination with air. as well as any other volumetric heat sources. As with the velocity field. additional scalar equations may be involved in your solution. The source term. Equation 23. Sh .g. and the turbulence variables (e.3. FLUENT automatically refines the time step for the integration of the volume fraction equation.38) The VOF model treats energy. Such problems also arise in cases where the properties vary by several orders of magnitude. k and or the Reynolds stresses) are shared by the phases throughout the field. E.4: Setting TimeDependent Parameters for the VOF Model. is shown below.39) α q ρq n E = X q=1 where Eq for each phase is based on the specific heat of that phase and the shared temperature.. You can choose to update the volume fraction once for each time step. a single set of trans port equations is solved. . as massaveraged variables: n X q=1 α q ρq E q (23.10. and temperature. the accuracy of the temperature near the interface is limited in cases where large temperature differences exist between the phases.7 TimeDependence For timede pendent VOF calculations. ∂ (ρE) + ∇ · (~v(ρE + p)) = ∇ · (keff ∇T ) + Sh ∂t (23. For example.23. In the case of turbulence quantities.31 is solved using an explicit timemarching scheme. These options are discussed in more detail in Section 23. contains contributions from radiation. Such large discrepancies in properties lead to equation sets with anisotropic coefficients. 23. the conductivities of the materials can differ by as much as four orders of magnitude. but you can influence this time step calculation by modifying the Coura nt number. The properties ρ and keff (effective thermal conductivi ty) are shared by the phases.5 Energy Equation The energy equation. which in turn can lead to convergence and precision limitations.
or through a UDF. The solver will include the additional tange ntial stress terms (causing what is termed as Marangoni convection) that arise due to the variation in surface tension coefficient. At the surface. The model can be augmented by the additional specification of the contact angles between the phases and the walls. Within the bubble. Surface Tension Surface tension arises as a result of attracti ve forces between molecules in a fluid. It can be shown that the pressure drop across the surface depends upon the surface tension coefficient.3. You can specify a surface tension coefficient as a consta nt. a formulation of the CSF model is used. however. With this model. In regions where two fluids are separated.310) where p1 and p2 are the pressures in the two fluids on either side of the interface. where the surface curvature is computed from local gradients in the surface normal at the interface. and the combined effect of the radial components of force across the entire spherical surface is to make the surface contract. The surface tension model in FLUENT is the continuum surface force (CSF) model proposed by Brackbill et al. that is required to maintain equilibrium in such instances.311) . [39]. defined as the gradient of αq . The surface tension is a force. σ. but one of them is not in the form of spherical bubbles. n = ∇αq (23. and where only the forces normal to the interface are considered. Let n be the surface normal. In FLUENT. the net force on a molecule due to its neighbors is zero. acting only at the surface. Variable surface tension coefficient effects are usually importa nt only in zero/nearzero gravity conditions.23. the surface tension acts to minimize free energy by decreasing the area of the interface. the addition of surface tension to the VOF calculation results in a source term in the momentum equation. R1 and R2 : p2 − p1 = σ 1 1 + R 1 R2 (23. the volume fraction of the q th phase. consider the special case where the surface tension is consta nt along the surface. To understand the origin of the source term. as a function of temperature. thereby increasing the pressure on the concave side of the surface. the net force is radially inward. and the surface curvature as measured by two radii in orthogonal directions. for example. Consider an air bubble in water. It acts to balance the radially inward intermolecular attracti ve force with the radially outward pressure gradient force across the surface.8 Surface Adhesion Tension and Wall The VOF model can also include the effects of surface tension along the interface between each pair of phases.
is defined in terms of the divergence of the unit normal.315 shows that the surface tension source term for a cell is proportional to the average densi ty in the cell. then κi = −κj and ∇αi = −∇αj . It has the following form: Fvol = X pairs ij. and Equation 23. Note that the calculation of surface tension effects on triangular and tetrahedral meshes is not as accurate as on quadrilateral and hexahedral meshes. .36. nˆ [39]: κ = ∇ · nˆ where nˆ = n n (23.313) (23. i< j σij αi ρi κj ∇αj + αj ρj κi ∇αi 1 (ρ i + ρ j ) 2 (23.312) The surface tension can be written in terms of the pressure jump across the surface. κ.314 simplifies to Fvol = σij ρκi ∇αi +ρj) (23. The region where surface tension effects are most importa nt should therefore be meshed with quadrilaterals or hexahedra. If only two phases are present in a cell. Equation 23.314) This expression allows for a smooth superposition of forces near cells where more than two phases are present.The curvature.315) 1(ρ i 2 where ρ is the volumeaveraged density computed using Equation 23. The force at the surface can be expressed as a volume force using the divergence theorem. It is this volume force that is the source term which is added to the momentum equation.
We.g.316) 1. . and the Weber number.W henSurfaceTension Effects Are Im p o rtant The importance of surface tension effects is determined based on the value of two dimensionless quantities: the Reynolds number. Several surface tension options are provided through the text user interface (TUI) using the solve/set/surfacetension command: solve −→ set −→surfacetension The surfacetension command prompts you for the following information: • whether you require nodebased smoothing The default value is no indicating that cellbased smoothing will be used for the VOF calculations. the qua ntity of interest is the Weber number: We = ρLU 2 σ (23. The default is yes which produces better results with surface tension compared to gradients that are calculated at the cell centers. liquid enters through the inlet with wall adhesion on). and the capillary number. or the Reynolds number. Surface tension effects can be neglected if Ca or We 1. FLUENT uses VOF gradients directly from the nodes to calculate the curvature for surface tension forces. Ca. This is useful in the cases where VOF smoothing causes a problem (e. For Re 1. • the smoothing relaxation factor The default is 1. • the number of smoothings The default value is 1.. Re. the qua ntity of interest is the capillary number: Ca = and for Re µU σ (23. Re. A higher value can be used in case of tetrahedral and triangular meshes in order to reduce any spurious velocities. • whether you want to use VOF gradients at the nodes for curvature calculations With this option.317) 1 where U is the freestream velocity.
if you are setting the contact angle between the oil and air phases in the Wall panel shown in Figure 23. The model is taken from work done by Brackbill et al. [39]. θw is measured inside the oil phase. The combination of this contact angle with the normally calculated surface normal one cell away from the wall determine the local curvature of the surface.2. as seen in Figure 23. and this curvature is used to adjust the body force term in the surface tension calculation.3.3. then the surface normal at the live cell next to the wall is nˆ = nˆ w cos θw + tˆw sin θw (23. the contact angle that the fluid is assumed to make with the wall is used to adjust the surface normal in cells near the wall. respectively. This socalled dynamic boundary condition results in the adjustme nt of the curvature of the surface near the wall.Wall Adhesion An option to specify a wall adhesion angle in conjunction with the surface tension model is also available in the VOF model. Rather than impose this boundary condition at the wall itself. For example.318) where nˆ w and tˆw are the unit vectors normal and tange ntial to the wall.3. If θw is the contact angle at the wall. The contact angle θw is the angle between the wall and the tange nt to the interface at the wall. . measured inside the phase listed in the left column under Wall Adhesion in the Momentum tab of the Wall panel.
3.2: The Wall Panel for a Mixture in a VOF Calculation with Wall Adhesion .Figure 23.
dams.interface OIL AIR AIR wall OR θW = 30 o interface θW = 30 o OIL wall Figure 23. and surfacepiercing structures in unbounded stream) using the VOF formulation and the open channel boundary condition. Open channel flows are characterized by the dimensionless Froude Number. V Fr = √ gy (23. the wave propagation and free surface behavior becomes important. in this case. These flows involve the existence of a free surface between the flowing fluid and fluid above it (generally the atmosphere). The denominator in Equation 23.3.g. the distance from the bottom of the channel to the free surface. The wave speed as seen by the fixed observer is defined as √ Vw = V ± gy (23. g is gravity. and y is a length scale. In such cases.319) where V is the velocity magnitude. rivers.319 is the propagation speed of the wave. This feature is mostly applicable to marine applications and the analysis of flows through drainage systems.3: Measuring the Contact Angle 23.3.9 Open ChannelFlow FLUENT can model the effects of open channel flow (e. Flow is generally governed by the forces of gravity and inertia.. which is defined as the ratio of inertia force and hydrostatic force.320) .
V > gy (thus Vw > 0). In this case. i. g is the gravity vector. V is the velocity magnitude.Based on the Froude number. V < gy (thus Vw < 0 or Vw > 0). respectively. and ρ0 is the reference densi ty.322) .e.. downstream conditions do not affect the flow upstream.e. the flow is known to be critical. downstream conditions might affect the flow upstream. In this case. From this. −→g  is the gravity magnitude. the dynamic pressure q is q= and the static pressure ps is −→ ps = (ρ − ρ0 )−→g (gˆ · ( b − −→a )) (23. the flow is known to be supercriti cal where disturbances cannot travel upstream. the character of the flow changes.323) ρ − ρ0 2 V 2 (23. open channel flows can be classified in the following three categories: √ • When F r < 1. gˆ is the unit vector of gravity. √ • When F r > 1. the flow is known to be subcritical where disturbances can travel upstream as well as downstream.32 21) where −→ b and −→a are the position vectors of the face centroid and any point on the free surface. direction of gravity.. In this case. ρ is the density of the mixture in the cell. where upstream propagating waves remain stationar y. • When F r = 1 (thus Vw = 0). i. UpstreamBoundaryConditions There are two options available for the upstream boundary condition for open channel flows: • pressure inlet • mass flow rate Pressure Inlet The total pressure p0 at the inlet can be given as p0 = 1 −→ (ρ − ρ0 )V2 + (ρ − ρ0 )−→g (gˆ · ( b − −→a )) (23. −→ free surface is assumed to be horizontal and normal to the Here.
• Calculate the volume fraction at the each face of boundary using the interpolated node values. otherwise the pressure is taken from the neighboring cell.326) (23. For subcritical inlet flows (Fr < 1).324) where the distance from the free surface to the reference position. . therefore this option has been disabled. the volume fraction value on the boundary can be calculated using the fixed height of the free surface from the bottom.323 and Equation 23. For supercritical inlet flows (Fr > 1). (23. otherwise you must specify the static pressure as the Gauge Pressure. is ylocal = −(−→a · gˆ) Mass Flow Rate The mass flow rate for each phase associated with the open channel flow is defined by m˙ phase = ρphase (Area phase )(V elocity) Volum eFractionSpecification In open channel flows. ylocal . For supercritical flows (Fr >1). if there are only two phases.which can be further expanded to ps = (ρ − ρ0 )−→g ((gˆ · −→ b ) + ylocal ) (23. then the pressure is taken from the pressure profile specified over the boundary. the pressure is always taken from the neighboring cell. FLUENT internally calculates the volume fraction based on the input parameters specified in the Boundary Conditions panel. FLUENT reconstructs the volume fraction values on the boundary by using the values from the neighboring cells.325. For subcritical outlet flows (Fr < 1). the static pressure is dictated by Equation 23. This can be accomplished using the following procedure: • Calculate the node values of volume fraction at the boundary using the cell values. Using the Free Surface Level.325) DownstreamBoundaryConditions Pressure Outlet Determining the static pressure is dependent on the Pressure Specification Method.
but assume local equilibrium over short spatial length scales.ting the flow rate weighting to 1. to understand the limitations of this boundary type: • You can only use single outflow boundaries at the outlet.4 M ixture odelTheory M O utfl w Boundary o Outflow boundary conditions can be used at the outlet of open channel flows to model flow exits where the details of the flow velocity and pressure are not known prior to solving the flow problem. It is importa nt. • There should be an initial flow field in the simulation to avoid convergence issues due to flow reversal at the outflow. • An outflow boundary condition can only be used with mass flow inlets. In addition. For example.23. If you choose the inlet as massflowinlet. this option is disabled. It can also be used to model homogeneous multiphase flows with very strong coupling and the phases moving at the same velocity. 23. such surfaces should be placed accordingly. then you can only use pressureoutlet at the outlet. then FLUENT will extra polate the required information from the interior. Backflow Volum eFractionSpecification FLUENT internally calculates the volume fraction values on the outlet boundary by using the neighboring cell values. which is achieved by set. Therefore.4 M ixtureM odelTheory 23. .4.1 Overviewand Limitationsof the M ixture Model Overview The mixture model is a simplified multiphase model that can be used to model multiphase flows where the phases move at different velocities. outflow splitting is not permitted in open channel flows with outflow boundaries. however. If the conditions are unknown at the outflow boundaries. In other words. Note that this only holds true for open channel flow. therefore. It is not compatible with pressure inlets and pressure outlets. which will result in an unreliable solution. if you choose the inlet as pressureinlet. the mixture model can be used to calculate nonNewtonian viscosity. then you can use either outflow or pressureoutlet boundary conditions at the outlet. • Note that the outflow boundary condition assumes that flow is fully developed in the direction perpendicular to the outflow boundary surface. The coupling between the phases should be strong.
• Solidification and melting cannot mixture model. and algebraic expressions for the relative velocities. This is applicable for liquidsolid flows. • When tracking particles in parallel. • The shell conduction model for walls cannot be used with the mixture model. • Only one of the phases can be defined as a compressible ideal gas. There is no limitation on using compressible liquids using userdefined functions.9: Parallel Pro. • Streamwise periodic flow with specified mass flow rate cannot be modeled when the mixture model is used (the user is allowed to specify a pressure drop). The mixture model allows you to select granular phases and calculates all properties of the granular phases. A full multiphase model may not be feasible when there is a wide distribution of the particulate phase or when the interphase laws are unknown or their reliability can be questioned.ture model if the shared memory option is enabled (Section 22. Limitations The following limitations apply to the mixture model in FLUENT: • You must use the pressurebased solver. be modeled in conjunction with the • The LES turbulence model cannot be used with the mixture model if the cavitation model is enabled. cyclone separators. and energy equations for the mixture. • The mixture model cannot be used for inviscid flows. Typical applications include sedime ntation. enables the compatibili ty of all multiphase flow models with the DPM model. continuity. The mixture model is a good substitute for the full Eulerian multiphase model in several cases. the volume fraction equations for the secondary phases. when running in parallel. The mixture model is not available with either of the densitybased solvers.cessing for the Discrete Phase Model). the DPM model cannot be used with the mix. • The relative velocity formulation cannot be used in combination with the MRF and mixture model (see Section 10.3. (Note that using the message passing option.1: Limitations ). A simpler model like the mixture model can perform as well as a full multiphase model while solving a smaller number of variables than the full multiphase model.11. and bubbly flows where the gas volume fraction remains low.M odelingM ultiphase Flows The mixture model can model n phases (fluid or particulate) by solving the momentum.) . particleladen flows with low loading.
It differs from the VOF model in two respects: • The mixture model allows the phases to be interpenetrating. ρm n X k=1 α k ρk (23. uses a singlefluid approach. depending on the space occupied by phase q and phase p. The volume fractions αq and αp for a control volume can therefore be equal to any value between 0 and 1. and the mixture model is then reduced to a homogeneous multiphase model. (Note that the phases can also be assumed to move at the same velocity.43) .2 ContinuityEquation The continuity equation for the mixture is ∂ (ρm ) + ∇ · (ρm~vm ) = 0 ∂t where ~vm is the massaveraged velocity: Pn kρk~vk =1 (23.42) ~vm = and ρm is the mixture densi ty: ρm = αk is the volume fraction of phase k. using the concept of slip velocities. like the VOF model. the momentum equation for the mixture.4. 23.23.41) αk (23. • The mixture model allows the phases to move at different velocities.4 M ixture odelTheory M The mixture model. the energy equation for the mixture. as well as algebraic expressions for the relative velocities (if the phases are moving at different velocities). and the volume fraction equation for the secondary phases.) The mixture model solves the continuity equation for the mixture.
47 represents energy transfer due to conduction. where hk is the sensible enthalpy for phase k. It can be expressed as h i ∂ (ρm~vm ) + ∇ · (ρm~vm~vm ) = −∇p + ∇ · µm ∇~vm + ∇~v T + m ∂t ρm~g + F~ + ∇ · X n k=1 αk ρk~vdr . F~ is a body force.4. The first term on the righthand side of Equation 23.k is the drift velocity for secondary phase k: ~vdr.k ! (23. In Equation 23.47) where keff is the effective conductivity ( αk (kk + kt )).4. SE includes any other volumetric heat sources. and Ek = hk for an incompressible phase. defined according to the turbulence model being used).46) 23.44) where n is the number of phases. and µm is the viscosity of the mixture: µm = n X k=1 αk µk (23.47.k~vdr . E k= h k− p ρk + vk 2 2 P (23.23.48) for a compressible phase.45) ~vdr.4 Energy Equation The energy equation for the mixture takes the following form: n n X ∂ X (αk ρk Ek ) + ∇ · (αk~vk (ρk Ek + p)) = ∇ · (keff ∇T ) + SE ∂t k=1 k=1 (23.3 M omentum Equation The momentum equation for the mixture can be obtained by summing the individual momentum equations for all phases. where kt is the turbule nt thermal conductivi ty. .k = ~vk − ~vm (23.
[229].413) τp (ρp − ρm ) ~a fdrag ρp (23.687 Re ≤ 1000 0.p = ~vpq − k=1 X ck~vqk (23. ~a is the secondaryphase particle’s acceleration. a local equilibrium between the phases should be reached over short spatial length scale.0183 Re Re > 1000 (23.5 Relative(Slip) Velocity and the D riftVelocity The relative velocity (also referred to as the slip velocity) is defined as the velocity of a secondary phase (p) relative to the velocity of the primary phase (q): ~vpq = ~vp − ~vq The mass fraction for any phase (k) is defined as ck = αk ρk ρm (23. The default drag function fdrag is taken from Schiller and Naumann [320]: ( fdrag = 1 + 0.411) FLUENT’s mixture model makes use of an algebraic slip formulation.414) and the acceleration ~a is of the form ~a = ~g − (~vm · ∇)~vm − ∂~vm ∂t (23.49) The drift velocity and the relative velocity (~vqp ) are connected by the following expression: n ~vdr.4. the form of the relative velocity is given by: ~vpq = where τp is the particle relaxation time ρp d2 p τp = 18µq (23.415) .23. The basic assumption of the algebraic slip mixture model is that to prescribe an algebraic relation for the relative velocity.412) d is the diameter of the particles (or droplets or bubbles) of secondary phase p.15 Re0. Following Manninen et al.410) (23.
in which the acceleration of the particle is given by gravity and/or a centrifugal force and the particulate relaxation time is modified to take into account the presence of other particles.4. if the slip velocity is not solved.417) . In addition.5.sion appearing in the momentum equation for the dispersed phase. The following choices are available: • SchillerNaumann (the default formulation) • MorsiAlexander • symmetric • consta nt • userdefined See Section 23. you can directly prescribe formulations for the drag function.6 VolumeFractionEquationfor the Secondary Phases From the continuity equation for secondary phase p. and Section 23.4: Interphase Exchange Coefficients for more information on these drag functions and their formulations. Note that. When you are solving a mixture multiphase calculation with slip velocity. FLUENT adds this dispersion to the relative velocity: ~vpq = 2 (ρp − ρm )dp νm ∇α q ~a 18µq fdrag αp σD − (23. In turbule nt flows the relative velocity should contain a diffusion term due to the disper.p ) + ∂t q=1 X (m˙ qp − m˙ pq ) (23.The simplest algebraic slip formulation is the socalled drift flux model. the mixture model is reduced to a homogeneous multiphase model. the mixture model can be customized (using userdefined functions) to use a formulation other than the algebraic slip method for the slip velocity.1: Defining the Phases for the Mixture Model for instructions on how to enable them. 23. the volume fraction equation for secondary phase p can be obtained: n ∂ (αp ρp ) + ∇ · (αp ρp~vm ) = −∇ · (αp ρp~vdr. See the separate UDF Manual for details.11.416) where (νm ) is the mixture turbule nt viscosity and (σD ) is a Prandtl dispersion coefficient.
4. The volume weighted averaged for the viscosity would now contain shear viscosity arising from particle momentum exchange due to translation and collision. [119] is also available: µs.7 G ranular Properties Since the conce ntration of particles is an importa nt factor in the calculation of the effective viscosity for the mixture.kin √ 10ρs ds Θs π 4 = 1 + g0. [363]: µs.421) .kin √ αs ds ρs Θs π 2 = 6 (3 − e ) 1 + (1 + ess ) (3ess − 1) αs ss g0.col = 4 αs ρs ds g0.fr (23. are added to give the solids shear viscosity: µs = µs. The collisional and kinetic parts.418) CollisionalViscosity The collisional part of the shear viscosity is modeled as [119.419) KineticViscosity FLUENT provides two expressions for the kinetic viscosity. 363] µs.ss (1 + ess ) Θs 5 π 1/2 (23. and the optional frictional part.420) The following optional expression from Gidaspow et al.kin + µs.ss 2 (23.ss 5 (23.col + µs.ss αs (1 + ess ) 5 96αs (1 + ess) g0. we may use the granular viscosity (see section on Eulerian granular flows) to get a value for the viscosity of the suspension.23. The default expression is from Syamlal et al.
4.total = N X q=1 pq (23.23.4.9 Solids Pressure The total solid pressure is calculated and included in the mixture momentum equations: Ps.ss 3/2 √ ρs αs2 Θs ds π (23.8 G ranular Temperature The viscosities need the specification of the granular temperature for the sth solids phase.422) [363].423) The transfer of the kinetic energy of random fluctuations in particle velocity from the sth solids phase to the lth fluid or solid phase is represented by φls [119]: φls = −3Kls Θs (23.424) FLUENT allows you to solve for the granular temperature with the following options: • algebraic formulation (the default) This is obtained by neglecting convection and diffusion in the trans port equation (Equation 23.425) where pq is presented in the section for granular flows by equation Equation 23. • consta nt granular temperature This is useful in very dense situations where the random fluctuations are small. γΘs . [221] γΘm = 2 12(1 − ess )g0. • UDF for granular temperature 23. Here we use an algebraic equation derived from the trans port equation by neglecting convection and diffusion and takes the form [363] 0 = (−ps I + τ s ) : ∇~vs − γΘs + φls where (−ps I + τ s ) : ∇~vs = the generation of energy by the solid stress tensor γΘs = the collisional dissipation of energy φls = the energy exchange between the lth fluid or solid phase and the sth solid phase The collisional dissipation of energy. This term is represented by the expression derived by Lun et al.548 .422) (23. represents the rate of energy dissipation within the sth solids phase due to collisions between particles.
1: Overview and Limitations of the Eulerian Model • Section 23.4: Eulerian Model for multiphase modeling strategies. With the Eulerian multiphase model.5. Any number of secondary phases can be modeled.5.11: Solution Method in FLUENT 23.5.3: Conservation Equations • Section 23. yet interacting phases. See Section 23.2: Volume Fractions • Section 23. however. .5 EulerianM odelTheory Details about the Eulerian multiphase model are presented in the following subsections: • Section 23.10: Turbulence Models • Section 23. in contrast to the EulerianLagrangian treatme nt that is used for the discrete phase model.8: Granular Temperature • Section 23.5.5.1 Overviewand Limitationsof the EulerianM odel Overview The Eulerian multiphase model in FLUENT allows for the modeling of multiple separate. An Eulerian treatme nt is used for each phase. gases.5.5.6: Maximum Packing Limit in Binary Mixtures • Section 23. provided that sufficient memory is available.5. The phases can be liquids. you may find that your solution is limited by convergence behavior. A granular flow is simply one that involves at least one phase that has been designated as a granular phase.7: Solids Shear Stresses • Section 23.5: Solids Pressure • Section 23.5.5. For complex multiphase flows. or solids in nearly any combination.23.4: Interphase Exchange Coefficients • Section 23. the number of secondary phases is limited only by memory requirements and convergence behavior.5.14.5 EulerianM odelTheory 23.9: Description of Heat Transfer • Section 23. FLUENT’s Eulerian multiphase model does not distinguish between fluidfluid and fluidsolid (granular) multiphase flows.5.
9: Par. • Inviscid flow is not allowed.ual. • Several interphase drag coefficient functions are available. • Streamwise periodic flow with specified mass flow rate cannot be modeled when the Eulerian model is used (the user is allowed to specify a pressure drop).turbulence models are available.M odelingM ultiphase Flows The FLUENT solution is based on the following: • A single pressure is shared by all phases. except for the following limitations: • The Reynolds Stress turbulence model is not available on a per phase basis.11.rian multiphase model if the shared memory option is enabled (Section 22. and may apply to all phases or to the mixture.) . the DPM model cannot be used with the Eule. enables the compatibili ty of all multiphase flow models with the DPM model. or a partial differential equation. • Momentum and continuity equations are solved for each phase. Limitations All other features available in FLUENT can be used in conjunction with the Eulerian multiphase model. You can select either an algebraic formulation.ing option. • When tracking particles in parallel. (Note that using the message pass. • Particle tracking (using the Lagrangian dispersed phase model) interacts only with the primary phase. (You can also modify the interphase drag coefficient through userdefined functions.allel Processing for the Discrete Phase Model). • The following parameters are available for granular phases: – Granular temperature (solids fluctuating energy) can be calculated for each solid phase. a consta nt. You can select appropriate models and userdefined functions for all properties.) • All of the k. a userdefined function. when running in parallel. as described in the separate UDF Man. which are appropriate for various types of multiphase regimes. • Melting and solidification are not allowed. Frictional viscosity for modeling granular flow is also available. – Solidphase shear and bulk viscosities are obtained by applying kinetic theory to granular flows.
2 VolumeFractions The description of multiphase flow as interpenetrating continua incorporates the concept of phasic volume fractions. . α2 . The derivation of the conservation equations can be done by ensemble averaging the local insta ntaneous balance for each of the phases [10] or by using the mixture theory approach [36].5 EulerianM odelTheory To change from a singlephase model. the introduction of the volume fractions α1 . where a single set of conservation equations for momentum. to a multiphase model. as well as mechanisms for the exchange of momen.tum.52) The effective density of phase q is ρˆq = αq ρq where ρq is the physical density of phase q. and mass between the phases.51) αq = 1 (23. .23. Volume fractions represent the space occupied by each phase. the original set must also be modified.5. The volume of phase q. additional sets of conservation equations must be introduced. Vq . . and the laws of conservation of mass and momentum are satisfied by each phase individually. In the process of introducing additional sets of conservation equations. among other things. (23. The modifications involve.53) . is defined by Z Vq = where n X q=1 V αq dV (23. heat. continuity and (optionally) energy is solved. αn for the multiple phases. 23. denoted here by αq .
~vpq is the interphase velocity.q ) (23.q is a lift F~vm.54 is zero. and m˙ qp characterizes the mass transfer from phase q to phase p.54) where ~vq is the velocity of phase q and m˙ pq characterizes the mass transfer from the pth to q th phase. but you can specify a consta nt or userdefined mass source for each phase.55) p=1 where τ q is the q th phase stressstrain tensor 2 τ q = αq µq (∇~vq + ∇~v T ) + αq (λq µq )∇ · ~vq I q 3 − (23. if m˙ pq < 0 (i. ~vpq = ~vp .e.7: Modeling Mass Transfer in Multiphase Flows for more information on the modeling of mass transfer in FLUENT’s general multiphase models. F~q is an external body force. followed by the solved equations themsel ves. between phases. phase p mass is being transferred to phase q). force. If m˙ pq > 0 (i. defined as follows. and p is the pressure shared by all phases.5. the source term Sq on the righthand side of Equation 23. ~vpq = ~vq .q is a virtual mass R~ pq is an interaction force force. See Section 23..q + F~vm. and you are able to specify these mechanisms separatel y. if m˙ qp > 0 then vqp = vq . By default. Equationsin GeneralForm Conservation Mass of The continuity equation for phase q is nX ∂ (αq ρq ) + ∇ · (αq ρq~vq ) = (m˙ pq − m˙ qp ) + Sq ∂t p=1 (23.56) Here µq and λq are the shear and bulk viscosity of phase q. Conservation Momentum of The momentum balance for phase q yields ∂ (αq ρq~vq ) + ∇ · (αq ρq~vq~vq ) = −αq ∇p + ∇ · τ q + αq ρq~g+ ∂t n X (R~ pq + m˙ pq~vpq − m˙ qp~vqp ) + (F~q + F~lift.3 ConservationEquations The general conservation equations from which the equations solved by FLUENT are derived are presented in this section. Likewise. phase q mass is being transferred to phase p). A similar term appears in the momentum and enthalpy equations.e.23. F~lift.. if m˙ qp < 0 .
.then vqp = vp .
In most cases.4: Interphase Exchange Coefficients). F~lift is not included. The lift force acting on a secondary phase p in a primary phase q is computed from [88] F~lift = −0. FLUENT uses a simple interaction term of the following form: n X p=1 R~ pq = n X Kpq (~vp − ~vq ) (23.5. if desired. cohesion.Equation 23. the lift force is insignificant compared to the drag force.55 must be closed with appropriate expressions for the interphase force R~ pq . This force depends on the friction. it may be appropriate to include this term. The lift force will be more significant for larger particles.q = −F~lift . These lift forces act on a particle mainly due to velocity gradients in the primaryphase flow field. Thus. i It is importa nt that if you include the lift force in your calculation. FLUENT can include the effect of lift forces on the secondary phase particles (or droplets or bubbles). so there is no reason to include this extra term.57) p=1 where Kpq (= Kqp ) is the interphase momentum exchange coefficient (described in Section 23. you need not include it everywhere in the computational domain since it is computationally expensive to converge. Lift Forces For multiphase flows. . the lift force is significant when the slip velocity is large in the vicinity of high strain rates for the primary phase. By default. The lift force and lift coefficient can be specified for each pair of phases.58) The lift force F~lift will be added to the righthand side of the momentum equation for both phases (F~lift.5ρq αp (~vq − ~vp ) × (∇ × ~vq ) (23.. if the phases separate quickly). If the lift force is significant (e. pressure. and is subject to the conditions that R~ pq = −R~ qp and R~ qq = 0.p ). the inclusion of lift forces is not appropriate for closely packed particles or for very small particles. For example. but the FLUENT model assumes that the particle diameter is much smaller than the interparticle spacing. and other effects. in the wall boundary layer for turbule nt bubbly flows in channels.g.
Virtual Mass Force For multiphase flows.510) The virtual mass force F~vm will be added to the righthand side of the momentum equation for both phases (F~vm.g. due to chemical reaction or radiation). The virtual mass effect is significant when the secondary phase density is much smaller than the primary phase density (e. .q = −F~vm. the enthalpy of the vapor at the temperature of the droplets.p ).. The inertia of the primaryphase mass encountered by the accelerating particles (or droplets or bubbles) exerts a “virtual mass force” on the particles [88]: F~vm = 0. FLUENT includes the “virtual mass effect” that occurs when a secondary phase p accelerates relative to the primary phase q. and hpq is the interphase enthalpy (e. F~vm is not included.511) where hq is the specific enthalpy of the q th phase. The heat exchange between phases must comply with the local balance conditions Qpq = −Qqp and Qqq = 0. for a transie nt bubble column).59) dt − dt The term dq dt denotes the phase material time derivative of the form dq (φ) (~v dt = ∂(φ) ∂t + q · ∇)φ (23.g. Conservation Energy of To describe the conservation of energy in Eulerian multiphase applications. Qpq is the intensity of heat exchange between the pth and q th phases.5αp ρq dq~v q dp~vp ! (23. in the case of evaporation). Sq is a source term that includes sources of enthalpy (e.. By default.g. ~qq is the heat flux. a separate enthalpy equation can be written for each phase: n X ∂ (αq ρq hq ) + ∇ · (αq ρq ~uq hq ) = ∂pq + τ q : ∇~uq − ∇ ·~qq + Sq + (Qpq + m˙ pq hpq − ∂t ∂t −αq m˙ qp hqp ) p=1 (23..
q . as solved by FLUENT. The solidphase stresses are derived by making an analogy between the random particle motion arising from particleparticle collisions and the thermal motion of molecules in a gas. viscosity. 51. This treatme nt is common to fluidfluid and granular flows.512) where ρrq is the phase reference density. 363]. Continuity Equation The volume fraction of each phase is calculated from a continuity equation: n X 1 ∂ (αq ρq ) + ∇ · (αq ρq~vq ) = (m˙ pq − m˙ qp ) ρrq ∂t p=1 (23. and pressure of the solid phase. are presented here for the general case of an nphase flow. The solution of this equation for each secondary phase. FluidFluid om entum M Equations The conservation of momentum for a fluid phase q is ∂ (αq ρq~vq ) + ∇ · (αq ρq~vq~vq ) = −αq ∇p + ∇ · τ q + αq ρq~g + ∂t n X (Kpq (~vp − ~vq ) + m˙ pq~vpq − m˙ qp~vqp ) + (23. and F~vm. 119. As is the case for a gas. F~q . The kinetic energy associated with the particle velocity fluctuations is represented . 79. FLUENT uses a multifluid granular model to describe the flow behavior of a fluidsolid mixture.55.52). 198.q + F~vm. along with the condition that the volume fractions sum to one (given by Equation 23. FluidSolid om entum M Equations Following the work of [7. allows for the calculation of the primaryphase volume fraction.EquationsSolved by FLUENT The equations for fluidfluid and granular multiphase flows. F~lift.513) p=1 (F~q + F~lift.q ) Here ~g is the acceleration due to gravity and τ q . taking into account the inelasticity of the granular phase.q are as defined for Equation 23. 267. or the volume averaged density of the q th phase in the solution domain. 221. the intensity of the particle velocity fluctuations determines the stresses.
N is the total number of phases. for granular flows. This has an impact on how each of the fluids is assigned to a particular phase. and F~vm.4 InterphaseExchange Coefficients It can be seen in Equations 23. in flows where there are unequal amounts of two fluids. the predomina nt fluid should be modeled as the primary fluid. FluidFluid Exchange Coefficient For fluidfluid flows. The conservation of momentum for the fluid phases is similar to Equation 23.513. since the sparser fluid is more likely to form droplets or bubbles.514) (F~s + F~lift . the fluidsolid and solidsolid exchange coefficients Kls . each secondary phase is assumed to form droplets or bubbles. Conservation Energy of The equation solved by FLUENT for the conservation of energy is Equation 23.511.514 that momentum exchange between the phases is based on the value of the fluidfluid exchange coefficient Kpq and.5.513 and 23. Kls = Ksl is the momentum exchange coefficient between fluid or solid phase l and solid phase s.516) . liquidliquid or gasliquid mixtures can be written in the following general form: Kpq = αq αp ρp f τp (23. 23. F~lift. and that for the sth solid phase is ∂ (αs ρs~vs ) + ∇ · (αs ρs~vs~vs ) = −αs ∇p − ∇ps + ∇ · τ s + αs ρs~g + ∂t N X l=1 (Kls (~vl − ~vs ) + m˙ ls~vls − m˙ sl~vsl ) + (23. the drag function. is defined differently for the different exchangecoefficient models (as described below) and τp . and F~q . the “particulate relaxation time”.s + F~vm. is defined as ρ d2 τp = p 18µq p (23. For example.s ) where ps is the sth solids pressure. The exchange coefficient for these types of bubbly.q .55.by a “pseudothermal” or granular temperature which is proportional to the mean square of the random motion of particles.q are as defined for Equation 23.515) where f .
515. and it is acceptable for general use for all fluidfluid pairs of phases. Kpq should tend to zero whenever the primary phase is not present within the domain. the drag function f is always multiplied by the volume fraction of the primary phase q. as is reflected in Equation 23.15 Re0.687 )/Re Re ≤ 1000 0.521) . It is this drag function that differs among the exchangecoefficient models. The relative Reynolds number for the primary phase q and secondary phase p is obtained from Re = ρq ~vp − ~vq dp µq (23. • For the model of Schiller and Naumann [320] f = where ( CD Re 24 (23. For all these situations.522) CD Re 24 (23. • For the Morsi and Alexander model [252] f = where CD = a1 + a2 a + 32 Re Re (23.where dp is the diameter of the bubbles or droplets of phase p. To enforce this.520) where µrp = αp µp + αr µr is the mixture viscosity of the phases p and r.44 Re > 1000 (23. Nearly all definitions of f include a drag coefficient (CD ) that is based on the relati ve Reynolds number (Re). The Schiller and Naumann model is the default method.519) The relative Reynolds number for secondary phases p and r is obtained from Re = ρrp ~vr − ~vp  drp µrp (23.518) and Re is the relative Reynolds number.517) CD = 24(1 + 0.
357.73. −47500 1000 < Re < 5000 0.and Re is defined by Equation 23. Note that if there is only one dispersed phase.8889 a1 .0903 1. 24. −116. the air is the continuous phase.520. Thus for a single dispersed phase. The symmetric model is recommended for flows in which the secondary (dispersed) phase in one region of the domain becomes the primary (continuous) phase in (dp +dq ) another. a2 . 0. −3. in the top half of the container. −1662.222. The a’s are defined as follows: 0. then dp = dq in Equation 23.687 )/Re Re ≤ 1000 0.5.33. 5416700 Re ≥ 10000 (23.50. a3 = 0 < Re < 0.546.62.6167.526) CD = 24(1 + 0. −490.3644. 46. .527) and Re is defined by Equation 23.46. but calculations with this model may be less stable than with the other models.520.15 Re0.1 < Re < 1 1 < Re < 10 0.5191. the air is the dispersed phase in the bottom half of the container. 98.523) The Morsi and Alexander model is the most complete. adjusting the function definition frequently over a large range of Reynolds numbers.690. 29. 578700 5000 < Re < 10000 0 . 22. if air is injected into the bottom of a container filled halfway with water.519 or 23.525) CD Re 24 (23.44 Re > 1000 (23.1 0.524) (αp ρp + αq ρq )( p2 q )2 18(αp µp + αq µq ) d +d (23.519 or 23. This model can also be used for the interaction between secondary phases. • For the symmetric model Kpq = where τpq = and f = where ( αp (α p ρp + αq ρq )f τpq (23. 148. 0 3. 2778 100 < Re < 1000 0.1667.67 10 < Re < 100 − 0. For 2 example. dp = dq and = dp .525.
the “particulate relaxation time”.e.You can specify different exchange coefficients for each pair of phases. If the exchange coefficient is equal to zero (i.8 CD = 0. s is for the sth solid phase. with correlations that are a function of the volume fraction and relative Reynolds number [305]: Res = ρl ds ~vs − ~vl  µl (23. FluidSolidExchange Coefficient The fluidsolid exchange coefficient Ksl can be written in the following general form: Ksl = α s ρs f τs (23. and ds is the diameter of the sth solid phase particles. . It is also possible to use userdefined functions to define exchange coefficients for each pair of phases. It is this drag function that differs among the exchangecoefficient models.532) where the subscript l is for the lth fluid phase.s 2 (23. and τs .528) where f is defined differently for the different exchangecoefficient models (as described below). the flow fields for the fluids will be computed independently. • For the SyamlalO’Brien model [362] f= CD Res αl 2 24vr. if no exchange coefficient is specified).530) where the drag function has a form derived by Dalla Valle [73] 4. with the only “interaction” being their complementary volume fractions within each computational cell.529) where ds is the diameter of particles of phase s. is defined as τs = 2 ρs ds 18µl (23. All definitions of f include a drag function (CD ) that is based on the relative Reynolds number (Res ).s (23..63 + q Res /v r.531) This model is based on measurements of the terminal velocities of particles in fluidized or settling beds.
06 Res + (0. This model is appropriate for dilute systems. and B = α2. the fluidsolid exchange coefficient is of the following form: Ksl = where CD = 3 CD 4 s αl ρl ~vs − ~vl  ds αl −2. This model is appropriate when the solids shear stresses are defined according to Syamlal et al. .535) (23. [363] (Equation 23.85.5 A − 0.s ! ~vs − ~vl  (23.85.537) for αl > 0.The fluidsolid exchange coefficient has the form 3αs αl ρl Ksl = C 2 4vr.12 Res (2B − A) + A2 (23.28 for αl ≤ 0.564).687 α l Res (23.65 l (23.534) with A = αl4.s ds D Res vr.15(αl Res )0.14 and B = 0.532.65 (23.538) i 24 h 1 + 0.533) where vr.s = 0. • For the model of Wen and Yu [396].536) (23.s is the terminal velocity correlation for the solid phase [113]: q vr.8αl1.06 Res )2 + 0.539) and Res is defined by Equation 23.
687 α l Res (23.5: Radial Distribution Function .ls 2 ~vl − ~vs  (23.• The Gidaspow model [119] is a combination of the Wen and Yu model [396] and the Ergun equation [96].5.ls 2π α ρ α ρ 8 s s l l 3 2π (ρ l d3 + ρs ds ) l Kls = where els Cfr. When αl > 0.ls l + ds ) g0.540) i 24 h 1 + 0.5: Solids Pressure and the radial distribution coefficient is described in Section 23.5. the fluidsolid exchange coefficient Ksl is of the following form: Ksl = where CD = When αl ≤ 0.8.8.75 ~vl  α d2 ds l s This model is recommended for dense fluidized beds. (23.ls = 0) = the diameter of the particles of solid l = the radial distribution coefficient Note that the coefficient of restitution is described in Section 23.15(αl Res )0.541) SolidSolidExchange Coefficient The solidsolid exchange coefficient Kls has the following form [361]: 3 (1 + els ) (d π 2 + Cfr. .542) 3 CD 4 s αl ρl ~vs − ~vl  ds αl −2.543) = the coefficient of restitution = the coefficient of friction between the lth and sth solidphase particles (C fr.65 (23.ls dl g0. αs (1 − αl )µl ρl αs ~vs − Ksl = 150 + 1.
where the solids volume fraction is less than its maximum allowed value). and Θs is the granular temperature. Equations derived empirically may not satisfy this property and need to be changed accordingly without deviating significantly from the original form. g0. a granular temperature is introduced into the model.9 for ess .. in the granularphase momentum equation.max .5. The solids pressure is composed of a kinetic term and a second term due to particle collisions: ps = αs ρs Θs + 2ρs (1 + ess )α2 g0. Θp .ss Θs s (23. a general solids pressure formulation in the presence of other phases could be of the form ps = αs ρs Θs [(1 + 4αs g0. and appears in the expression for the solids pressure and viscosities. Θq ) d (23. The function g0. From [118]. however there is a need to provide a better formulation so that some properties may feel the presence of other phases.544) where ess is the coefficient of restitution for particle collisions. A known problem is that N solids phases with identical properties should be consistent when the same phases are described by a single solids phase.ss (described below in more detail) is a distribution function that governs the transition from the “compressible” condition with α < αs. ∇ps . the above expression does not take into account the effect of other phases.max . to the “incompressible” condition with α = αs.ss Θs s and [226] 1 [(1 + ess )(1 − ess + 2µf ric )]] (23. Because a Maxwellian velocity distribution is used for the particles. mq . where no further decrease in the spacing can occur.max . and will be described later in this section. Other formulations that are also available in FLUENT are [363] ps = 2ρs (1 + ess )α2 g0. A derivation of the expressions from the Boltzman equations for a granular mixture are beyond the scope of this manual.23.547) . where the spacing between the solid particles can continue to decrease.5 Solids Pressure For granular flows in the compressible regime (i.545) 3 g0. The granular temperature Θs is proportional to the kinetic energy of the fluctuating particle motion. but the value can be adjusted to suit the particle type.ss ) + pq = αq ρq Θq + N Xπ p=1 (23.63 is the default for αs. FLUENT uses a default value of 0. A value of 0.pqqp 3 nq np (1 + eqp )f (mp .ss is the radial distribution function.546) 2 When more than one solids phase are calculated. a solids pressure is calculated independently and used for the pressure gradient term. but you can modify it during the problem setup.e.
RadialDistri ution Function b The radial distribution function. s → 0 and g0 → ∞. For now.550 it can be observed that for a dilute solid phase s → ∞. χ is equal to 1 for a rare gas.qp )ρq Θq (23. np . The above expression reverts to the one solids phase expression when N = 1 and q = p but also has the property of feeling the presence of other phases.pq αq αp ρq Θq (23. it follows that pq = αq ρq Θq + ( N X d3 p=1 pq 3 dq pc. and therefore g0 → 1. nq are the number of particles. The radial distribution function is closely connected to the factor χ of Chapman and Cowling’s [51] theory of nonuniform gases.550) where s is the distance between grains. This function may also be interpreted as the nondimensional distance between spheres: g0 = s + dp s (23. . we have to simplify this expression so that it depends only on the granular temperature of phase q pq = αq ρq Θq + 2 N X d3 pq d +d d3 q p=1 (1 + epq )g0. is a correction factor that modifies the probabili ty of collisions between grains when the solid granular phase becomes dense.548) Since all models need to be cast in the general form.where dpq = p 2 q is the average diameter.qp is the collisional part of the pressure between phases q and p.549) where pc. From Equation 23. and f is a function of the masses of the particles and their granular temperatures. In the limit when the solid phase compacts. g0 . and increases and tends to infinity when the molecules are so close together that motion is not possible. mp and mq are the masses of the particles in phases p and q.
max This is an empirical function and does not extends easily to n phases.59α2 + 4.555) • The following equation [363] is available: 3( g0. For two identical phases with the property that αq = α1 + α2 .ll = 0.551) s.max ) 2 d k=1 k (23. use [267]: g0 = 1 − αs α ! 1 −1 3 (23.553) • The following expression is also available [151]: g0.52α3 s 1− αs.678 s (23. pq = p1 + p2 .554) • Also available [226].5αs + 4.ll = 1 − where αs α s. is the following: 1 + 2.556) 1 k=1 dk + dk dl (1 − αs ) (1 − α )2 (d + dk ) s j .In the literature there is no unique formulation for the radial distribution function. In order to correct this problem.kl = PN αk ) (23. max αs 3 N 1 X αk + d l 2 k=1 dk g0. N X k=1 αk (23.ll = N 1 3 X αk + dl αs (1 −αs. FLUENT uses the following consistent formulation: g0. the above function is not consistent for the calculation of the partial pressures p1 and p2 .max ! 1 −1 N 3 1 X αk + dl 2 k=1 dk (23. slightly modified for n solids phases. FLUENT has a number of options: • For one solids phase.552) αs = and k are solids phases only.
559) (23. 23.max + [1 ∗(α 1.556 reverts to the [47] derivation. while Equation 23.554 and Equation 23.555 compare well with [6] experimental data.554 and Equation 23.max d1 1.max ) (23.max d2 ](1 − α1.561) The packing limit is used for the calculation of the radial distribution function.max )α2.560) otherwise.When the number of solid phases is greater than 1.max (23.max − = (α 1.max )α2. For a binary mixture FLUENT uses the correlations proposed by [99]. .5.max − α2.max + (1 − α1.555 are extended to g0.ll + dl g0.max +α 2. the mixture composition is defined as X1 = αα1 2 1 +α where X1 <= α1.max )α2.6 M axi um Packing Lim itin BinaryM ixtures m The packing limit is not a fixed qua ntity and may change according to the number of particles present within a given volume and the diameter of the particles.558) (α1.557) It is interesting to note that equations Equation 23.max )(1 − X1 ) + α1.552.lm = dm g0. For a binary mixture with diameters d1 > d2 . the maximum packing limit for the binary mixture is s [1 − d2 ](α + (1 − α1.max + (1 − α1. Equation 23.max ) d1 X )α2. Small particles accumulate in between larger particles increasing the packing limit.max The maximum packing limit for the mixture is given by s αs.max (23.max ) 1 α1.mm dm + dl (23. Equation 23.
563) KineticViscosity FLUENT provides two expressions for the kinetic part.col = 4 αs ρs ds g0. The collisional and kinetic parts.ss (1 + ess ) Θs 3 π 1/2 (23.col + µs. expression or use a userdefined function.kin √ 10ρs ds Θs π 4 = 1 + g0.566) Note that the bulk viscosity is set to a consta nt value of zero.5.ss (1 + ess ) Θs 5 π 1/2 (23. and the optional frictional part. [119] is also available: µs. The default expression is from Syamlal et al.562) CollisionalViscosity The collisional part of the shear viscosity is modeled as [119. 363] µs.565) Bulk Viscosity The solids bulk viscosity accounts for the resistance of the granular particles to compression and expansion.kin + µs. [363]: µs. by default. It is also possible to select the Lun et al.fr (23. [221]: λs = 4 αs ρs ds g0.ss 5 (23. are added to give the solids shear viscosity: µs = µs.564) The following optional expression from Gidaspow et al. A frictional component of viscosity can be included to account for the viscousplastic transition that occurs when particles solid phase reach the maximum solid volume fraction.7 Solids Shear Stresses The tum also of a solids stress tensor contains shear and bulk viscosities arising from particle momenexchange due to translation and collision. It has the following form from Lun et al.ss αs (1 + ess ) ) g0.23.kin √ αs ds ρs Θs π 2 = 6 (3 − e ) 1 + (1 + ess ) (3ess − 1) αs ss g0.ss 5 96αs (1 + ess 2 (23. .
.569) (23. The application of the modified Coulomb law leads to an expression of the form µf riction = Pf riction sin φ √ 2 I2D (23. FLUENT extends the formulation of the frictional viscosity and employs other models. φ is the angle of internal friction. The application of kinetic theory to granular flows is no longer relevant since particles are in contact and the resulting frictional stresses need to be taken into account.5 when the flow is threedimensional and the maximum packing limit is about 0. See the separate UDF Manual for details.567) where ps is the solids pressure. and I2D is the second invariant of the deviatoric stress tensor.571) Where φ is the angle of internal friction and I2D is the second invariant of the deviatoric stress tensor. instantaneous collisions are less importa nt. The solids shear viscosity computed by FLUENT does not. by default.Frictional Viscosity In dense flow at low shear. In granular flows with high solids volume fraction. The frictional stresses are usually written in Newtonian form: ~ T τf riction = −Pf icti on I~+ µ riction (∇~us + (∇~u ) ) r f s (23.fr = √ 2 I2D (23. Then PS = Pkinetic + Pf riction µS = µkinetic + µf riction (23.568) The frictional stress is added to the stress predicted by the kinetic theory when the solids volume fraction exceeds a critical value. This value is normally set to 0. as well as providing new hooks for UDFs. FLUENT uses Schaeffer’s [318] expression: ps sin φ µs. It is also possible to specify a consta nt or userdefined frictional viscosity. where the secondary volume fraction for a solid phase nears the packing limit. while the frictional viscosity can be derived from the first principles. the generation of stress is mainly due to friction between particles. If the frictional viscosity is included in the calculation.570) The derivation of the frictional pressure is mainly semiempirical. account for the friction between particles.63.
574) (23. n=2 and p = 3 [266]. The radial distribution function is an importa nt parameter in the description of the solids pressure resulting from granular kinetic theory. however a complete description would require the introduction of more physics to capture the elastic regime with the calculation of the yield stress and the use of the flowrule.Two additional models are available in FLUENT: the Johnson and Jackson [165] model for frictional pressure and Syamlal et al [363].5.573) The second model that is employed is Syamlal et al [363].mi n )n Pf riction = F r s (α s.564. described in Equation 23. i The introduction of the frictional viscosity helps in the description of fric. The Johnson and Jackson [165] model for frictional pressure is defined as (α − αs. as it has the desired effect.572) With coefficient Fr = 0. [221] or Gidaspow [118] the radial function tends to infinity as the volume fraction tends to the packing limit. . These effects can be added by the user via UDFs to model static regime.max − αs )p (23.1αs The frictional viscosity for this model is of the form µf riction = Pf riction sin φ (23. Comparing the two models results in the frictional normal stress differing by orders of magnitude. It would then be possible to use this pressure directly in the calculation of the frictional viscosity. The critical value for the solids volume fraction is 0.05. The coefficient Fr was modified to make it a function of the volume fraction: F r = 0. This approach is also available in FLUENT by default.tional flows. Small time steps are required to get good convergence behavior. If we use the models of Lun et al.
576) where η= 1 (1 + ess ) 2 FLUENT uses the following expression if the optional model of Gidaspow et al. kΘs is given by √ 15ds ρs αs Θs π 2 16 (41 − 33η)ηαs g0.23.ss + 5 15π ) k Θs (23. [221] γΘm = 12(1 − e2 )g0. γΘs . [119] is enabled: q s kΘs 2 150ρs ds (Θπ) 6 2 = 1 + αs g0. When the default Syamlal et al.ss Θs π (23. model [363] is used.ss 5 384(1 + ess )g0. the diffusion coefficient for granular energy. This term is represented by the expression derived by Lun et al.5.ss = 4(41 − 33η) 1 + η2 (4η − 3)αs g0.577) The collisional dissipation of energy.ss (1 + es ) + 2ρs αs ds (1 + ess )g0.578) ss .575 contains the term kΘs ∇Θs describing the diffusive flux of granular energy. The trans port equation derived from kinetic theory takes the form [79] " # 3 ∂ 2 ∂t where (ρs αs Θs ) + ∇ · (ρs αs~vs Θs = (−ps I +τ s ) : ∇~vs +∇·(kΘs ∇Θs )−γΘs +φls (23.8 G ranular Temperature The granular temperature for the sth solids phase is proportional to the kinetic energy of the random motion of the particles.ss 2 3/2 √ ρs αs Θs ds π (23. represents the rate of energy dissipation within the sth solids phase due to collisions between particles.5) 75) (−ps I + τ s ) : ∇~vs = kΘs ∇Θs = γ Θs = φls = the generation of energy by the solid stress tensor the diffusion of energy (kΘs is the diffusion coefficient) the collisional dissipation of energy the energy exchange between the lth fluid or solid phase and the sth solid phase Equation 23.
and g0 is the radial distribution function that is model dependent. See Chapter 7: Boundary Conditions for details.579) FLUENT allows the user to solve for the granular temperature with the following options: • algebraic formulation (the default) It is obtained by neglecting convection and diffusion in the trans port equation. defined by Z Hq = cp. φ is the specularity coefficient between the particle and the wall. • consta nt granular temperature This is useful in very dense situations where the random fluctuations are small. we may write the shear force at the wall in the following form: q π√ αs ~ (23.q is the specific heat at consta nt pressure of phase q. 3 s (1 − esw )ρs g0 Θ2 s αs.max Here U~s. The general boundary condition for granular temperature at the wall takes the form [165] qs = q 3 π√ α π√ α 2 3φ s ρs g0 Θs )U~s. τ~ =− 3φ s 6 αs.9 Descriptionof HeatTransfer The internal energy balance for phase q is written in terms of the phase enthalpy.575 and it is allowed to choose different options for it properties. · U~s. • partial differential equation This is given by Equation 23. Equation 23.580) ρs g0 Θs Us.q dTq (23.582) where cp. .511.581) 23. Equation 23. αs. is the particle slip velocity parallel to the wall.575 [363]. • UDF for granular temperature For a granular phase s.max is the volume fraction for the particles at maximum packing. The thermal boundary conditions used with multiphase flows are the same as those for a singlephase flow.max αs.5.The transfer of the kinetic energy of random fluctuations in particle velocity from the sth solids phase to the lth fluid or solid phase is represented by φls [119]: φls = −3Kls Θs (23.max 6 − 4 (23.
584.586) In the case of granular flows (where p = s). applicable to a porosity range of 0. and Pr is the Prandtl number of the q th phase: Pr = cp q µ q κq (23. Nup . The Nusselt number is typically determined from one of the many correlations reported in the literature.4αf + 1.2α2 )Re 0.0 and a Reynolds number of up to 105 : Nus = (7 − 10αf + 5α2 )(1 + 0.585) where Rep is the relative Reynolds number based on the diameter of the pth phase and the relative velocity u~p − u~q . To enforce this. by hpq = 6κq αp αq Nu p dp 2 (23. FLUENT uses the correlation of Ranz and Marshall [295. The heat transfer coefficient is related to the pth phase Nusselt number.33 − 2. In the case of fluidfluid multiphase.The Heat Exchange Coefficient The rate of energy transfer between phases is assumed to be a function of the temperature difference Qpq = hpq (Tp − Tq ) (23.35–1.583) where hpq (= hqp ) is the heat transfer coefficient between the pth phase and the q th phase.6Rep1/2 Pr 1/3 (23.7 Pr 1/3 (23. 296]: Nup = 2.584) Here κq is the thermal conductivity of the q th phase. hpq should tend to zero whenever one of the phases is not present within the domain. .587) f s f s The Prandtl number is defined as above with q = f .2 Pr 1/3 ) + (1.0 + 0. as reflected in Equation 23.7Re0. FLUENT uses a Nusselt number correlation by Gunn [128]. hpq is always multiplied by the volume fraction of the primary phase q. For all these situations.
FLUENT provides three methods for modeling turbulence in multiphase flows within the context of the k. The k.23. In comparison to singlephase flows. as described in Chapter 12: Modeling Turbulence.models. In addition. the number of terms to be modeled in the momentum equations in multiphase flows is large. the dispersed turbulence model.TurbulenceModels FLUENT provides three turbulence model options in the context of the k. and this makes the modeling of turbulence in multiphase simulations extremely complex.5.turbulence model options are: • mixture turbulence model (the default) • dispersed turbulence model • turbulence model for each phase i Note that the descriptions of each method below are presented based on the standard k. the choice of model depends on the importance of the secondaryphase turbulence in your application. The RSM turbulence model options are: • mixture turbulence model (the default) • dispersed turbulence model For either category.10 TurbulenceModels To describe the effects of turbule nt fluctuations of velocities and scalar quantities in a single phase. FLUENT provides two turbulence options within the context of the Reynolds stress models (RSM). FLUENT uses various types of closure models. and are therefore not presented explicitly. k.models are similar.model.models: the mixture turbulence model (the default). The multiphase modifications to the RNG and realizable k. . or a perphase turbulence model.
589) (23.590) αi ρi~vi (23.m = µt.Model for the singlephase k.k. .m = ρm Cµ k2 (23. for stratified (or nearly stratified) multiphase flows.591) αi ρi ~vm = i=1 N X i=1 the turbule nt viscosity. µt.m . and it is applicable when phases separate.m − ρm (ρm k) + ∇ · (ρm~vm k) = ∇ µt.M ixture Turbulence Model The mixture turbulence model is the default multiphase turbulence model.m − C2 ρm ) (23. In these cases.m (∇~vm + (∇~vm ) T ) : ∇~vm (23. and when the density ratio between phases is close to 1. Gk. are computed from ρm = and N X N X i=1 αi ρ i (23.m ∇ ∂t σk k · and ∂ µt.model.4. is computed from Gk.593) The constants in these equations are the same as those described in Section 12. It represents the first extension of the singlephase k.m . using mixture properties and mixture velocities is sufficient to capture importa nt features of the turbule nt flow.588) k where the mixture density and velocity.m (ρm ) + ∇ · (ρm~vm ) = ∇ ∇ ∂t σ · + (C1 Gk. The k and equations describing this model are as follows: ∂ + Gk. is computed from µt. ρm and ~vm .model.1: Standard k.592) and the production of turbulence kinetic energy.
• a phaseweighted averaging process The choice of averaging process has an impact on the modeling of dispersion in turbule nt multiphase flows.model for the continuous phase Turbule nt predictions for the continuous phase are obtained using the standard k. It is possible to take into account the dispersion of the dispersed phases trans ported by the turbule nt fluid motion. turbule nt fluctuations in the volume fractions do not appear.k. The model is applicable when there is clearly one primary continuous phase and the rest are dispersed dilute secondary phases. A twostep averaging process leads to the appearance of fluctuations in the phase volume fractions. When the twostep averaging process is used with a phaseweighted average for the turbulence. • Tchentheory correlations for the dispersed phases Predictions for turbulence quantities for the dispersed phases are obtained using the Tchen theory of dispersion of discrete particles by homogeneous turbulence [141]. interparticle collisions are negligible and the domina nt process in the random motion of the secondary phases is the influence of the primaryphase turbulence. Assumptions The dispersed method for modeling turbulence in FLUENT assumes the following: • a modified k. • interphase turbule nt momentum transfer In turbule nt multiphase flows. In this case. Fluctuating quantities of the secondary phases can therefore be given in terms of the mean characteristics of the primary phase and the ratio of the particle relaxation time and eddyparticle interaction time. FLUENT uses phaseweighted averaging.Dispe rsed TurbulenceModel The dispersed turbulence model is the appropriate model when the conce ntrations of the secondary phases are dilute. the momentum exchange terms contain the correlation between the instantaneous distribution of the dispersed phases and the turbule nt fluid motion. . however.model suppleme nted with extra terms that include the interphase turbule nt momentum transfer. so no volume fraction fluctuations are introduced into the continuity equations.
q = where q 3 kq Cµ 2 q (23.596) is the dissipation rate and Cµ = 0. Reynolds stress tensor for continuous phase q takes the following form: T 2 0 τ q 0= − (ρq kq + ρq µt.q = 3/2 3 kq Cµ 2 q (23.597) Turbule nt predictions are obtained from the modified k.q ∇ · U~ q )I + ρq µt.q ∇ ) + α q q q (C G − C2 ρq q ) + αq ρq Π q (αq ρq q ) + ∇ · (αq ρq U~ q q ) = ∇ · kq 1 k.q q q (αq ρq kq ) + ∇ · (αq ρq U~ q kq ) = ∇ · ∂t σk (αq and ∂ µt.eling Turbule nt Production in the k.q (∇U~ q ) 3 + ∇U~ q (23.q is the production of turbule nt kinetic energy.Turbulence in the Co ntinuous Phase The The eddy viscosity model is used to calculate averaged fluctuating quantities. and Gk. The length scale of the turbule nt eddies is s Lt.q ∂t σ (23.595) and a characteristic time of the energetic turbule nt eddies is defined as τt.q ∇k ) + α G − α ρ q q k.599) (αq Here Πkq and Π q represent the influence of the dispersed phases on the continuous phase q.model. q + αq ρq Πkq (23.09.598) .q is written in terms of the turbule nt kinetic energy of phase q: µt.q = ρq Cµ 2 kq q (23. The turbule nt viscosity µt.Models.model: ∂ µt.4: Mod. All other terms have the same meaning as in the singlephase k.4. as defined in Section 12.594) where U~ q is the phaseweighted velocity.
The term Πkq can be derived from the insta ntaneous equation of the continuous phase and takes the following form.5103) The Lagrangian integral time scale calculated along particle trajectories.5105) (23. ~vpq is the relative velocity.pq = q (1 + Cβ ξ 2 ) where ξ= τt.q (23. [95]: Π q = C3 where C3 = 1.pq = −1 αp ρq K pq ρp + CV ρq ! (23. and the turbule nt kinetic energy of each dispersed phase. Π q is modeled according to Elgobashi et al.5104) . and ~vdr is the drift velocity (defined by Equation 23.2.q τt.5114 below). mainly affected by the crossingtr ajectory effect [69].5109 below).q ~vpq  Lt.5102) Time and length scales that characterize the motion are used to evaluate dispersion coefficients. correlation functions. Turbulence in the Dis persed Phase q kq Πkq (23. The characteristic particle relaxation time connected with inertial effects acting on a dispersed phase p is defined as τF.5100) q α q ρq (kpq − 2kq + ~vpq · ~vdr ) (23. is defined as τt.5101) where klq is the covariance of the velocities of the continuous phase q and the dispersed phase l (calculated from Equation 23. where M represents the number of secondary phases: Πkq = which can be simplified to Πkq = M X Kpq p=1 M X Kpq p=1 α q ρq (< ~v 0 0 ·p ~v 0 0 > +( U~ p − U~ q ) · ~vdr ) (23.
FLUENT assumes Dp = Dq = Dt. and σpq is a dispersion Prandtl number.57) is modeled as follows.5113) The second term on the righthand side of Equation 23.5113 contains the drift velocity: Dq ~vdr = − ∇αp − ∇αq σpq αq σpq αp Dp ! (23.5112) CV and CV = 0. Interphase Turbule nt M e ntum Transfer om The turbule nt drag term for multiphase flows (Kpq (~vp −~vq ) in Equation 23.pq τF.5106) where θ is the angle between the mean particle velocity and the mean relative velocity.5114) Here Dp and Dq are diffusivities. When using Tchen theory in multiphase flows.and Cβ = 1.pq + 2 kp − 3 ρ b = (1 + CV ) p + ρq 1 kpq τt.pq 3 (23. .8 − 1.pq (23.pq = (23.5108) ! b + ηpq 1 + ηpq (23.5107) Following Simonin [333].5109) (23.5 is the addedmass coefficient.75. FLUENT writes the turbulence quantities for dispersed phase p as follows: b2 + ηpq 1 + ηpq ! kp = kq kpq = 2kq Dt.pq 3 !−1 Dp = Dt.35 cos2 θ (23. The ratio between these two characteristic times is written as ηpq = τt. for dispersed phase p and continuous phase q: Kpq (~vp − ~vq ) = Kpq (U~ p − U~ q ) − Kpq~vdr (23.5111) (23.5110) 1 b kpq τF.pq and the default value for σpq is 0.
The drift velocity results from turbule nt fluctuations in the volume fraction. When multiplied by the exchange coefficient Kpq , it serves as a correction to the momentum exchange term for turbule nt flows. This correction is not included, by default, but you can enable it during the problem setup. You can enable the effect of drift velocity by performing the following: 1. If it is not already done, set the kepsilon Multiphase Model to Dispersed in the Viscous panel. 2. Enter the multiphaseoptions text command in the console window. define −→ models −→ viscous −→ multiphaseturbulence −→ multiphaseoptions
/define/models/viscous/multiphaseturbulence> multiphaseEnable dispersion force in momentum? yes options [no] Enable interphase kepsilon source? yes [no]
The effect of the drift velocity is influenced both by the momentum equation and, to a lesser extent, the turbulence equation. Therefore, you should answer yes to both questions to take into account the effect of drift velocity. k TurbulenceM odelfor Each Phase The most general multiphase turbulence model solves a set of k and trans port equations for each phase. This turbulence model is the appropriate choice when the turbulence transfer among the phases plays a domina nt role. Note that, since FLUENT is solving two additional trans port equations for each secondary phase, the perphase turbulence model is more computationally intensive than the dispersed turbulence model. Trans port Equations The Reynolds stress tensor and turbule nt viscosity are computed using Equations 23.594 and 23.595. Turbulence predictions are obtained from ∂ µt,q ∇k ) + (α G − α ρ ) + q q k,q q q q (αq ρq kq ) + ∇ · (αq ρq U~ q kq ) = ∇ · ∂t σk (αq
X
N N
Klq (Clq kl − Cql kq ) − − U~ q ) ·
l=1 l=1
X
Klq (U~ l
µt,l αl σl
∇αl + U~ q ) ·
X
N
Klq (U~ l −
µt,q αq σq
∇αq (23.5115)
l=1
and ∂ µt,q ∇ ) + q q (αq ρq q ) + ∇ · (αq ρq U~ q q ) = ∇ · kq ∂t σ (αq C3
N X
"
C1 αq Gk,q − C2 αq ρq
q
+
! #
Klq (Clq kl − Cql kq ) − U~ q ) ·
l=1
NX
Klq (U~ l −
µt,l αl σl
∇αl + q) ·
N X
µt,q Klq (U~ l − U~ ∇αq
l=1
l=1
αq σ q (23.5116)
The terms Clq and Cql can be approximated as ηlq Clq = 2, Cql = 2 1 + ηlq where ηlq is defined by Equation 23.5107. Interphase Turbule nt M e ntum Transfer om
!
(23.5117)
The turbule nt drag term (Kpq (~vp − ~vq ) in Equation 23.57) is modeled as follows, where l is the dispersed phase (replacing p in Equation 23.57) and q is the continuous phase:
N X l=1
Klq (~vl − ~vq ) =
l=1
N
X
Klq (U~ l − U~ q ) −
l=1
N
X
Klq~vdr,lq
(23.5118)
Here U~ and U~ are phaseweighted velocities, and ~v l q dr,lq is the drift velocity for phase l (computed using Equation 23.5114, substituting l for p). Note that FLUENT will compute the diffusivities Dl and Dq directly from the trans port equations, rather than using Tchen theory (as it does for the dispersed turbulence model). As noted above, the drift velocity results from turbule nt fluctuations in the volume fraction. When multiplied by the exchange coefficient Klq , it serves as a correction to the momentum exchange term for turbule nt flows. This correction is not included, by default, but you can enable it during the problem setup. The turbulence model for each phase in FLUENT accounts for the effect of the turbulence field of one phase on the other(s). If you want to modify or enhance the interaction of the multiple turbulence fields and interphase turbule nt momentum transfer, you can supply these terms using userdefined functions.
RSM Turbulence Models
Multiphase turbulence modeling typically involves two equation models that are based on singlephase models and often cannot accurately capture the underlying flow physics. Additional turbulence modeling for multiphase flows is diminished even more when the basic underlying singlephase model cannot capture the complex physics of the flow. In such situations, the logical next step is to combine the Reynolds stress model with the multiphase algorithm in order to handle challenging situations in which both factors, RSM for turbulence and the Eulerian multiphase formulation, are a precondition for accurate predictions [65]. The phaseaveraged continuity and momentum equations for a continuous phase are: ∂ (α ρ ) + ∇ · (αc ρc ˜ c ) = 0 U ∂t c c (23.5119)
O ∂ (αc ρrmc U˜c ) + ∇ · (αc ρrmc U˜c U˜c ) = −αc ∇p˜ + ∇ · τ˜t + FDc (23.5120) c ∂t
For simplicity, the laminar stressstrain tensor and other body forces such as gravity have been omitted from Equations 23.511923.5120. The tilde denotes phaseaveraged variables while an overbar (e.g., αc ) reflects timeaveraged values. In general, any variable Φ can have a phaseaverage value defined as ˜ Φc = α c Φc αc (23.5121)
Considering only two phases for simplicity, the drag force between the continuous and the dispersed phases can be defined as:
"
FDc = Kdc
α u0 (U˜d − U˜c ) d d− − αd
0 cu c
! #
(23.5122)
αc
where Kdc is the drag coefficient. Several terms in the Equation 23.5122 need to be modeled in order to close the phaseaveraged momentum equations. Full descriptions of all modeling assumptions can be found in [64]. This section only describes the different modeling definition of the turbule nt stresses τ˜t that appears in Equation 23.5120. The turbule nt stress that appears in the momentum equations need to be defined on a perphase basis and can be calculated as: τ˜tk = −αk ρk R˜ k,ij (23.5123)
where the subscript k is replaced by c for the primary (i.e., continuous) phase or by d for any secondary (i.e., dispersed) phases. As is the case for singlephase flows, the current multiphase Reynolds stress model (RSM) also solves the trans port equations for Reynolds stresses Rij . FLUENT includes two methods for modeling turbulence in multiphase flows within the context of the RSM model: the dispersed turbulence model, and the mixture turbulence model. R SMDispe rsed TurbulenceModel The dispersed turbulence model is used when the concentrations of the secondary phase are dilute and the primary phase turbulence is regarded as the domina nt process. Consequently, the trans port equations for turbulence quantities are only solved for the primary (continuous) phase, while the predictions of turbulence quantities for dispersed phases are obtained using the Tchen theory. The trans port equation for the primary phase Reynolds stresses in the case of the dispersed model are:
! " #
∂ (αρRij ) + ∂ (αρU˜ R˜ ) = ˜ k ij ∂t ∂xk −αρ
R
k
˜ ˜ ∂U +
ij
˜ ∂xk R jk
∂ ∂U˜i ∂ + αµ (R˜ ij ) ∂xk ∂xk ∂xk
∂ ∂u0i ∂uj0 0 0 0 − [αρui uj u ] + αp( + ) k ∂xk ∂xj ∂xi − αρ ˜ij + ΠR,ij
(23.5124)
The variables in Equation 23.5124 are per continuous phase c and the subscript is omitted for clarity. The last term of Equation 23.5124, ΠR,ij , takes into account the interaction between the continuous and the dispersed phase turbulence. A general model for this term can be of the form: ΠR,ij = Kdc C1,dc (R dc,ij − Rc,ij ) + Kdc C2,dc adc,i bdc,j (23.5125)
where C1 and C2 are unknown coefficients, adc,i is the relative velocity, bdc,j represents the drift or the relative velocity, and Rdc,ij is the unknown particulatefluid velocity correlation. To simplify this unknown term, the following assumption has been made: ΠR,ij = 2 δij Πk 3 (23.5126)
where δij is the Kronecker delta, and Πk represents the modified version of the original Simonin model [333]. Πkc = Kdc (k˜dc − 2k˜c + V˜rel · V˜drift ) (23.5127)
but in a segregated fashion. The modeling of ˜ together with all other unknown terms in Equation 23.where K˜ c represents the turbule nt kinetic energy of the continuous k˜dc is phase. The mixture density. i=1 PN αi ρi U˜i i=1 αi ρi (23. R SMM ixture Turbulence Model The main assumption for the mixture model is that all phases share the same turbulence field which consequently means that the term ΠR in the Reynolds stress trans port equations (Equation 23. the ˜rel and V˜drift stand for continuousdispersed phase velocity covariance and finally. can be expressed as ρm = N X i=1 αi ρi (23.5130) l=1 . The velocities are solved coupled by phases.5128) while mixture velocities can be expressed as PN U˜ m = where N is the number of species. The block algebraic multigrid scheme used by the densitybased solver described in [394] is used to solve a vector equation formed by the velocity components of all phases simultaneously.5129) 23.5. the pressurecorrection equation takes the form n X 1 ( ∂ ∂t k=1 ρrk α ρ + ∇ · α ρ 0 + ∇ · α ρX − ~v ( (m˙ k k k ~v k k k k k ∗ n lk ) − m˙ )) kl =0 (23. FLUENT uses the phase coupled SIMPLE (PCSIMPLE) algorithm [379] for the pressurevelocity coupling. Apart from that. the equations maintain the same form but with phase properties and phase velocities being replaced with mixture properties and mixture velocities. Then. V the relative and the drift velocities.11 Solution Methodin FLUENT For Eulerian multiphase calculations. the trans port equation for the turbule nt kinetic energy dissipation rate ( ˜) is required. a pressure correction equation is built based on total volume continuity rather than mass continuity. PCSIMPLE is an extension of the SIMPLE algorithm [276] to multiphase flows. The PressureCorrection Equation For incompressible multiphase flow. In order to achieve full closure. respectively.5127 are modeled in the same way as in [64]. Pressure and velocities are then corrected so as to satisfy the continuity constrai nt. for example.5124) is neglected.
Modeling wet steam is very importa nt in the analysis and design of steam turbines. VolumeFractions The volume fractions are obtained from the phase continuity equations.5132) 23. k is the velocity correction for the k th phase.k αk = X nb (anb. a condensation process will take place shortly after the state path crosses the vaporsaturation line. The increase in steam turbine exit wetness can cause severe erosion to the turbine blades at the lowpressure stages.13: Setting Up the Wet Steam Model. The expansion process causes the superheated dry steam to first subcool and then nucleate to form a twophase mixture of saturated vapor and fine liquid droplets known as wet steam.5131) In order to satisfy the condition that all the volume fractions sum to one. the equation of the k th volume fraction is ap. and the number of liquiddroplets per unit volume (η). and a reduction in aerodynamic efficiency of the turbine stages operating in the wet steam region [250]. Information about enabling the model and using your own property functions and data with the wet steam model is provided in Section 23. which involves the formation of liquiddroplets in a homogeneous nonequilibrium condensation process. and ~v ∗ is k ~v 0 the value of ~vk at the current iteration.6 Wet Steam M odelTheory 23.k αnb. Solution settings .1 Overview and Limitations of the Wet Steam Model Overview During the rapid expansion of steam. is based on the classical nonisothermal nucleation theor y. in addition to two trans port equations for the liquidphase massfraction (β). In discretized form. The flow mixture is modeled using the compressible NavierStokes equations. n X k=1 αk = 1 (23.6. This section describes the theoretical aspects of the wet steam model.M odelingM ultiphase Flows where ρrk is the phase reference density for the k th phase (defined as the total volume average density of phase k). FLUENT has adopted the EulerianEulerian approach for modeling wet steam flow.k ) + bk = Rk (23. The velocity corrections are themselves expressed as functions of the pressure corrections. The phase change model.
• The mass fraction of the condensed phase. it is assumed that the volume of the condensed liquid phase is negligible.14. is small (β < 0. From the preceding assumptions. The primary phase is the gaseousphase consisting of watervapor (denoted by the subscript v) while the secondary phase is the liquidphase consisting of condensedwater droplets (denoted by the subscript l).5: Wet Steam Model. massflow inlet. • When the wet steam model is active. and pressure outlet are the only inflow and outflow boundary conditions available. • Since droplet sizes are typically very small (from approximately 0.2). Therefore.6. • The interactions between droplets are neglected. Postprocessing variables are described in Section 23. then it must be done in the Materials panel before activating the wet steam model.1: ModelSpecific Variables . 23. β (also known as wetness factor). The following assumptions are made in this model: • The velocity slip between the droplets and gaseousphase is negligible. Limitations The following restrictions and limitations currently apply to the wet steam model in FLUENT: • The wet steam model is available for the densitybased solvers only. it follows that the mixture density (ρ) can be related to the vapor density (ρv ) by the following equation: ρ= ρv (1 − β) (23.61) .1 microns to approximately 100 microns). • Pressure inlet.15. if solid properties need to be set and adjusted. the access to the Materials panel is restricted because the fluid mixture properties are determined from the built in steam prop.2 Wet Steam Flow Equations The wet steam is a mixture of twophases.erty functions or from the userdefined wet steam property functions.and strategies for the wet steam model can be found in Section 23.
Equation 23.64 form a closed system of equations which.61. Equation 23.66) .v. Together.63) where Γ is the mass generation rate due to condensation and evaporation (kg per unit volume per second).65) 3 (23. along with Equation 23.In addition.w.61 and the average droplet volume Vd are combined in the following expression: β η =(1 − β)Vd (ρl /ρ v ) where ρl is the liquid density and the average droplet volume is defined as Vd = where rd is the droplet radius. The flow equations are solved using the same densitybased solver algorithms employed for general compressible flows.62) where Q=( P. permit the calculation of the wet steam flow field. The mixture flow is governed by the compressible NavierStokes equations given in vector form by Equation 25. 3 πrd (23.u.64) where I is the nucleation rate (number of new droplets per unit volume per second). and Equation 23.54: ∂W ∂ ∂Q ∂t Z V I Z Q dV + [F − G] · dA = H dV V (23. To model wet steam.63.62. two additional trans port equations are needed [152]. The second trans port equation models the evolution of the number density of the droplets per unit volume: ∂ρη + ∇ · (ρ−→v η) = ρI ∂t (23. the temperature and the pressure of the mixture will be equivalent to the temperature and pressure of the vaporphase. To determine the number of droplets per unit volume.T) are mixture quantities. Equation 23. The first trans port equation governs the mass fraction of the condensed liquid phase (β): ∂ρβ + ∇ · (ρ−→v β) = Γ ∂t (23.
.e. when the state path crosses the saturated. ρl is the condensed liquid density (also evaluated at temperature T ). Therefore. and the supersaturation ratio S can take on values greater than one.23.6. r∗ = 2σ ρ RT ln S l (23. and r∗ is the KelvinHelmholtz critical droplet radius. An expression for r∗ is given by [410].3 Phase Change Model The following is assumed in the phase change model: • The condensation is homogeneous (i. • The droplet is assumed to be spherical. and S is the super saturation ratio defined as the ratio of vapor pressure to the equilibrium saturation pressure: s= P Psat (T ) (23. The mass generation rate Γ in the classical nucleation theory during the nonequilibrium condensation process is given by the sum of mass increase due to nucleation (the formation of critically sized droplets) and also due to growth/demise of these droplets [152]. • The droplet is surrounded by infinite vapor space.69) The expansion process is usually very rapid. • The droplet growth is based on average represe ntative mean radii. Γ is written as: Γ= 4 ∂r πρl I r∗ 3 + 4πρl ηr2 3 ∂t (23.68) where σ is the liquid surface tension evaluated at temperature T . • The heat capacity of the fine droplet is negligible compared with the late nt heat released in condensation. . no impurities present to form nuclei).67) where r is the average radius of the droplet.vapor line. the process will depart from equilibrium. Therefore. above which the droplet will grow and below which the droplet will evaporate.
611) where qc is evaporation coefficient.5 RT ! (23.4 BuiltinThermo dynamic Wet Steam Properties There are many equations that describe the thermodynamic state and properties of steam. is given by: I= qc ρv 2 (23. kb is the Boltzmann consta nt. Mm is mass of one molecule. FLUENT uses a simpler form of the thermodynamic state equations [409] for efficient CFD calculations that are accurate over a wide range of temperatures and pressures.612) where hlv is the specific enthalpy of evaporation at pressure p and γ is the ratio of specific heat capacities. . the transfer of mass from the vapor to the droplets and the transfer of heat from the droplets to the vapor in the form of late nt heat.6. The nucleation rate described by the steadystate classical homogeneous nucleation theory [410] and corrected for nonisothermal effects. These equations are described below. 23. and ρl is the liquid density at temperature T . This energy transfer relation was presented in [408] and used in [152] and can be written as: ∂r γ + 1 C (T − T ) √P = p 0 ∂t hlv ρl 2πRT 2γ where T0 is the droplet temperature. The classical homogeneous nucleation theory describes the formation of a liquidphase in the form of droplets from a supersaturated phase in the absence of impurities or foreign particles. While some of these equations are accurate in generating property tables. A nonisothermal correction factor.The condensation process involves two mechanisms.610) !s 2σ (1 + θ) ρl e Mm 3 π − 2 4 π r∗ σ 3 bT K (23. they are not suitable for fast CFD computations. Therefore. is given by: 2(γ − 1) hlv θ= (γ + 1) RT ! hlv − 0. θ. σ is the liquid surface tension.
and b= 1.0015. The vapor isobaric specific heat capacity C pv is given by: C 2 C pv = C p0 (T ) + R [(1 − αv T )(B − B1 ) − B2 ] ρv (1 − 2αv T )C + αv T C1 + − 2 ρv 2 (23.616) The vapor specific enthalpy.000942. τ =647. a1 = 0.0.618) 2 v The isobaric specific heat at zero pressure is defined by the following empirical equation: . hv is given by: hv = h0 (T ) + RT (B − B1 )ρv + (C − 1 2 )ρv 2 (23.286 with T given in Kelvin. τ = 1500 with T given in Kelvin. is given by [409]: P = ρv RT (1 + Bρv + C ρv 2 ) (23.615) T where C is given in m6 /kg 2 .0004882. α = 10000.8978.613) where B.5E06. The two empirical functions that define the virial coefficients B and C cover the temperature range from 273 K to 1073 K.16. which relates the pressure to the vapor density and the temperature. a= 1. C = a(τ − τ0 )e −ατ + b (23. and a3 = 0. α=11. sv is given by: sv = s0 (T ) − R ln ρv + (B + B1 )ρv + ( C + C1 ) ρ 2 " # (23. τo = 0.Equationof State The steam equation of state used in the solver.772.614) where B is given in m3 /kg. T a2 = 0.617) The vapor specific entropy. and C are the second and the third virial coefficients given by the following empirical functions: B = a (1 + −1 1 τ α ) + a2 eτ(1 − e−τ ) 2 + a3 τ 5 (23.
619) .C p0 (T ) = X 6 i=1 ai T i−2 (23.
38930E04.0. The liquid surface tension equation was obtained from [408]. µl and K tl . a5 = 2. (23.47276. The example provided in Section 23. While the values of C pl . and C2 = T 2 dT 2 . a4 = 2. a1 = 46.13.620) s0 (T ) = where hc and sc are arbitrary consta nts. µl and K tl were curve fit using published data from [91] and then written in polynomial forms. SaturatedLiquidLine At the saturated liquidline. 2 2 Both h0 (T ) and s0 (T ) are functions of temperature and they are defined by: Z h0 (T ) = Z C p0 dT + hc C p0 dT + s c T (23.where C p0 is in KJ/kg K. the liquid density. and a6 = 9. and dB dC dB dC B1 = T dT . ρl . The example provided in Section 23. s. C1 = T dT . C p. a3 = 8. . B2 = T 2dT 2 .622) where φ represents any of the following thermodynamic properties: h.5: UDWSPF Example contains functions called wetst cpl(). surface tension. and thermal conductivity must be defined. was obtained from [301].5: UDWSPF Example contains a function called wetst satP() that represents the formulation for the saturation pressure. and wetst ktl() that represent formulations for C pl .46619E10. C v.13.19989E07. µ or K t.621) The vapor dynamic viscosity µv and thermal conductivity K tv are also functions of temperature and were obtained from [408].70466E14. M ixture Properties The mixture properties are related to vapor and liquid properties via the wetness factor using the following mixing law: φm = φl β + (1 − β)φv (23. The equation for liquid density. a2 = 1. wetst mul(). specific heat C p. SaturatedVapor Line The saturation pressure equation as a function of temperature was obtained from [301]. dynamic viscosity.
Eulerian multiphase. No source term is added for other scalars such as turbulence or userdefined scalars. See Section 23.1: Source Terms due to Mass Transfer • Section 23.7.3: UDFPrescri bed Mass Transfer and the separate UDF Manual for more information about the modeling of mass transfer via userdefined functions. species.7 ModelingMass Transferin Multiphase Flows This section describes the modeling of mass transfer in the framework of FLUENT’s general multiphase models (i. and energy equations.7.7.1 Source Transfer Terms due to Mass FLUENT adds contributions due to mass transfer only to the momentum.23.7.7. or define your own mass transfer model via userdefined functions. You can use models available in FLUENT (e. 23. the mass transfer will be with the bulk phase. There are numerous kinds of mass transfer processes that can be modeled in FLUENT.4: Cavitation Models i Note that FLUENT’s curre nt cavitation model can only be used in the framework of the mixture multiphase model. mixture multiphase.g. Information about mass transfer is presented in the following subsections: • Section 23.72) (23. In case a particular phase does not have a mixture material associated with it..e.2: Unidirectional Consta nt Rate Mass Transfer • Section 23. Mass Equation The contribution to the mass source for phase p in a cell is mp = −mpi qj and for phase q is mq = mpi qj (23.7. Let mpi qj be the mass transfer rate per unit volume from the ith species of phase p to the j th species of phase q.3: UDFPrescri bed Mass Transfer • Section 23.71) . VOF multiphase). FLUENT’s cavitation model).
74) (23.23.7 M odelingM ass Transferin M ultiphase Flows M omentum Equation For VOF or mixture models. . Species Equation The species source in a cell for species i of phase p is mi = −mpi qj p and for species j of phase q is mj = mpi qj q (23.75) (23.73) Energy Equation For all multiphase models. there is no momentum source. The transfer of these scalar quantities due to mass transfer could be modeled using userdefined source terms. the following energy sources are added.76) where hf p and hf q are the formation enthalpies of species i of phase p and species j of phase q respectively and hi is the enthalpy of species i of phase p (with reference to the p formation enthal py). the momentum source in a cell for phase p is mp~up = −mpi qj ~up and for phase q is mq ~uq = mpi qj ~up (23.77) (23. The energy source in a cell for phase p is Hp = −mpi qj (h i ) p and for phase q is i j f Hq = mpi qj (hpi + hf p− h q ) i j (23.78) OtherScalar Equations No source/sink terms are added for turbulence quantities and other scalars. For the Eulerian model.
e.79) and r˙ is a consta nt rate of particle shrinking or swelling.i ρq where yp. then λpq = r˙αp yp. If phase p is a mixture material and a mass transfer mechanism is defined for species i of phase p. . This is not available for the VOF model.710) (23. (23.i is the mass fraction of species i in phase p. evaporation.7. This contribution is based on the assumption that the mass “created” or “destroyed” will have the same momentum and energy of the phase from which it was created or destroyed.2 Unidirectional ConstantRate Mass Transfer The unidirectional mass transfer model defines a positive mass flow rate per unit volume from phase p to phase q: m˙pq = max[0.3 UDFPrescribed Mass Transfer Because there is no universal model for mass transfer. such as the rate of burning of a liquid droplet.711) 23. λpq ] − max[0. or scalar equations. If you would like to input your source terms directly into momentum. then the appropriate path is to use UDFs for userdefined sources for all equations. etc.23. FLUENT provides a UDF that you can use to input models for different types of mass transfer. −λpq ] where λpq = r˙αp ρq (23. FLUENT will automatically add the source contribution to all relevant momentum and scalar equations. condensation.g. rather than the UDF for mass transfer. See the separate UDF Manual for more information about UDFbased mass transfer in multiphase. boiling.7. Note that when using this UDF. energy.
It accounts for all firstorder effects (i. which under decreasing pressure may grow and form cavities. developed by Singhal et al. [334].23. which may fall below the saturated vapor pressure. You can use FLUENT’s current cavitation model to include cavitation effects in twophase flows when the mixture model is used. The cavitation model can be used with the mixture multiphase model (with or without slip velocities). The process of rupturing the liquid by a decrease of pressure at consta nt temperature is called cavitation.7. or flows with multiphase species trans port applications. • the extended cavitation model capabili ty This includes a description of the extension of the cavitation model for multiphase (Nphase) flows. very large and steep density variations happen in the lowpressure/c avitating regions. Overviewof the CavitationModel A liquid at consta nt temperature can be subjected to a decreasing pressure. variable fluid densi ty flows. and noncondensable gases). or nuclei. phase change. The cavitation model implemented here is based on the socalled “full cavitation model”. However.. In such processes.4 CavitationModels This section provides information about the cavitation model used in FLUENT.e. bubble dynamics. unlike the original approach [334] assuming singlephase. isothermal. It has the capability to account for multiphase (Nphase) flows or flows with multiphase species trans port. The complete cavitation model capability in FLUENT can be presented in two parts: • the basic cavitation model This includes a description of the fundame ntal modeling approach and the standard twophase cavitation model. turbule nt pressure fluctuations. and the thermal effects and compressibility of both liquid and gas phases. The liquid also contains the microbubbles of noncondensable (dissolved or ingested) gases. . the cavitation model in FLUENT is under the framework of multiphase flows. the effects of slip velocities between the liquid and gaseous phases.
the following assumptions are made: • The system under investigation involves only two phases (a liquid and its vapor). The following limitations apply to the cavitation model in FLUENT: • The cavitation model cannot be used with the VOF model. See the separate UDF Manual for more information on userdefined density functions. The cavitation model offers the following capabilities: • The cavitation model accounts for the mass transfer between a single liquid and its vapor. liquid surface tension coefficient) can be either a consta nt or a function of temperature. • Both liquid and vapor phases can be incompressible or compressible. • The cavitation model can only used for cavitating flow occurring in a single liquid fluid. It is always preferable to solve for cavitation using the mixture model without slip velocity. the density can be described using a userdefined function. • It can be solved with the mixture energy equation. . • The parameters used in the mass transfer model for cavitation (vaporization pres.Basic CavitationModel In the standard twophase cavitation model. • Both bubble formation (evaporation) and collapse (condensation) are taken into account in the model. because the surface tracking schemes for the VOF model are incompatible with the interpenetrating continua assumption of the cavitation model. • The cavitation model can be used only for multiphase simulations that use the mixture model. • With the cavitation model. the secondary phase must be vapor. The noncon.sure.densible gases are assumed to always be compressible. • It is compatible with all the available turbulence models in FLUENT. and a certain fraction of separately modeled noncondensable gases. slip velocities can be turned on if the problem suggests that there is significant slip between phases. • It is fully compatible with dynamic mesh and nonconformal interfaces. • The mass fraction of noncondensable gases is known in advance. For compressible liquids. the primary phase must be liquid.
local static pressure and are given by: when p < psat Vch Re = Ce ρl ρv σ when p > psat Vch Rc = Cc ρl ρv σ s s 2(psat − p)(1 − f ) 3ρl (23.VaporM ass Fractionand VaporTranspo rt The working fluid is assumed to be a mixture of liquid. Vch is a characteristic the surface tension approximated the the localsat is the liquid saturation vapor = √k). The default values are Ce = 0.01. and Re and Rc are the vapor generation and condensation rate terms (or phase change rates).e. Standard governing equations in the mixture model and the mixture turbulence model describe the flow and account for the effects of turbulence. v~v is the velocity vector of the vapor phase. A vapor trans port equation governs the vapor mass fraction.02 and Cc = 0. given by: ∂ (ρf ) + ∇(ρv~v f ) = ∇(γ∇f ) + Re − Rc ∂t (23. (i. and Ce and Cc are empirical consta nts. and limiting bubble size considerations (interface surface area per unit volume of vapor) [334]. These rates are functions of the instantaneous.714) where the suffixes l and v denote the liquid and vapor phases. Vch pressure at the given temperature.713) 2(p − psat ) f 3ρl (23. γ is the effective exchange coefficient. vapor and noncondensable gases. The rate expressions are derived from the RayleighPlesset equations. f . which is coefficient of by liquid.712) where ρ is the mixture densi ty. . p turbulence intensity. σ is velocity.
715) where ρl ..g. Even a very small amount (e. and αg are the respective volume fractions. It may be noted that the noncondensable gas is not defined as a phase or a material. By default. and αl .714 and the volume fraction (α i ) in Equation 23.716) 1 (psat + pturb ) 2 (23. However.39ρk where k is the local turbulence kinetic energy. then the molecular weight can be changed by using a text command. αv + αg ) is commonly referred to as the void fraction (α). ρ. ρv .. the molecular weight and temperature are required. αv . Effectsof NoncondensableGases The operating liquid usually contains small finite amounts of noncondensable gases (e. 10 ppm) of noncondensable gases can have significant effects on the cavitating flow field due to expansion at low pressures (following the ideal gas law). aeration). and ρg are the densities of the liquid.g. and the noncondensable gases. For more information.717) (23.e. with the gaseous phase comprising of the liquid vapor and the noncondensable gases. the working fluid is assumed to be a mixture of the liquid phase and the gaseous phase. When using the ideal gas law to compute the noncondensable gas density. FLUENT’s cavitation model accounts for the turbulenceinduced pressure fluctuations by simply raising the phasechange threshold pressure from psat to pv = where pturb = 0. the vapor. contact your FLUENT support engineer. The relationship between the mass fraction (f i ) in Equations 23. In the present approach. The density of the mixture. dissolved gases.718) The combined volume fraction of vapor and gas (i.0. the gas is assumed to be air and the molecular weight is set to 29. .717 is αi = fi ρi ρ (23.TurbulenceInduced Pressure Fluctuations Significant effect of turbulence on cavitating flows has been reported [311].. is calculated as ρ = αv ρv + αg ρg + (1 − αv − αg )ρl (23.712–23. if the noncondensable gas is not air. respectively.
720) . • Open the Solution Initialization panel. the default value is set to 300 K when the energy equation is not activated. isothermal flow). If the temperature is different. Define −→ Models −→Energy.719) (23. turn off the energy equation by deselecting Energy in the list. By so doing. you can change the temperature in FLUENT in the following way: • Activate the energy equation. • Open the Solution Controls panel.. • In the Solution Controls panel.. the final phase rate expressions are written as: when p < pv 2(pv − p)(1 − f − f ) k v g Re = Ce ρl ρv 3ρl σ when p > pv k 2(p − pv ) ρl ρl Rc = Cc fv 3ρl σ √ s √ s (23.. • In the Solution Initialization.As for the temperature. set the initial value as a desired temperature. Solve −→ Initialize −→Initialize.. Phase Change Rates After accounting for the effects of turbulenceinduced pressure fluctuations and noncondensable gases. FLUENT uses the initial values for the temperature.e.... Solve −→ Controls −→Solution. but still a consta nt (i.. under Equations.
0.2 – 0. The pressure is set close to the highest pressure among the inlets and outlets to avoid unexpected low pressure and cavitating spots. The density and the vaporization mass (source term in the vapor equation) can also be relaxed to improve convergence. In addition. The following is a list of factors that must be considered when using the cavitation model. which.g. the relaxation factor for density is set between the values of 0. say in the range 0. and increase them to reasonable values after a sufficient number of iterations. in many cases. This can be achieved by performing the following steps: 1. • initial conditions Though no special initial condition settings are required. when the saturation pressure of a liquid at a . along with tips to help address potential numerical problems: • relaxation factors In general..4. Set near zero relaxation factors for the vaporization mass and for density. between 0. are then usually very difficult for the model to correct. and near zero saturation pressure all cause unfavorable effects on solution convergence. several factors greatly influence numerical stability. usually. Obtain a converged / nearconverged solution for a single phase liquid flow. and then switch on the cavitation model.g. if the liquid is purified of noncondensable gases. it may be beneficial to obtain a realistic pressure field before substa ntial cavities are formed. while for the vaporization mass values between 0.Additional uidelinesfor the CavitationModel G In practical applications of the cavitation model. enhance numerical stability and lead to more realistic results. A value of zero for the mass fraction of noncondensable gases should generally be avoided.0 may be appropriate. In some cases. 10−8 ) may be used to replace the default value of 1. small relaxation factors are advised for momentum equations. once present. large ratio of liquid to vapor density. For instance. even smaller relaxation factors may be required for all the equations.05 – 0. • noncondensable gases Noncondensable gases are usually present in liquids. higher mass fractions of the noncondensable gases may. Typically. For some extreme cases. 2. 15 ppm) of noncondensable gases can have significant effects on both the physical realism and the convergence characteristics of the solution. In particular.5×10−5 .3 and 1.1 and 1. In fact. it is suggested that the vapor fraction is always set to inlet values. high pressure difference between the inlet and exit.7. The relaxation factor for the pressurecorrection equation should usually be larger than those for momentum equations. Also. poor initial conditions very often lead to an unrealistic pressure field and unexpected cavitating zones. in complicated cases. a much smaller value (e. Even a small amount (e..
there exist other gaseous species in the systems investigated. the basic twophase cavitation model needs to be extended to a multiphase (Nphase) flows.7. For more information. FLUENT sets the maximum pressure limit to 5. contact your FLUENT support engineer. in a ventilated supercavitating vehicle. For some very complicated cases. For instance. In order to be able to predict those type of cavitating flows. • limits for dependent variables In many cases.4. it is more desirable to use the following pressure discretization schemes in cavitation applications: – body force weighted – second order – PRESTO! The standard and linear schemes generally are not very effective in complex cavitating flows. ExtendedCavitationModelCapability In many practical applications. this factor is set to 0. setting the pressure upper limit to a reasonable value can help convergence greatly at the early stage of the solution.0×1010 Pascal. It is advised to always limit the maximum pressure when it is possible. or a multiphase species trans port cavitation model. which should work well for most of the cases. a special relaxation factor is introduced for the pressure correction equation. • pressure discretization schemes As for many multiphase flows. By default. • the relaxation factor for the pressure correction equation For cavitating flows. . the incoming flow is a mixture of a liquid and some gaseous species. when cavitation occurs. you may experience the divergence of the AMG solver. noncondensable gases will play a crucial role both numerically and physically. Also in some cases. air is injected into a liquid to stabilize or increase the cavitation along the vehicle surfaces. By default. You can set the value of this relaxation factor by typing a text command.certain temperature is zero or very small. this value may be reduced to no less than 0. however. Under those circumstances.
• Only one secondary phase can be defined as compressible gas phase.721) • The primary phase can only be a single liquid. while a user. M ultiphase Species Transpo t CavitationModel r In some cases.defined density may be applied to all the phases. • The basic cavitation model is still used to model the phase changes between the liquid and vapor. for the multiphase cavitation model also (23. The detailed description of the multiphase species trans port approach can be found in Section 23.8: Modeling Species Trans port in Multiphase Flows. The multiphase species trans port cavitation model can be summarized as follows: • All the assumptions/limitations apply here. In addition to the primary liquid and secondary vapor phase. the multiphase species trans port approach offers an option to handle these type of applications by assuming that there is one compressible gas phase with multiple species. and the noncondensable gas needs to be modeled by a separate compressible gas phase. the general trans port equation governing the mass fraction fi given by: ∂ (ρfi ) + ∇(ρv~v fi ) = ∇(γfi ∇fi ) + Sfi ∂t where Sfi is a (userdefined) source term. the mass fraction needs to be set to 0. • For an noncavitating phase i. Since only one compressible gas phase is allowed in the general multiphase approach. Sfi = 0.M ultiphase CavitationModel The multiphase cavitation model is an extension of the basic twophase cavitation model to multiphase flows. • All the secondary phases allow more than one species. It is desirable to consider them all compressible. By default. To exclude noncondensable gases from the system. • The predescribed noncondensable gases can still be included in the system. there are several gas phase components in a system. . more secondary gaseous phases can be included into the computational system under the following assumptions/limitations: • Mass transfer (cavitation) only occurs between the first and the second phases.
For most multiphase species trans port problems. can also be applied to multiphase flows. and R is the heterogeneous reaction rate. 23. but each species has its own pressure (i.9. partial pressure). αq is the volume fraction for phase q and Si q is the rate of creation by addition from the dispersed phase plus any userdefined sources. As a result. The reacta nts and the products belong to the same mixture material (set in the Species Model panel). predict the local mass fraction of each species.81) J~i p=1 where Ri q is the net rate of production of homogeneous species i by chemical reaction for phase q.23. FLUENT treats homogeneous gas phase chemical reactions the same as a singlephase chemical reaction..719 are the partial density and pressure of the vapor.11).e. and hence the same phase. as described in Chapter 14: Modeling Species Trans port and FiniteRate Chemistry. • The mass transfer between other phases or species are modeled with the standard mass transfer approach. boundary con. m˙ qj pi is the mass transfer source between species i and j from phase q to p. You can choose to solve the conservation equations for chemical species in multiphase flows by having FLUENT. see Chapter 14: Modeling Species Trans port and FiniteRate Chemistry. For more information. either as a phase or a species.ditions for a particular species are set in the associated phase boundary condition panel (see Chapter 23. Yi k .8 M odelingSpecies Transpo t in M ultiphase r Flows • The vapor. the vapor density and the pressure used in Equation 23. the phase shares the same pressure as the other phases. The setup of a homogeneous gas phase chemical reaction in FLUENT is the same as it is for a single phase.8 ModelingSpecies Transpo in M ultiphase rt Flows Species trans port. must be in the second phase. The generalized chemical species conservation equation (Equation 14. through the solution of a convectiondiffusion equation for the ith species. for each phase k. when applied to a multiphase mixture can be represented in the following form: n X ∂ q q q q (ρ α Yi )+∇·(ρq αq~v q Yi q ) = −∇·αq +αq Ri q +αq Si q + (m˙ pi qj − m˙ qj pi )+R ∂t (23. and postprocessing and .8: Defining Multiphase Boundary Conditions). In the standard model. The reaction rate is scaled by the volume fraction of the particular phase in the cell. the zero consta nt rate should be chosen. • For the phases with multiple species. • The mass transfer between a liquid and a vapor phase/s pecies is modeled by the basic cavitation model. In addition.
The mass transfer and heterogeneous reactions will be associated with the bulk fluid for phases with a single fluid material. see Section 23. For multiphase species trans port simulations. The species of different phases is entirely independent. 23.chemistry models of homogeneous reactions are available for multiphase species trans port. turn on Volumetric under Reactions in the Species Model panel.9. i To turn off reactions for a particular phase. The species equations are solved in those phases that are assigned a mixture material.reporting of results is performed on a perphase basis (see Section 23.1 Limitations The following limitations flows: exist for the modeling of species trans port for multiphase • The nonpremixed. Explicit relationships between species of different phases can be specified through mass transfer and heterogeneous reactions. while keeping the reactions active for other phases. respectively. see Chapter 14: Modeling Species Trans port and FiniteRate Chemistry . select none from the Reactions dropdown list. and Particle Surface reactions. The species equation above is solved for the mass fraction of the species in a particular phase.6: Specifying Heterogeneous Reactions . • The discrete phase model (DPM) is not compatible with multiphase species trans. FLUENT treats multiphase surface reactions as it would a singlephase reaction. For more information. partiallypremixed combustion.7: Including Mass Transfer Effects and Section 23. Wall Surface. • The stiff chemistry solver is not available for multiphase species reactions. There is no implicit relationship between them even if they share the same name.8. Then.port. or the composition PDF trans port species trans port models are not available for multiphase species reactions. premixed.15: Postprocessing for Multiphase Modeling).9. For more information on mass transfer and heterogeneous reactions. . Some phases may have a fluid material associated with them instead of a mixture material. • Only the laminar finiterate. finiterate/eddydissipation and eddydissipation turbulence. in the Materials panel. The reaction rate is scaled with the volume fraction of the particular phase in the cell. the Species Model panel allows you to include Volumetric.
p represents the product.86) (23. or the bulk phase itself if the phase does not have a mixture material associated with it. The general expression for the mass source for the ith phase is Sri = − R Spi = R X ri r r γ j Mj (23.8. FLUENT loops through all the mass transfer mechanisms to compute the net mass source/sink of each species in each phase. . The net mass source/sink of a species is used to compute species and mass source terms.2: Unidirectional Consta nt Rate Mass Transfer) or through userdefined functions. multiple mass transfer mechanisms need to be input. Mass Transfer Mass source for the phases are given by: S1 = R(cM c − aM a ) S2 = R(dM d − bMb ) (23. Instead of a matrix. M is the molecular weight.85) (23.82) where S is the mass source. For other equations.87) X pi p p γj Mj Si = Spi + Sri where γ is the stoichiometric coefficient. Each mass transfer mechanism defines the mass transfer phenomenon from one entity to another entity. the trans port due to mass transfer needs to be explicitly modeled by the user.7.84) (23.83) (23. and r represents the reacta nt.type input.2 Mass and M omentumTransferwith M ultiphase Species Transport The FLUENT multiphase mass transfer model accommodates mass transfer between species belonging to different phases. An entity is either a particular species in a phase. The mass transfer phenomenon could be specified either through the inbuilt unidirectional “consta ntrate” mass transfer (Section 23. and R is the reaction rate.23. Source Terms due to Heterogeneous Reactions Consider the following reaction: aA + bB → cC + dD Let as assume that A and C belong to phase 1 and B and D to phase 2. FLUENT will also automatically add the source contribution to all relevant momentum and energy equations based on that assumption that the momentum and energy carried along with the transferred mass.
~unet . then the above term will be zero. Species Transfer The general expression for source for k th species in the j th phase is S ri = −R k S pk = R i X rk i p p γ j Mj k k r γ r Mj j k k (23.814) (23.810) (23.811) The general expression is u~ Si = Spi ~unet − R γj Mj ~ui ri X r r (23. of the reacta nts is given by: ~unet = aM a~u1 + bMb~u2 aM a + bMb (23. Momentum transfer for the phases is then given by: S ~ = R(cMc~unet − aM a~u1 ) u 1 = R(dMd~unet − bMb~u2 ) S~ u 2 (23.89) where j represents the j th item (either a reacta nt or a product).812) If we assume that there is no momentum transfer.815) X pk i k S i = Spik + Srik .88) The general expression for the net velocity of the reacta nts is given by: ~unet γr M r ~ur = P jrjr r γj Mj r P j (23. The net velocity. but we can assume that the reacta nts mix (conserving mome ntum) and the products take momentum in the ratio of the rate of their formation.813) (23.M om entum Transfer Momentum transfer is more complicated.
The general expression for the heat source is: H Si R = Spi Hnet − X ri r γr Mj Hr j j ! X p p p f + γ j Mj h j pi (23.H eatTransfer For heat transfer. we need to consider the formation enthalpies of the reacta nts and products as well: The net enthalpy of the reacta nts is given by: Hnet = aM a (H a + hf ) + bMb (Hb + h ) a b aM a + bMb f (23.821) SiH = Spi Hnet − R X pi p p γj Mj hf j p (23. heat transfer for the phases are given by: c H S1 = R(cM c Hnet − aM a H − cMc hf ) a − dMd hdf ) H = R(dM H d net − bMb H S2 b (23.818) (23.820) If we assume that there is no heat transfer. The general expression for Hnet is: P r Hnet = γ r M r (H r + hf j j )j j P r r r γ Mj j r (23. Thus the expression for Hnet will be: P Hnet = H The expression Si will be γ r M jr hfj j P r r r γ Mj j r r (23.819) The last term in the above equations appears because our enthalpy is with reference to the formation enthal py.822) .816) where hf represents the formation enthalpy.817) If we assume that this enthalpy gets distributed to the products in the ratio of their mass production rates. and H represents the enthal py. we can assume that the different species only carry their formation enthalpies with them.
See Sections 23. specify the VOF formulation as well. See Section 8..3: Modeling Turbulence for details.. 3. Define the phases. See Section 23.12. you will need to create a new material for it in the fluid materials category (not the solid materials category). 2.. See Sections 23. create a new material. Define −→ Models −→Viscous. (Eulerian model only) If the flow is turbule nt.1: Additional Guidelines for Eulerian Multiphase Simulations for guidelines on simplifying Eulerian multiphase simulations..23. define the multiphase turbulence model.g. Enable the multiphase model you want to use (VOF. Define −→Materials.2: Using the Materials Panel for details about copying from the database and creating new materials. 1. For information about inputs related to other models that you are using in conjunction with the multiphase model.1 and 23. Define −→ Models −→Multiphase.. and specify any interaction between them (e.9 Steps for Using a Multiphase Model The procedure for setting up and solving a general multiphase problem is outlined below.3 for additional information about specifying material properties for a compressible phase (VOF and mixture models only). mixture. . slip velocity functions if you are using the mixture model. or drag functions if you are using the Eulerian model).10.9. 4. or Eulerian) and specify the number of phases.3–23.1.12.12. and described in detail in the subsections that follow.1 for details.. If the material you want to use is not in the database. Define −→Phases.9.. Copy the material representing each phase from the materials database.10. See also Section 23. see the appropriate sections for those models.11. Remember that only the steps that are pertine nt to general multiphase calculations are shown here. i If your model includes a particulate (granular) phase.5 and 23. See Sections 23. It is possible to turn off reactions in some materials by selecting none in the Reactions dropdown list under Properties in the Materials panel. surface tension if you are using the VOF model.. For the VOF model.2 for details..
2: Solving a Homogeneous Multiphase Flow • Section 23.10. See Section 23. See Sections 23. Information is presented in the following subsections: • Section 23.9 Steps for Using a M ultiphase odel M 5.14. 6. Define −→Boundary Conditions. See Sections 23.9. turn on gravity and specify the gravitational acceleration.9.14 and 23.6: Specifying Heterogeneous Reactions • Section 23. including the secondaryphase volume fractions at flow boundaries and (if you are modeling wall adhesion in a VOF simulation) the contact angles at walls.. Specify the boundary conditions. Postprocessing and reporting of results is available for each phase that is selected.. Calculate a solution and examine the results.9.5: Modeling Multiphase Species Transport • Section 23.14 for details..23. mixture.4 and 23. 7..1: Setting Initial Volume Fractions for details. 9.. See Section 23. If body forces are present. Define −→Operating Conditions.9.9. Set any modelspecific solution parameters.7: Including Mass Transfer Effects • Section 23. Initialize the solution and set the initial volume fractions for the secondary phases.4: Including Body Forces • Section 23..1: Enabling the Multiphase Model • Section 23.9. 8.8: Defining Multiphase Boundary Conditions .15 for details..9. This section provides instructions and guidelines for using the VOF. Solve −→ Initialize −→Patch.3: Defining the Phases • Section 23. and Eulerian multiphase models.9. Solve −→ Controls −→Solution..9.9. See Section 23.8: Defining Multiphase Boundary Conditions for details.4: Including Body Forces for details.
mixture. select Volume of Fluid. Figure 23.M odelingM ultiphase Flows 23.9.9. .. Mix.1 Enablingthe Multiphase Model To enable the VOF.1: The Multiphase Model Panel The panel will expand to show the relevant inputs for the selected multiphase model.ture. or Eulerian as the Model in the Multiphase Model panel (Figure 23.9. or Eulerian multiphase model. Define −→ Models −→Multiphase..1).
2 Solving a Homogeneous Multiphase Flow If you are using the mixture model. one in which the phases all move at the same velocity). 23.9.4: Including Body Forces) If you selected the mixture model.9 Steps for Using a M ultiphase odel M If you selected the VOF model.4.10.23. the inputs are as follows: • number of phases • whether or not to compute the slip velocities (see Section 23.9. you have the option to disable the calculation of slip velocities and solve a homogeneous multiphase flow (i.5: Relative (Slip) Velocity and the Drift Velocity. FLUENT will compute the slip velocities for the secondary phases. as described in Section 23. turn off Slip Velocity under Mixture Parameters. the input is the following: • number of phases To specify the number of phases for the multiphase calculation.geneous Multiphase Flow) • (optional) implicit body force formulation (see Section 23. By default.1: Choosing a VOF Formulation ) • (optional) implicit body force formulation (see Section 23. You can specify up to 20 phases.9. . enter the appropriate value in the Number of Phases field.9. If you want to solve a homogeneous multiphase flow..4: Including Body Forces) If you selected the Eulerian model. the inputs are as follows: • number of phases • VOF formulation (see Section 23.e.2: Solving a Homo.
9.9. and Eulerian models. respectively. slip velocity for the mixture model.10..2 for the VOF. ..2)..M odelingM ultiphase Flows 23.3 Definingthe Phases To define the phases (including their material properties) and any interphase interaction (e. click the Interaction..3. To specify any interaction between the phases. surface tension and wall adhesion for the VOF model. and secondaryphase indicates that the selected item is a secondary phase.1.9. Define −→Phases. Figure 23. Instructions for defining the phases and interaction are provided in Sections 23. 23. mixture.11. and 23.2: The Phases Panel Each item in the Phase list in this panel is one of two types. button. drag functions for the mixture and the Eulerian models).12.. you will use the Phases panel (Figure 23. as indicated in the Type list: primaryphase indicates that the selected item is the primary phase.g.
. turn on Gravity in the Operating Conditions panel and specify the Gravitational Acceleration. This treatme nt im.9 Steps for Using a M ultiphase odel M 23. . i For VOF and mixture calculations involving body forces. (This excludes the buildup of hydrostatic pressure within the lightest phase. See Section 23..23. and set the Operating Density to be the density of the lightest phase. Define −→Operating Conditions. This results in extra body force correction terms in Equation 25.4 IncludingBody Forces When large body forces (e.9. Segregated algorithms converge poorly unless partial equilibrium of pressure gradient and body forces is taken into account.413. and allows the flow to achieve a realistic pressure field very early in the iterative process.4: Including Body Forces for details. with an additional term involving corrections to the body force.) If any of the phases is compressible. it is recom.g.. gravity or surface tension forces) exist in multiphase flows. the body force and pressure gradient terms in the momentum equation are almost in equilibrium.9. FLUENT provides an optional “implicit body force” treatme nt that can account for this effect. Equation 25. set the Operating Density to zero. you should also turn on the Specified Operating Density option in the Operating Conditions panel. To include this body force.proves solution convergence by accounting for the partial equilibrium of the pressure gradient and body forces in the momentum equations.mended that you also turn on the Implicit Body Force treatme nt for the Body Force Formulation in the Multiphase Model panel.411. improving the roundoff accuracy for the momentum balance. with the contributions of convective and viscous terms small in comparison. For VOF calculations. The basic procedure involves augmenting the correction equation for the face flow rate. making the solution more robust.
.5 ModelingMultiphase Species Transpo rt FLUENT lets you describe a multiphase species trans port and volumetric reaction (Section 23.8: Modeling Species Trans port in Multiphase Flows) in a fashion that is similar to setting up a singlephase chemical reaction using the Species Model panel (e.3: The Species Model Panel with a Multiphase Model Enabled 1. 2. Figure 23...3). 4. Turn on Volumetric under Reactions.9.M odelingM ultiphase Flows 23.9.. Select Species Transport under Model. button to display the Phase Properties panel (Figure 23. & Figure 23.g.4).9. .9.. 3. Click the Set. Define −→ Models −→ Species −→Transport Reaction. Select a specific phase using the Phase dropdown list under Phase Properties.
9.9. FiniteRate/EddyDissipation (for turbule nt flows) computes both the Arrhenius rate and the mixing rate and uses the smaller of the two. the material for each phase is listed in the Material dropdown list. Three models are available: Laminar FiniteRate computes only the Arrhenius rate (see Equation 14. additional inputs may also be required depending on your modeling needs. See. you can choose the material that you want to use for a specific phase.126 and 14.4: The Phase Properties Panel In the Phase Properties panel.6: Specifying Heterogeneous Reactions for more information defining heterogeneous reactions. choose the TurbulenceChemistry Interaction model. EddyDissipation (for turbule nt flows) computes only the mixing rate (see Equations 14.23. If you want to inspect or edit any of the properties of any of the materials. for example. When modeling multiphase species trans port. Section 23.7: Including Mass Transfer Effects for more information on mass transfer effects. button. In the Species Model panel.9. .18) and neglects turbulence. From this list.9 Steps for Using a M ultiphase odel M Figure 23. 5.. then you need to open the Materials panel by clicking the Edit. The dropdown list contains all of the materials that have been defined for your simulation.. or Section 23.chemistry interaction.127).
M odelingM ultiphase Flows 23. Set the ID of each reaction you want to define.5). 1.9. 5.2). and particle surface reactions) in the Total Number of Heterogeneous Reactions field. Set the total number of reactions (volumetric reactions..) .9.. (Again.6 Specifying Heterogeneous Reactions You can use FLUENT to define multiple heterogeneous reactions and stoichiometry using the Phase Interaction panel (e.. button to open the Phase Interaction panel. or type in the value and press <Enter>.g.9. (Use the arrows to change the value.9. click the Interaction. if you type in the value be sure to press <Enter>. Specify the Reaction Name of each reaction that you want to define. Figure 23. 3. Click the Reactions tab in the Phase Interaction panel.. wall surface reactions. Figure 23.) 4. In the Phases panel (Figure 23. Define −→Phases..5: The Phase Interaction Panel for Heterogeneous Reactions 2.
defining mass transfer through cavitation is described separately in Section 23.9.11. (The stoichiometric coefficient 0 0 is the consta nt νi. . This assumption can be deacti vated using a text command. indicate the Phase and Species and the stoichiometric coefficient for each of your reacta nts and products. Select each reacta nt or product in the Reaction tab and then set its stoichiometric coefficient in the Stoich. mixture multiphase.7 Including Effects M ass Transfer As discussed in Section 23. contact your FLUENT support engineer. For each reaction.e.r or ν0i. 8. For each reaction.16. or VOF multiphase) can be modeled in one of three ways: • unidirectional consta nt rate mass transfer (not available for VOF calculations) • UDFprescribed mass transfer • mass transfer through cavitation (only valid for the mixture multiphase model) Because of the different procedures and limitations involved. i FLUENT assumes that the reacta nts are mixed thoroughly before reacting together.2: Including Cavitation Effects. For more information. For more information. thus the heat and momentum transfer is based on this assumption.r in Equation 14.23.. i The heterogeneous reaction rates can only be specified using a userdefined function. see the separate UDF Manual.7: Modeling Mass Transfer in Multiphase Flows. mass transfer effects in the framework of FLUENT’s general multiphase models (i.9 Steps for Using a M ultiphase odel M 6. specify how many reacta nts and products are involved in the reaction by increasing the value of the Number of Reactants and the Number of Products. A UDF is available for an Arrhenius. Coefficient field.) 7. Eulerian multiphase. 23. For each reaction. indicate an applicable userdefined function using the Reaction Rate Function dropdown list.type reaction with rate exponents that are equivalent to the stoichiometric coefficients.
. either as unidirectional consta nt or using a UDF. Figure 23.. If species trans port is part of the simulation.2). Figure 23. 1. specify the phase of the destination material phase under To Phase.g.6: The Phase Interaction Panel for Mass Transfer 2.. specify the phase of the source material under From Phase. and the source phase is composed of a mixture material. you will need to use the Phase Interaction panel (e. 6. then specify the species of the source phase mixture material in the corresponding Species dropdown list. For each mechanism. For each mechanism. . Define −→Phases. Note also that the same pair of phases can have multiple mass transfer mechanisms and you have the ability to activate and deacti vate the mechanisms of your choice.6).. In the Phases panel (Figure 23. 4. You can include any number of mass transfer mechanisms in your simulation.. button to open the Phase Interaction panel.9. Specify the Number of Mass Transfer Mechanisms. 3.9.9.To define mass transfer in a multiphase simulation. Click the Mass tab in the Phase Interaction panel. 5. click the Interaction.
i Note that including species trans port effects in the mass trans port of multi. unidirectional mass transfer. through a userdefined function. If a particular phase does not have a species associated with it. . then the mass transfer throughout the system will be performed by the bulk fluid material. This approach is a more involved but more powerful. FLUENT assumes that the reacta nts are mixed thoroughly before reacting together. userdefined allows you to impleme nt a correlation reflecting a model of your choice. and the destination phase is composed of a mixture material. you can define a mass transfer mechanism between species from different phases.7. Define −→ Models −→ Species −→Transport & Reaction.phase simulation requires that Species Transport be turned on in the Species Model panel. energy. then specify the species of the destination phase mixture material in the corresponding Species dropdown list. contact your FLUENT support engineer. For more information. When your model involves the trans port of multiphase species. The following choices are available: constantrate enables a consta nt. thus the heat and momentum transfer is based on this assumption. and turbulence are also trans ported with the mass that is transferred. i Note that momentum. For each mass transfer mechanism. allowing you to split the source terms according to a model of your choice.. Another option to model mass transfer between phases is through the use of userdefined sources and their inclusion in the relevant conservation equations. If species trans port is part of the simulation. This assumption can be deactivated using a text command. FLUENT will automatically include the terms needed to model mass transfer in all relevant conservation equations.. 8. select the desired mass transfer correlation under Mechanism.
.7: The Boundary Conditions Panel BoundaryConditionsfor the M ixtureand the Individual Phases The conditions you need to specify for the mixture and those you need to specify for the individual phases will depend on which of the three multiphase models you are using.7). while other conditions are shared by all phases (i. the mixture).e. Details for each model are provided below. Define Conditions.9.. but the procedure for setting multiphase boundary conditions is slightly different than for singlephase models.. You will need to set some conditions separately for individual phases. −→Boundary Figure 23.8 Defining Conditions Multiphase Boundary Multiphase boundary conditions are set in the Boundary Conditions panel (Figure 23. as described in detail below..23.9.9.
porous jump. or wall zone.1.9. you will need to set the mass flow rate or mass flux for each individual phase. See Chapter 7: Boundary Conditions for details about the relevant conditions for each type of boundary. solid. radiator. all conditions are specified for the mixture.26: Boundary Profiles). • For a mass flow inlet. i Note that if you read a VOF case that was set up in a version of FLUENT prior to 6. intake fan. the conditions you need to specify for each type of zone are listed below and summarized in Table 23.19. there are no conditions to be specified for the primary phase. All other conditions are specified for the mixture. • For a wall zone. or velocity inlet. • For an exhaust fan. • For a fluid zone. Note that the pressure farfield boundary is not available with the VOF model. The resistance coefficients and direction vectors. you can specify the contact angle if wall adhesion option is enabled. • For an axis. a profile (see Section 7. If the fluid zone is not porous. There are no conditions to be set for the individual phases.VO F Model If you are using the VOF model. you will need to redefine the conditions at the mass flow inlets. . If the fluid zone is porous. fan. all other conditions are specified for the mixture. you will need to set the volume fraction as a consta nt. pressure outlet. and all other sources are specified for the mixture.1. you will enable the Porous Zone option in the Fluid panel for the mixture. however. mass sources are specified for the individual phases. All other conditions are specified for the mixture. outlet vent. The porosity inputs (if relevant) are also specified for the mixture. For each secondary phase. All other conditions are specified for the mixture. are specified separately for each phase. See Section 7. inlet vent.6: User Inputs for Porous Media for details about these inputs. symmetry. or a userdefined function (see the separate UDF Manual). pressure inlet. periodic. outflow.
intake fan. pressure inlet. inlet vent. other porous inputs not available mass source.1: PhaseSpecific and Mixture Conditions for the VOF Model Type exhaust fan. outlet vent. other porous inputs not available porous zone. all others .Table 23. outflow. velocity inlet mass flow inlet axis. radiator. solid.9. pressure outlet. wall pressure farfield fluid Primary nothing Phase Secondary Phase volume fraction M ixture all others mass flow/flux nothing mass flow/flux nothing all others all others not available mass source. symmetry. porous jump. periodic. porosity. fan.
• For a fluid zone. are specified separately for each phase. Note that the pressure farfield boundary is not available with the mixture model. there are no conditions to be specified for the primary phase. See Section 7. or pressure inlet. If the fluid zone is not porous. you will specify the velocity for the individual phases. • For a velocity inlet. There are no conditions to be set for the individual phases. For each secondary phase. i Note that if you read a mixture multiphase case that was set up in a version of FLUENT previous to 6. • For an inlet vent.1. For each secondary phase. mass sources are specified for the individual phases. you will need to set the volume fraction (as described above).2. outflow. intake fan. Outflow boundary conditions are not available for the cavitation model.9. symmetry. All other conditions are specified for the mixture. you will need to set the mass flow rate or mass flux for each individual phase. you will enable the Porous Zone option in the Fluid panel for the mixture. The porosity inputs (if relevant) are also specified for the mixture. All other conditions are specified for the mixture. outlet vent. or pressure outlet. fan. • For a mass flow inlet. or wall zone. The resistance coefficients and direction vectors. you will need to redefine the conditions at the mass flow inlets. All other conditions are specified for the mixture. you will specify the coordinate system (3D only) and flowdirection components for the individual phases. If the fluid zone is porous. a profile (see Section 7. All other conditions are specified for the mixture. For each secondary phase. or a userdefined function (see the separate UDF Manual). radiator. periodic.6: User Inputs for Porous Media for details about these inputs. however. you will specify for the mixture which direction specification method will be used at this boundary (Normal to Boundary or Direction Vector). See Chapter 7: Boundary Conditions for details about the relevant conditions for each type of boundary. all conditions are specified for the mixture. . All other conditions are specified for the mixture. all other conditions are specified for the mixture. the conditions you need to specify for each type of zone are listed below and summarized in Table 23. If you select the Direction Vector specification method. • For an axis. solid.26: Boundary Profiles).19. porous jump. and all other sources are specified for the mixture. you will need to set the volume fraction as a consta nt. you will need to set the volume fraction (as described above). • For an exhaust fan.M ixture Model If you are using the mixture model.
other porous inputs not available mass source. porous jump. spec. method. pressure inlet mass flow inlet velocity inlet axis. other porous inputs not available porous zone. volume fraction nothing dir. volume fraction mass flow/flux velocity. periodic. system.9.2: PhaseSpecific and Mixture Conditions for the Mixture Model Type exhaust fan. outlet vent. fan. symmetry. all others . outflow (n/a for cavitation model). system. intake fan. wall pressure farfield fluid Primary nothing Phase Secondary Phase volume fraction M ixture all others coord. flow direction mass flow/flux velocity nothing coord. radiator. flow direction. pressure outlet inlet vent.Table 23. all others all others all others all others not available mass source. solid. porosity.
If the phase is granular.10 and 23. If you are using the mixture turbulence model. • For an inlet vent. you will need to specify them for the primary phase and for each secondary phase. If you are using the mixture turbulence model. If heat transfer is on. If you are using the dispersed turbulence model. you will need to specify them for the primary phase.9.5. there are no conditions to be specified for the primary phase if you are modeling laminar flow or using the mixture turbulence model (the default multiphase turbulence model). you will also need to set its granular temperature. as indicated in Tables 23.Eulerian Model If you are using the Eulerian model.9. If the phase is granular.5. All other conditions are specified for the mixture.9. If you are using the dispersed turbulence model. If you are using the perphase turbulence model.6. except for backflow total temperature if heat transfer is on. For each secondary phase. you will also need to set the total temperature for the individual phases. you will also need to set the backflow total temperature. If you select the Direction Vector specification method.3 for more information about multiphase turbulence models.9. See Sections 23.6.9. If heat transfer is on. you will need to set the mass flow rate or mass flux for each individual phase. or pressure inlet. you will need to specify them for the primary phase and for each secondary phase. • For a mass flow inlet.12. If you are using the perphase turbulence model. you will need to set the volume fraction as a consta nt.4. and 23. • For an exhaust fan. Note that the specification of turbulence parameters will depend on which of the three multiphase turbulence models you are using. you will need to specify the turbulence boundary conditions for the mixture.3.4–23. or a userdefined function (see the separate UDF Manual). you will specify the coordinate system (3D only) and flowdirection components for the individual phases. the conditions you need to specify for each type of zone are listed below and summarized in Tables 23. 23. you will specify for the mixture which direction specification method will be used at this boundary (Normal to Boundary or Direction Vector).9. or pressure outlet. 23. All other conditions are specified for the mixture. since the energy equations are solved for each phase.26: Boundary Profiles). outlet vent. a profile (see Section 7. you will also need to set its granular temperature. you will need to specify the turbulence boundary conditions for the mixture. you will need to set the volume fraction (as described above). . intake fan. For each secondary phase. you will need to specify them for the primary phase. You will also need to specify the temperature of each phase.
9. Otherwise. If heat transfer is on. FLUENT will calculate the phase velocities. . it is 1.0. as shown in Figure 23.0. i If a secondary phase has zero mass flux (i. neither Phase Velocity Ratio nor Volume Fraction will affect the solution.9. you can specify the slip velocity between phases.e. the Eulerian model is used to run a single phase case). By entering a ratio that is greater than 1.0 to indicate a smaller secondary phase velocity.. which means velocities are the same (no slip).8: – Velocity Ratio The value for the phase velocity ratio is the secondary phase to primary phase velocity ratio. Figure 23. When you select a mass flow inlet boundary for the secondary phase. you are indicating a larger secondary phase velocity. By default. you can enter a ratio that is less than 1. you will also need to set the total temperature for the individual phases. you will specify the velocity for the individual phases.8: MassFlow Inlet Boundary Condition Panel • For a velocity inlet. two options will be available for the Slip Velocity Specification Method. – Volume Fraction If you specify the volume fraction at an inlet.For mass flow inlet boundary conditions.
6: User Inputs for Porous Media for details about these inputs. . unless you are using the mixture turbulence model or the dispersed tur. If the fluid zone is not porous. All other conditions are specified for the mixture. See Section 7. The porosity inputs (if relevant) are also specified for the mixture. you will need to specify them for the primary phase and for each secondary phase. solid. however. All other conditions are specified for the mixture. all conditions are specified for the mixture. you will also need to set its granular temperature. source terms and fixed values for turbulence are specified instead for the mixture.For each secondary phase. If you are using the dispersed turbulence model. if heat transfer is on. all source terms and fixed values are specified for the individual phases. and mass flow inlet boundaries are not available with the Eulerian model. Note that the pressure farfield. • For a wall zone. All other conditions are specified for the mixture. you will enable the Porous Zone option in the Fluid panel for the mixture. you will need to set the volume fraction (as described above). you will need to specify them for the primary phase. See Chapter 7: Boundary Conditions for details about the relevant conditions for each type of boundary.19. porous jump. radiator. If you are using the mixture turbulence model. If the phase is granular. If you are using the perphase turbulence model. • For an axis.bulence model. fan. you will need to specify the turbulence boundary conditions for the mixture. all other conditions are specified for the mixture. shear conditions are specified for the individual phases. are specified separately for each phase. periodic. including thermal boundary conditions. There are no conditions to be set for the individual phases. outflow. If the fluid zone is porous. The resistance coefficients and direction vectors. they are specified only for the primary phase. or symmetry zone. If you are using the mixture turbulence model. • For a fluid zone. If you are using the dispersed turbulence model.
porosity. gran. volume fraction. all fixed values. fan. intake fan. spec. outlet vent. temperature (tot. system. temperature) Secondary Phase volume fraction. temperature) velocity. temperature) nothing M ixture all others coord. temperature) all others axis. temperature) dir. system. mass flow inlet fluid nothing all others shear condition not available shear condition not available all others not available all source terms. all others . gran. periodic.3: PhaseSpecific and Mixture Conditions for the Eulerian Model (for Laminar Flow) Type exhaust fan. all fixed values. volume fraction. temperature (tot. temperature (tot. temperature) coord. flow direction (tot. outflow. method. pressure inlet Primary Phase (tot. all others velocity inlet velocity (tot. solid. other porous inputs porous zone. pressure outlet inlet vent. radiator. porous jump. gran. symmetry wall pressure farfield.9. flow direction. other porous inputs all source terms.Table 23.
porous jump. mass flow inlet fluid nothing all others shear condition not available shear condition not available all others not available other source terms. temperature (tot. flow direction (tot. outlet vent. intake fan. temperature) Secondary Phase volume fraction. gran. gran. system. outflow. all others velocity inlet velocity (tot. other fixed values. other fixed values. spec. temperature (tot. periodic. system. pressure outlet inlet vent. temperature) coord.9. porous zone. gran. other porous inputs other source terms. radiator. temperature (tot. all others .4: PhaseSpecific and Mixture Conditions for the Eulerian Model (with the Mixture Turbulence Model) Type exhaust fan. other porous inputs source terms for turbulence. porosity. temperature) velocity. volume fraction. method. symmetry wall pressure farfield. fixed values for turbulence. temperature) nothing M ixture all others coord. temperature) all others axis. volume fraction. fan.Table 23. temperature) dir. flow direction. solid. pressure inlet Primary Phase (tot.
gran. fixed values. intake fan. temperature (tot. temperature (tot. temperature) velocity. all others . mass. temperature) nothing M ixture all others dir. solid. flow direction.9. pressure inlet Primary Phase turb. radiator. system. fan. volume fraction. mass flow inlet fluid all others shear condition not available shear condition not available all others not available momentum. parameters. temperature) velocity.Table 23. periodic. system. method. gran. temperature (tot. gran. porosity. spec. other porous inputs momentum and mass sources. turb. temperature) coord. porous jump. outflow. momentum and mass fixed values. flow direction. pressure outlet inlet vent. symmetry wall pressure farfield. all others velocity inlet all others axis. turb. volume fraction. other porous inputs porous zone. turb. temperature) nothing Secondary Phase volume fraction. (tot. parameters (tot. turb. sources. mass. parameters (tot. outlet vent. momentum. temperature) coord.5: PhaseSpecific and Mixture Conditions for the Eulerian Model (with the Dispersed Turbulence Model) Type exhaust fan.
porosity. all others velocity inlet velocity. other porous inputs momentum. momentum. turb.Table 23. temperature (tot. temperature) velocity. turb. temperature) dir. momentum. gran. intake fan. mass. fixed values. parameters (tot. fan. flow direction. outlet vent. temperature) nothing M ixture all others coord. method. sources. volume fraction. fixed values. porous jump. periodic. temperature) Secondary Phase volume fraction. turb. turb. radiator. turb. temperature (tot. pressure outlet inlet vent. turb. system. temperature) coord. temperature) all others axis. parameters (tot. gran. mass. parameters. all others . sources. other porous inputs porous zone.6: PhaseSpecific and Mixture Conditions for the Eulerian Model (with the PerPhase Turbulence Model) Type exhaust fan. spec. parameters.9. parameters (tot. parameters. solid. mass. turb. turb. symmetry wall pressure farfield. mass flow inlet fluid nothing all others shear condition not available shear condition not available all others not available momentum. flow direction. system. mass. pressure inlet Primary Phase turb. turb. temperature (tot. gran. volume fraction. outflow.
9: The Pressure Inlet Panel for a Mixture Note that only those conditions that apply to all phases. For i For a VOF calculation. measured inside the phase listed in the left column under Wall Adhesion in the Momentum tab of the Wall panel. the Pressure Inlet panel).Steps for Setting BoundaryConditions The steps you need to perform for each boundary are as follows: 1. to open the corresponding panel (e. if necessary. Figure 23. will appear in this panel.g.. if you enabled the Wall Adhesion option in . (See above for information about which conditions need to be set for the mixture.9. The contact angle (θw in Figure 23. and the corresponding panel will open automaticall y.9.10) or a UDF (see the UDF manual for more information). confirm the change (when prompted).g. specify the mixture boundary conditions. choose the correct zone type in the Type list.. (c) In the corresponding panel for the zone type you have selected (e. select mixture.9).. shown in Figure 23. 2.9.3.) (a) In the Phase dropdown list. the Phase Interaction panel. Select the boundary in the Zone list in the Boundary Conditions panel. you can specify the contact angle at the wall for each pair of phases as a consta nt (as shown in Figure 23. the Pressure Inlet panel. (b) If the current Type for this zone is correct. Set the conditions for the mixture at this boundary. otherwise.. click Set. as described above.3) is the angle between the wall and the tange nt to the interface at the wall.
example.10: The Wall Panel for a Mixture in a VOF Calculation with Wall Adhesion The default value for all pairs is 90 degrees.9. if you are setting the contact angle between the oil and air phases in the Wall panel shown in Figure 23. the interface is normal to the adjace nt wall). Figure 23. . θw is measured inside the oil phase.e..10. as is common with water in a glass. corresponds to water creeping up the side of a container. (d) Click OK when you are done setting the mixture boundary conditions.9. which is equivalent to no wall adhesion effects (i. A contact angle of 45◦ . for example.
(b) Click Set. will appear in this panel. when you select one of the individual phases (rather than the mixture).9. if necessary.. Note that only those conditions that apply to the individual phase.. i Note that..9.3.) (a) In the Phase dropdown list. and it applies to all of the individual phases. (See above for information about which conditions need to be set for the individual phases. as described above.11: The Pressure Inlet Panel for a Phase (c) Specify the conditions for the phase.. Figure 23. (d) Click OK when you are done setting the phasespecific boundary conditions.11). the zone type is specified for the mixture. select the phase (e. . to open the panel for this phase’s conditions (e. shown in Figure 23. only one type of zone appears in the Type list. It is not possible to assign phasespecific zone types at a given boundary.g.g. Set the conditions for each phase at this boundary. the Pressure Inlet panel. water).
In the To Zones list. . button.10. This will open the Copy BCs panel.matically result in the boundary conditions for the other phases and the mixture being copied as well. FLUENT will set all of the selected phase’s (or mixture’s) boundary conditions on the zones selected in the To Zones list to be the same as that phase’s conditions on the zone selected in the From Zone list. select the zone that has the conditions you want to copy.1. 4. 3.5: Copying Boundary Conditions for additional information about copying boundary conditions. In the Phase dropdown list.10 S ettin g U p the VO F M odel Steps for Copying BoundaryConditions The steps for copying boundary conditions for a multiphase flow are slightly different from those described in Section 7. i Note that copying the boundary conditions for one phase does not auto.10 Setting Up the VO F Model 23.1. select the zone or zones to which you want to copy the conditions. In the Boundary Conditions panel. Click Copy.) See Section 7. (You cannot copy a subset of the conditions. 2.. The modified steps are listed below: 1. select the appropriate VOF Scheme under VOF Parameters in the Multiphase Model panel. 5.5: Copying Boundary Conditions for a singlephase flow. You need to copy the conditions for each phase on each boundary of interest.23. click the Copy.1 Choosing a VO F Formulation To specify the VOF formulation to be used. including limitations.. In the From Zone list. The VOF formulations that are available in FLUENT are the Explicit and Implicit schemes. select the phase for which you want to copy the conditions (either mixture or one of the individual phases). such as only the thermal conditions. 23.
use the following text command: solve −→ set −→expert You will be asked a series of questions. Note that FLUENT will automatically turn on the unsteady formulation with firstorder discretization for time in the Solver panel. • Timede pendent with the donoracceptor interpolation scheme: This formulation should be used instead of the timede pendent formulation with the geometric re. Note that the CICSAM scheme or the modified HRIC scheme can be computationally inexpensive when compared to the geometric reconstruction scheme and improves the robustness and stability of the calculations. you should use the timede pendent explicit scheme. . then many more discretization schemes will be available for your selection. This formulation can also be used for other cases in which the geometric reconstruction scheme does not give satisfactory results. then select GeoReconstruct as the Volume Fraction Discretization scheme in the Solution Controls panel. For such cases. The DonorAcceptor scheme is used when Explicit is selected as the VOF Scheme in the Multiphase panel.mulation should be used whenever you are interested in the timeaccurate transie nt behavior of the VOF solution. You can now use this formulation by selecting DonorAcceptor as the Volume Fraction Discretization in the Solution Controls panel. or the flow calculation becomes unstable. it cannot be used for a hybrid mesh containing twisted hexahedral cells. Modified HRIC and CICSAM.M odelingM ultiphase Flows ExplicitSchemes • Timede pendent with the explicit interpolation scheme: Since the donoracceptor scheme is available only for quadrilateral and hexahedral meshes. To make it available. are available in the Solution Controls panel when the explicit VOF scheme is selected. The Volume Fraction discretizations. make sure Explicit is selected as the VOF Scheme in the Multiphase panel. • The CICSAM scheme gives interface sharpness of the same level as the geometric reconstruction scheme and is particularly suitable for flows with high viscosity ratios between the phases. this formulation is not available in the GUI.construction scheme if your mesh contains highly twisted hexahedral cells. Initially. To use this formulation. • Timede pendent with the geometric reconstruction interpolation scheme: This for. the donoracceptor scheme may provide more accurate results. one of which is Allow selection of all applicable discretization schemes? [no] If your response is yes. For such cases.
if you are using a conformal grid (i. and the final steadystate solution is not affected by the initial flow conditions and there is a distinct inflow boundary for each phase.. in order to obtain a sharper interface. described below.23.10 S ettin g U p the VO F M odel To use this formulation. To reduce this diffusivity. you may want to consider turning the geometric reconstruction scheme back on after calculating a solution with the implicit scheme. . i For the geometric reconstruction and donoracceptor schemes. it is recommended that you use the secondorder discretization scheme for the volume fraction equations. To use this formulation. menu item). you must ensure that there are no twosided (zerothickness) walls within the domain.8.6: Slitting Face Zones. then select Modified HRIC as the Volume Fraction Discretization in the Solution Controls panel. • Steadystate with the implicit interpolation scheme: This formulation can be used if you are looking for a steadystate solution. You should take the precautions described above to improve the sharpness of the interface. but the final steadystate solution is dependent on the initial flow conditions and/or you do not have a distinct inflow boundary for each phase. While the explicit timede pendent formulation is less computationally expensive than the geometric reconstruction scheme. Note that the implicit modified HRIC scheme can be used as a robust alternati ve to the explicit geometric reconstruction scheme. The issues discussed above for the explicit timede pendent formulation also apply to the implicit steadystate and timede pendent formulations. In addition.e. you are not interested in the intermediate transie nt flow behavior. To use this formulation. as described in Section 6. then select CICSAM as the Volume Fraction Discretization in the Solution Controls panel. select Implicit as the VOF Scheme in the Multiphase panel. i Im plicitSchemes • Timede pendent with the implicit interpolation scheme: This formulation can be used if you are looking for a steadystate solution and you are not interested in the intermediate transie nt flow behavior. If there are. the interface between phases will not be as sharp as that predicted with the geometric reconstruction scheme. you will need to slit them.. and enable an Unsteady calculation in the Solver panel (opened with the Define/Models/Solver. select Explicit as the VOF Scheme in the Multiphase panel. if the grid node locations are identical at the boundaries where two subdomains meet). select Implicit as the VOF Scheme..
turn on Volume of Fluid. Explicit. To start using the open channel flow boundary condition. .M odelingM ultiphase Flows Examples To help you determine the best formulation to use for your problem.. (a) Open the Operating Conditions panel. (a) Open the Multiphase Model panel..2 ModelingOpen ChannelFlows Using the VOF formulation. open channel flows can be modeled in FLUENT. perform the following: 1. select either Implicit. select Open Channel Flow. Define −→Operating Conditions. Under VOF Parameters. • flow around a ship’s hull Use the steadystate with the implicit interpolation scheme. 3. 23.10. • shape of the liquid interface in a centrifuge Use the timede pendent with the implicit interpolation scheme.. (c) Under VOF Scheme. (b) Turn on Gravity and set the gravitational acceleration fields.. examples that use different formulations are listed below: • jet breakup Use the explicit scheme (timede pendent with the geometric reconstruction scheme or the donoracceptor) if problems occur with the geometric reconstruction scheme. 2. Turn on gravity. Enable the volume of fluid model. Define −→ Models −→Multiphase. (b) Under Model.
This inlet group is recognized by the parameter Inlet Group ID.10 S ettin g U p the VO F M odel In order to set specific parameters for a particular boundary for open channel flows. Free Surface Level.g.23.10. Bottom Level Flow Rate Weighting pressure outlet mass flow inlet outflow DefiningInletGroups Open channel systems involve the flowing fluid (the secondary phase) and the fluid above it (the primary phase). If both phases enter through the separate inlets (e. i In threephase flows. Table 23. Free Surface Level. Secondary Phase for Inlet. only one secondary phase is allowed to pass through one inlet group. On the other hand. .1 summarizes the types of boundaries available to the open channel flow boundary condition.g.10. Bottom Level Inlet Group ID.. then the inlet itself represents the inlet group. Free Surface Level. Secondary Phase for Inlet. these two inlets form an inlet group. Table 23. For more information on setting boundary condition parameters. turn on the Open Channel Flow option in the corresponding boundary condition panel. which will be same for both the inlets that make up the inlet group. see Chapter 7: Boundary Conditions . inletphase2 and inletphase1).1: Open Channel Boundary Parameters for the VOF Model Boundary Type pressure inlet Parameter Inlet Group ID. Velocity Magnitude Outlet Group ID.. and the additional parameters needed to model open channel flow. Pressure Specification Method. if both the phases enter through the same inlet (e. inletcombined). Bottom Level. Flow Specification Method.
For instance.M odelingM ultiphase Flows DefiningOutletGroups Outletgroups can be defined in the same manner as the inlet groups. The first inlet group consists of water and air entering through the same inlet (a single face zone). consider the case of two inlet groups for a particular problem. In this case. then the Outlet Group ID will be the same for each outlet of that group. the Outlet Group ID is used to identify the different outlets that are part of the same outlet group. In the case where the same outlet group has separate outlets (different face zones) for each phase. when both phases enter through the same outlet (single face zone). separate outlets for each phase are not recommended in threephase flows. • You should specify a different Inlet Group ID for each distinct inlet group. when both phases enter through the same inlet (single face zone).. For instance.e. but each uses a different inlet (oilinlet and airinlet) for each phase. In the case where the same inlet group has separate inlets (different face zones) for each phase. The second inlet group consists of oil and air entering through the same inlet group. In this case. i. . then those phases are part of one inlet group and you would set the Inlet Group ID to 1 for that inlet (or inlet group). use the following guidelines: • Since the Inlet Group ID is used to identify the inlets of the same inlet group. then those phases are part of one outlet group and you would set the Outlet Group ID to 1 for that outlet (or outlet group). Setting the OutletGroup For pressure outlet boundaries. the outlet should represent the outlet group. Bottom Level. then the Inlet Group ID will be the same for each inlet of that group. you would specify the same Inlet Group ID of 2 for both of the inlets that belong to the inlet group. Setting the InletGroup For pressure inlets and mass flow inlets. general information such as Free Surface Level. you would specify an inlet group ID of 1 for that inlet (or inlet group). i In threephase flows. or the mass flow rate for each phase should be the same for each inlet of the same inlet group. When specifying the inlet group. For example. the Inlet Group ID is used to identify the different inlets that are part of the same inlet group.
Likewise. when all the phases are leaving through the same outlet. consider the case of two outlet groups for a particular problem. If the liquid’s free surface level lies above the origin. and pressure inlet.10 S ettin g U p the VO F M odel When specifying the outlet group.23. you would specify an outlet number of 1 for that outlet (or outlet group). if the liquid’s free surface level lies below the origin. In this case. 2. but each uses a different outlet (oiloutlet and airoutlet) for each phase. We can simply calculate the free surface level in two steps: 1. Determining Free Surface Level the For the appropriate boundary. general information such as Free Surface Level or Bottom Level should be the same for each outlet of the same outlet group. Determine the absolute value of height from the free surface to the origin in the direction of gravity. then the Free Surface Level is positive (see Figure 23.1). then the Free Surface Level is negati ve. For example.10. you would specify the same Outlet Group ID of 2 for both of the outlets that belong to the outlet group. . use the following guidelines: • Since the Outlet Group ID is used to identify the outlets of the same outlet group. In this case. is represented by ylocal in Equation 23. including pressure outlet. i For threephase flows. • You should specify a different Outlet Group ID for each distinct outlet group. mass flow inlet.101) where −→a is the position vector of any point on the free surface. Here we assume a horizontal free surface that is normal to the direction of gravity. The second outlet group consists of oil and air exiting through the same outlet group. ylocal = −(−→a · gˆ) (23. the outlet should consist only of a single face zone. This parameter is available for all relevant boundaries. Apply the correct sign based on whether the free surface level is above or below the origin. and gˆ is the unit vector in the direction of the force of gravity. you need to specify the Free Surface Level value. The Free Surface Level. The first inlet group consists of water and air exiting from the same outlet (a single face zone).325.
10. 2. then the Bottom Level is positive (see Figure 23. and gˆ is the unit vector of gravity. We can simply calculate the bottom level in two steps: 1.325. you need to specify the Bottom Level value.1). then the Bottom Level is negati ve.102) −→ where b is the position vector of any point on the bottom of the channel.M odelingM ultiphase Flows Determining Bottom Level the For the appropriate boundary. Free Surface Level (positive) g Bottom Level (positive) Reference Level Origin Figure 23. −→ ybottom = −( b · gˆ) (23. Apply the correct sign based on whether the bottom level is above or below the origin. if the channel’s bottom lies below the origin. is represented by a relation similar to Equation 23. This parameter is available for all relevant boundaries. mass flow inlet.1: Determining the Free Surface Level and the Bottom Level . Determine the absolute value of depth from the bottom level to the origin in the direction of gravity. Here we assume a horizontal free surface that is normal to the direction of gravity.10. and pressure inlet. including pressure outlet. Likewise. If the channel’s bottom lies above the origin. The Bottom Level.
Note that only one secondary phase is allowed to pass through one inlet group. you would choose oil as the secondary phase. a hydrostatic pressure profile) • by specifying the consta nt pressure i This option is not available in the case of threephase flows since the pres. This is to be specified as the magnitude of the upstream inlet velocity in the flow. is used as an option for describing the flow. and oil and water as the secondary phases. Consider also that there are two inlet groups: • water and air • oil and air For the former inlet group.e. the Secondary Phase for Inlet field is significant in cases of threephase flows.103) where V is the velocity magnitude and g is the gravity magnitude. you would choose water as the secondary phase. the outlet pressure can be specified in one of two ways: • by prescribing the local height (i. along with the velocity. For the latter inlet group. Determining Secondary Phase for the Inlet the For pressure inlets and mass flow inlets.. Determining Velocity Magnitude the Pressure inlet boundaries require the Velocity Magnitude for calculating the dynamic pressure at the boundary. .Specifying the Total Height The total height.sure on the boundary is taken from the neighboring cell. Consider a problem involving a threephase flow consisting of air as the primary phase. i Choosing the Pressure SpecificationMethod For a pressure outlet boundary. The total height is given as ytot = ylocal + V2 2g (23.
Recommendationsfor Setting Up an Open ChannelFlow Problem The following list represents a list of recommendations for solving problems using the open channel flow boundary condition: • In the cases where the inlet group has a different inlet for each phase of fluid. For finer meshes.e. subcritical (upstream and downstream). the If the final conditions of the flow can be predicted by other means. • The conservation of the Bernoulli integral does not provide the conservation of mass flow rate for the pressure boundary. time is very dependent on the initial conditions. The pressure boundary should be selected when steady and nonoscillating drag is the main objective.gent solution. you can start the solution by assuming outlet corresponds to the level of liquid at the inlet. for problems having consta nt mass flow rate. • Only the heavier phase should be selected as the secondary phase. • Specifying the top boundary as the pressure outlet can sometimes lead to a diver. • The solution begins with an estimated pressure profile at the outlet boundar y. • In the case of threephase flows. that the level of liquid at the The convergence and solution When the flow is completely applications for instance. In the case of a coarser mesh. and Mass Flow Rate) for each inlet should correspond to all other inlets that belong to the inlet group. only one secondary phase is allowed to enter through one inlet group (i. So. the solution time can be significantly reduced by using the proper boundary condition. In general. then the parameter values (such as Free Surface Level.. the mixed inflow of different secondary phases is not allowed). . the mass flow rate boundary condition is a better option.Limitations The following list summarizes some issues and limitations associated with the open channel boundary condition. Bottom Level. the mass flow rate comes closer to the actual value. there can be a significant difference in mass flow rate from the actual mass flow rate. in marine above approach is recommended. This may be due to the corner singularity at the pressure boundary in the air region or due to the inability to specify local flow direction correctly if the air enters through the top locally.
a smaller time step is recommended. then patching some regions with inlet Free Surface Level values and some regions with outlet Free Surface Level values could be useful for some problems. i In general.e. i . to consider how your choice will affect the ease of problem setup.11. See Sections 23.10. Patch the inlet velocity again in the full domain. a material that uses the compressible ideal gas law for density) for more than one of the phases. Recall that only one of the phases can be a compressible ideal gas. Patch the domain using a volume fraction value of the secondary phase to 1. Be sure that you do not select a compressible ideal gas material (i. up to the Free Surface Level specified at the inlet. You can increase the time step once the solution becomes more stable. and providing the inlet velocity. 3. 23. it may be more convenient to make that phase a secondary phase. Also.10. If the Free Surface Level values are different at the inlet and outlet. you can specify the primary and secondary phases whichever way you prefer. then perform the following initialization procedure: 1.• The initialization procedure is very critical in the open channel analysis. if one of the phases is a compressible ideal gas.5 and 23.3 Definingthe Phases for the VO F Model Instructions for specifying the necessary information for the primary and secondary phases and their interaction in a VOF calculation are provided below. 2.3 for details. It is a good idea.. If you are interested in the final steady state solution. but now the initial conditions are dependent on the user. • For the initial stability of the solution. if you are planning to patch an initial volume fraction of 1 for one phase in a portion of the domain. For example. Initialize the domain by setting the volume fraction of the secondary phase to 0. especially in more complicated problems. it is recommended that you specify it as the primary phase to improve solution stabili ty. The same steps for initialization are also recommended for unsteady flows.
10.... remember to click Change before closing the Material panel. 4. (b) In the Material panel.2: The Primary Phase Panel 3.. In the Primary Phase panel. and the Material panel will open.) i If you make changes to the properties. Define the material properties for the Phase Material. Figure 23.Definingthe Prim aryPhase To define the primary phase in a VOF calculation. and Section 23. Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list. to open the Primary Phase panel (Figure 23. (a) Click Edit.10. check the properties. 5. (See Chapter 8: Physical Properties for general information about setting material properties.10. Click Set. perform the following steps: 1.2). .10. 6. Section 23. enter a Name for the phase. and modify them if necessary.5: Modeling Compressible Flows for specific information related to compressible VOF calculations. 2. Select phase1 in the Phase list.. Click OK in the Primary Phase panel.6: Modeling Solidification/Melting for specific information related to melting/solidification VOF calculations.
2. to open the Secondary Phase panel (Figure 23.3: The Secondary Phase Panel for the VOF Model 3. 6.3).10.10. following the procedure outlined above for setting the material properties for the primary phase. phase2) in the Phase list. Click Set. Click OK in the Secondary Phase panel.g. Select the phase (e. 4. Figure 23. Define the material properties for the Phase Material. . Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list.. In the Secondary Phase panel..Defininga Secondary Phase To define a secondary phase in a VOF calculation.. enter a Name for the phase. 5. perform the following steps: 1.
8: When Surface Tension Effects Are Importa nt. i Note that the calculation of surface tension effects will be more accurate if you use a quadrilateral or hexahedral mesh in the area(s) of the computational domain where surface tension is significant.3.10. or the Weber number. If you cannot use a quadrilateral or hexahedral mesh for the entire domain.3. the importance of surface tension effects depends on the value of the capillary number.4). see Section 23. to open the Phase Interaction panel (Figure 23. click Interaction..4: The Phase Interaction Panel for the VOF Model (Surface Tension Tab) ..317).8: Surface Tension and Wall Adhesion. with quadrilaterals or hexahedra in the affected areas.3. Ca (defined by Equation 23. then you should use a hybrid mesh.10. Surface tension effects can be neglected if Ca 1 or We 1. Figure 23. We (defined by Equation 23. FLUENT also offers an option to use VOF gradients at the nodes for curvature calculations on meshes when more accuracy is desired. For more information. as described in Section 23.316). If you want to include the effects of surface tension along the interface between one or more pairs of phases.8: Surface Tension and Wall Adhesion.Including Surface Tension and Wall Adhesion Effects As discussed in Section 23.
as discussed in Section 25. 3. If you choose the Explicit scheme. piecewise linear. When Wall Adhesion is enabled. piecewise polynomial. For each pair of phases between which you want to include the effects of surface tension.. You also have control over the time step used for the volume fraction calculation. Alternati vely you can specify a temperature dependent. Models .4: Including Body Forces for details. representing no surface tension effects along the interface between the two phases. Define −→ −→Solver. FLUENT will turn on the firstorder unsteady formulation for you automaticall y. Click the Surface Tension tab.4 Setting TimeDependent Paramete s for the VO F M odel r If you are using the timede pendent VOF formulation in FLUENT. This treatme nt improves solution convergence by accounting for the partial equilibrium of the pressure gradient and surface tension forces in the momentum equations. depending upon your inputs to the model. If you want to include wall adhesion.10.8: Defining Multiphase Boundary Conditions ). an explicit solution for the volume fraction is obtained either once each time step or once each iteration. 23.9. it is recommended that you also turn on the Implicit Body Force treatme nt for the Body Force Formulation in the Multiphase Model panel.Perform the following steps to include surface tension (and. you will need to specify the contact angle at each wall as a boundary condition (as described in Section 23. so you need not visit the Solver panel yourself. specify a consta nt surface tension coefficient. i For calculations involving surface tension. wall adhesion) effects along the interface between one or more pairs of phases: 1. See Section 23. and the separate UDF Manual for more information on userdefined functions.3. turn on the Wall Adhesion option..8: Surface Tension and Wall Adhesion for more information on surface tension. you will need to enable the Unsteady option in the Solver panel (and choose the appropriate Unsteady Formulation. 2. if appropriate. or a userdefined surface tension coefficient.9. See Section 23. All surface tension coefficients are equal to 0 by default. polynomial.17.1: User Inputs for TimeDependent Problems). To compute a timede pendent VOF solution.
since the volume fraction fields will not change from iteration to iteration. although at a greater computational cost. FLUENT will solve the volume fraction equation(s) once for each time step. and requires more computational effort per time step than the default choice. i If you are using using the solve vof every iteration? and/or remeshing. Based upon this time and your input for the maximum allowed Courant Number in the Multiphase Models panel.port equations will not be completely updated each iteration. FLUENT will refine the time step for VOF automaticall y. sliding meshes. based on your input for the maximum Courant Number allowed near the free surface. if the maximum allowed Coura nt number is 0. FLUENT divides the volume of each cell by the sum of the outgoing fluxes. use the text command: define −→ models −→ multiphase −→ and select vof as the model.There are two inputs for the timede pendent calculation for the VOF model: • By default. Note that these inputs are not required when the implicit scheme is used. as described above. or dynamic meshes with layering option will yield more accurate results. This means that the convective flux coefficients appearing in the other trans.25 (the default). • When FLUENT performs a timede pendent VOF calculation. The Coura nt number is a dimensionless number that compares the time step in a calculation to the characteristic time of transit of a fluid element across a control volume: ∆t ∆xcell /v fluid (23. enter yes. The resulting time represents the time it would take for the fluid to empty out of the cell. This choice is the less stable of the two.104) In the region near the fluid interface. a time step is computed for use in the VOF calculation. When prompted to solve vof every iteration?. The smallest such time is used as the characteristic time of transit for a fluid element across a control volume. For example. the convective flux coefficients in the other trans port equations will be updated based on the updated volume fractions at each iteration. . the time step will be chosen to be at most onefourth the minimum transit time for any cell near the interface. the time step used for the volume fraction calculation will not be the same as the time step used for the rest of the trans port equations. If you want FLUENT to solve the volume fraction equation(s) at every iteration within a time step. When FLUENT solves these equations every iteration.
See Section 9. • If you specify the total pressure at a boundary (e. note the following: • Only one of the phases can be defined as a compressible ideal gas (i. and as static temperature for the other phases (which are incompressible)..10.6: Compressible Flows for more information about compressible flows. you will need to redefine the conditions at the mass flow inlets. See Section 23.8: Defining Multiphase Boundary Conditions for more information on defining conditions for a mass flow inlet in VOF multiphase calculations. you can select the ideal gas law for the density of only one phase’s material).g. you will need to specify mass flow or mass flux for each individual phase. it is better (although not re. There is no limitation on using compressible liquids using userdefined functions.quired) if the primary phase is a compressible ideal gas. i Note that if you read a case file that was set up in a version of FLUENT previous to 6.e. for a pressure inlet or intake fan) the specified value for temperature at that boundary will be used as total temperature for the compressible phase. .. • When using the VOF model.1.9. • For each mass flow inlet. for stability reasons.5 ModelingCompressibleFlows If you are using the VOF model for a compressible flow.23.
and solidus temperature (Tsolidus ) to zero.1 Definingthe Phases for the M ixture Model Instructions for specifying the necessary information for the primary and secondary phases and their interaction for a mixture model calculation are provided below.11.3: Defining the Primary Phase for details. . Select the phase (e.e. Be sure that you do not select a compressible ideal gas material (i. a material that uses the compressible ideal gas law for density) for more than one of the phases. liquid or vapor) secondary phase in a mixture multiphase calculation.. See Chapter 24: Modeling Solidification and Melting for more information about melting and solidification. perform the following steps: 1. 2.3: Modeling Compressible Flows for details... See Section 23. note the following: • It is possible to model melting or solidification in a single phase or in multiple phases. liquidus temperature (Tliquidus ).11..11 Setting Up the M ixture odel M 23.6 ModelingSolidification/Melting If you are including melting or solidification in your VOF calculation.10.10. 23. to open the Secondary Phase panel (Figure 23. Click Set. See Section 23. phase2) in the Phase list. you must set the late nt heat (L).e. • For phases that are not melting or solidifying.g.23..1). Defininga Nongranular Secondary Phase To define a nongranular (i.11. i Recall that only one of the phases can be a compressible ideal gas. Definingthe Prim aryPhase The procedure for defining the primary phase in a mixture model calculation is the same as for a VOF calculation.
11 Setting Up the M ixture odel M Figure 23. and it will not be available to you. 6. 5.3: Defining the Primary Phase). In the Secondary Phase panel. droplets. enter a Name for the phase.1: The Secondary Phase Panel for the Mixture Model 3.11. See the separate UDF Manual for details about userdefined functions. Click OK in the Secondary Phase panel. In the Secondary Phase panel. specify the Diameter of the bubbles. this input is not necessary. . You can specify a consta nt value.412). Define the material properties for the Phase Material. 7. following the same procedure you used to set the material properties for the primary phase (see Section 23. or particles of this phase (dp in Equation 23. you need to specify only the density. as mentioned in Section 23. Note that when you are using the mixture model without slip velocity. For a particulate phase (which must be placed in the fluid materials category. you can ignore the values for the other properties.23.9: Steps for Using a Multiphase Model). or use a userdefined function. Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list. since they will not be used.10. 4.
M odelingM ultiphase Flows Defininga G ranular Secondary Phase To define a granular (i. Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list. perform the following steps: 1. enter a Name for the phase. . In the Secondary Phase panel..g.. Select the phase (e.11. phase2) in the Phase list. Click Set.2)...e. 2. 4. to open the Secondary Phase panel (Figure 23.2: The Secondary Phase Panel for a Granular Phase Using the Mixture Model 3. Figure 23.11. particulate) secondary phase in a mixture model multiphase calculation.
∇ps . since they will not be used. following the same procedure you used to set the material properties for the primary phase (see Section 23. ∂Ps ∂αs (23.420. You can select constant in the dropdown list and specify a consta nt value. See the separate UDF Manual for details about userdefined functions. Define the material properties for the Phase Material. you can ignore the values for the other properties.9: Steps for Using a Multiphase Model). select syamlalobrien to compute the value using Equation 23.111) . or userdefined option. as mentioned in Section 23. select gidaspow to compute the value using Equation 23. in the granularphase momentum equation. Solids Pressure specifies the pressure gradient term. Choose either the lunetal. See the separate UDF Manual for details about userdefined functions. the syamlalobrien. Choose either the derived or userdefined options. the maahmadi.421. you need to specify only the density. the arastoopour. or select userdefined to use a userdefined function. or a userdefined option.3: Defining the Primary Phase). 6.11 Setting Up the M ixture odel M 5.23. Elasticity Modulus is defined as G= with G ≥ 0.10. You can select constant (the default) in the dropdown list and specify a consta nt value. 7. specify the following properties of the particles of this phase: Diameter specifies the diameter of the particles. Turn on the Granular option. Radial Distribution specifies a correction factor that modifies the probability of collisions between grains when the solid granular phase becomes dense. In the Secondary Phase panel. or the userdefined option.418). the maahmadi. Choose either the lunetal. i Note that all properties for granular flows can utilize userdefined functions (UDFs). the constant.kin in Equation 23. or select userdefined to use a userdefined function. Granular Temperature specifies temperature for the solids phase and is proportional to the kinetic energy of the random motion of the particles. For a granular phase (which must be placed in the fluid materials category. Granular Viscosity specifies the kinetic part of the granular viscosity of the particles (µ s. the syamlalobrien. Choose either the algebraic.
12. which is the default value in FLUENT.. smaller spheres can fill the small gaps between larger spheres.max ). to open the Phase Interaction panel (Figure 23. the packing limit is about 0.63. For monodispersed spheres. 8.3).11. Figure 23.11. so you may need to increase the maximum packing limit. and then click the Drag tab. In polydispersed cases.. DefiningDrag Between Phases For mixture multiphase flows with slip velocity.3: The Phase Interaction Panel for the Mixture Model (Drag Tab) . Click OK in the Secondary Phase panel. you can specify the drag function to be used in the calculation. See Section 23.2: Defining the Phases for the Eulerian Model. To specify drag laws. click Interaction. however.4. The functions available here are a subset of those discussed in Section 23.5: Relati ve (Slip) Velocity and the Drift Velocity for more information.Packing Limit specifies the maximum volume fraction for the granular phase (αs.
4. • Select none if the secondary phase has the same velocity as the primary phase (i.. and then click the Slip tab.4: The Phase Interaction Panel for the Mixture Model (Slip Tab) Under Slip Velocity. • Select userdefined to use a userdefined function for the slip velocity. • Select maninnenetal (the default) to use the algebraic slip method of Manninen et al. Figure 23. [229].11. See the separate UDF Manual for details.Definingthe Slip Velocity If you are solving for slip velocities during the mixture calculation. click Interaction. described in Section 23.5: Relative (Slip) Velocity and the Drift Velocity. you can specify the slip velocity function for each secondary phase with respect to the primary phase by choosing the appropriate item in the adjace nt dropdown list.4). and you want to modify the slip velocity definition.e. no slip velocity).11... to open the Phase Interaction panel (Figure 23. .
you will specify three parameters to be used in the calculation of mass transfer due to cavitation. turn on the Cavitation option in the Mass tab of the Phase Interaction panel.23. . it is possible to include the effects of cavitation.2 IncludingCavitationEffects For mixture model calculations.7.11.11. Under Cavitation Parameters in the Phase Interaction panel. which depends on the purity of the liquid. Figure 23. depending mainly on temperature. the Surface Tension Coefficient .5: The Phase Interaction Panel for Mass Transfer with Cavitation Enabled When you are using FLUENT’s cavitation model. NonCondensable Gas Mass Fraction is the mass fraction of dissolved gases.715). The default value of psat is 2540 Pa. To enable the cavitation model. the vaporization pressure for water at ambient temperature. set the Vaporization Pressure (psat in Equation 23. and the NonCondensable Gas Mass Fraction . using FLUENT’scavitation model described in Section 23. Note that psat and the surface tension are properties of the liquid.4: Cavitation Models.
you should choose the constantrate option in the Mechanism dropdown list and leave the value as 0 for both mass transfer mechanisms before turning on the Cavitation option. • mass transfer from vapor to liquid.7: Including Mass Transfer Effects.When multiple species are included in one or more secondary phases.11.11. Figure 23. It may be noted. In the Phase Interaction panel (Figure 23.6).9. that for cavitation problems. the mass transfer mechanism must be defined before turning on the Cavitation option.6: The Phase Interaction Panel for Mass Transfer with Cavitation Disabled . This is defined in the same way as described Section 23. or the heat transfer due to phase change needs to be taken into account. however. at least two mass transfer mechanisms are defined: • mass transfer from liquid to vapor.
See Section 9. .8: Defining Multiphase Boundary Conditions for more information on defining conditions for a mass flow inlet in mixture multiphase calculations.3 ModelingCompressibleFlows If you are using the mixture model for a compressible flow. note the following: • Only one of the phases can be defined as a compressible ideal gas (i. and as static temperature for the other phases (which are incompressible). i Note that if you read a case file that was set up in a version of FLUENT previous to 6. • For each mass flow inlet. you can select the ideal gas law for the density of only one phase’s material). you will need to redefine the conditions at the mass flow inlets.11.e.9. See Section 23..23.. you will need to specify mass flow or mass flux for each individual phase.g. for a pressure inlet or intake fan) the specified value for temperature at that boundary will be used as total temperature for the compressible phase. There is no limitation on using compressible liquids using userdefined functions.1.6: Compressible Flows for more information about compressible flows. • If you specify the total pressure at a boundary (e.
See Section 23.10. Some suggestions for simplifying a multiphase flow problem are listed below: • Use a hexahedral or quadrilateral mesh (instead of a tetrahedral or triangular mesh). You may find that even a very simple approximation will provide you with useful information about your problem.12 Setting Up the EulerianM odel 23. you can start with simple approximations and work your way up to the final form of the problem definition.23. See Section 23.1 Additional Guidelines EulerianMultiphase for Simulations Once you have determined that the Eulerian multiphase model is appropriate for your problem (as described in Section 23.2: Choosing a General Multiphase Model). Before setting up your problem. .2 Definingthe Phases for the EulerianM odel Instructions for specifying the necessary information for the primary and secondary phases and their interaction for an Eulerian multiphase calculation are provided below. The required computational effort depends strongly on the number of trans port equations being solved and the degree of coupling. you should consider the computational effort required to solve your multiphase problem.14. Instead of trying to solve your multiphase flow in all of its complexity on your first solution attempt.12. • Reduce the number of phases. For the Eulerian multiphase model. 23. which has a large number of highly coupled trans port equations. Definingthe Prim aryPhase The procedure for defining the primary phase in an Eulerian multiphase calculation is the same as for a VOF calculation. computational expense will be high.4: Eulerian Model for more solution strategies for Eulerian multiphase calculations. try to reduce the problem stateme nt to the simplest form possible.3: Defining the Primary Phase for details.12 Setting Up the EulerianM odel 23.12.
following the same procedure you used to set the material properties for the primary phase (see Section 23. 4. perform the following steps: 1. to open the Secondary Phase panel (Figure 23.e. 5. You can specify a consta nt value. 6. 2. In the Secondary Phase panel. or use a userdefined function.. See the separate UDF Manual for details about userdefined functions.M odelingM ultiphase Flows Defininga Nongranular Secondary Phase To define a nongranular (i.1: The Secondary Phase Panel for a Nongranular Phase 3. Click Set.. specify the Diameter of the bubbles or droplets of this phase. Click OK in the Secondary Phase panel. Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list. Figure 23. enter a Name for the phase. . liquid or vapor) secondary phase in an Eulerian multiphase calculation. Define the material properties for the Phase Material. Select the phase (e. In the Secondary Phase panel.12.3: Defining the Primary Phase )...g.10.1). phase2) in the Phase list. 7.12.
23.12 Setting Up the EulerianM odel
Defininga G ranular Secondary Phase
To define a granular (i.e., particulate) secondary phase in an Eulerian multiphase calculation, perform the following steps: 1. Select the phase (e.g., phase2) in the Phase list. 2. Click Set... to open the Secondary Phase panel (Figure 23.12.2).
Figure 23.12.2: The Secondary Phase Panel for a Granular Phase
3. In the Secondary Phase panel, enter a Name for the phase. 4. Specify which material the phase contains by choosing the appropriate material in the Phase Material dropdown list.
M odelingM ultiphase Flows 5. Define the material properties for the Phase Material, following the same procedure you used to set the material properties for the primary phase (see Section 23.10.3: Defining the Primary Phase). For a granular phase (which must be placed in the fluid materials category, as mentioned in Section 23.9: Steps for Using a Multiphase Model), you need to specify only the density; you can ignore the values for the other properties, since they will not be used. Note that all properties for granular flows can utilize userdefined functions (UDFs). See the separate UDF Manual for details about userdefined functions. 6. Turn on the Granular option. 7. (optional) Turn on the Packed Bed option if you want to freeze the velocity field for the granular phase. Note that when you select the packed bed option for a phase, you should also use the fixed velocity option with a value of zero for all velocity components for all interior cell zones for that phase. 8. Specify the Granular Temperature Model. Choose either the default Phase Property option or the Partial Differential Equation option. See Section 23.5.8: Granular Temperature for details. 9. In the Secondary Phase panel, specify the following properties of the particles of this phase: Diameter specifies the diameter of the particles. You can select constant in the dropdown list and specify a consta nt value, or select userdefined to use a userdefined function. See the separate UDF Manual for details about userdefined functions. Granular Viscosity specifies the kinetic part of the granular viscosity of the particles (µ s,kin in Equation 23.562). You can select constant (the default) in the dropdown list and specify a consta nt value, select syamlalobrien to compute the value using Equation 23.564, select gidaspow to compute the value using Equation 23.565, or select userdefined to use a userdefined function. Granular Bulk Viscosity specifies the solids bulk viscosity (λq in Equation 23.56). You can select constant (the default) in the dropdown list and specify a consta nt value, select lunetal to compute the value using Equation 23.566, or select userdefined to use a userdefined function. Frictional Viscosity specifies a shear viscosity based on the viscousplastic flow (µs,fr in Equation 23.562). By default, the frictional viscosity is neglected, as indicated by the default selection of none in the dropdown list. If you want to include the frictional viscosity, you can select constant and specify a consta nt value, select schaeffer to compute the value using Equation 23.567, select johnsonetal to compute the value using Equation 23.572, or select userdefined to use a userdefined function.
i
23.12 Setting Up the EulerianM odel Angle of Internal Friction specifies a consta nt value for the angle φ used in Schaeffer’s expression for frictional viscosity (Equation 23.567). This parameter is relevant only if you have selected schaeffer or userdefined for the Frictional Viscosity. Frictional Pressure specifies the pressure gradient term, ∇Pf rictio n , in the granularphase momentum equation. Choose none to exclude frictional pressure from your calculation, johnsonetal to apply Equation 23.572, syamlalobrien to apply Equation 23.530, basedktgf, where the frictional pressure is defined by the kinetic theory [79]. The solids pressure tends to a large value near the packing limit, depending on the model selected for the radial distribution function. You must hook a userdefined function when selecting the userdefined option. See the separate UDF manual for information on hooking a UDF. Frictional Modulus is defined as G= ∂Pf riction ∂αf riction (23.121)
with G ≥ 0, which is the derived option. You can also specify a userdefined function for the frictional modulus. Friction Packing Limit specifies the maximum volume fraction for the granular phase (αs,max ). For monodispersed spheres, the packing limit is about 0.63, which is the default value in FLUENT. In polydispersed cases, however, smaller spheres can fill the small gaps between larger spheres, so you may need to increase the maximum packing limit. Granular Conductivity specifies the solids granular conductivity (kΘs in Equation 23.575). You can select syamlalobrien to compute the value using Equation 23.576, select gidaspow to compute the value using Equation 23.577, or select userdefined to use a userdefined function. Note, however, that FLU ENT currently uses an algebraic relation for the granular temperature. This has been obtained by neglecting convection and diffusion in the trans port equation, Equation 23.575 [363]. Granular Temperature specifies temperature for the solids phase and is proportional to the kinetic energy of the random motion of the particles. Choose either the algebraic, the constant, or userdefined option. Solids Pressure specifies the pressure gradient term, ∇ps , in the granularphase momentum equation. Choose either the lunetal, the syamlalobrien, the maahmadi, none, or a userdefined option. Radial Distribution specifies a correction factor that modifies the probability of collisions between grains when the solid granular phase becomes dense. Choose either the lunetal, the syamlalobrien, the maahmadi, the arastoopour, or a userdefined option.
you will also need to specify the restitution coefficient(s) for particle collisions. Click OK in the Secondary Phase panel. ∂Ps ∂αs (23. For monodispersed spheres. click Interaction.12.3: The Phase Interaction Panel for the Eulerian Model . Definingthe Interaction Between Phases For both granular and nongranular flows.Elasticity Modulus is defined as G= with G ≥ 0.max ). however. to open the Phase Interaction panel (Figure 23. To specify these parameters.122) Packing Limit specifies the maximum volume fraction for the granular phase (αs.12. the packing limit is about 0. which is the default value in FLUENT. Figure 23.. so you may need to increase the maximum packing limit. It is also possible to include an optional lift force and/or virtual mass force (described below) for both granular and nongranular flows.. For granular flows. you will need to specify the drag function to be used in the calculation of the momentum exchange coefficients. In polydispersed cases. 10.63.3). smaller spheres can fill the small gaps between larger spheres.
The symmetric model is recommended for flows in which the secondary (dispersed) phase in one region of the domain becomes the primary (continuous) phase in another.522. • Select morsialexander to use the fluidfluid drag function described by Equa. the air is the dispersed phase in the bottom half of the container. if air is injected into the bottom of a container filled halfway with water.tion 23. The SyamlalO’Brien model is recommended for use in conjunc.tion 23. • Select wenyu to use the fluidsolid drag function described by Equation 23.541. The Schiller and Naumann model is the default method. . but calculations with this model may be less stable than with the other models. The Wen and Yu model is applicable for dilute phase flows. adjusting the function definition frequently over a large range of Reynolds numbers. • Select symmetric to use the fluidfluid drag function described by Equation 23. the air is the continuous phase. The Morsi and Alexander model is the most complete. • Select schillernaumann to use the fluidfluid drag function described by Equa.543. The symmetric SyamlalO’Brien model is appropriate for a pair of solid phases.Specifyingthe DragFunction FLUENT allows you to specify a drag function for each pair of phases.tion with the SyamlalO’Brien model for granular viscosity. 2.518.539. and then specify the value in the text field. For each pair of phases. The Gidaspow model is recommended for dense fluidized beds. • Select gidaspow to use the fluidsolid drag function described by Equation 23. select the appropriate drag function from the corresponding dropdown list. Perform the following steps: 1. For example. • Select syamlalobriensymmetric to use the solidsolid drag function described by Equation 23. • Select syamlalobrien to use the fluidsolid drag function described by Equa. in which the total secondary phase volume fraction is significantly lower than that of the primary phase. in the top half of the container. • Select constant to specify a consta nt value for the drag function. Click the Drag tab to display the Drag Function inputs. and it is acceptable for general use in all fluidfluid multiphase calculations.tion 23.531.527.
. specify a consta nt restitution coefficient.g. you may want to include this effect. In addition to specifying the restitution coefficient for collisions between each pair of granular phases. i Note that the lift force will be more significant for larger particles. 2. Thus. These lift forces act on a particle. the inclusion of lift forces is not appropriate for closely packed particles or for very small particles. Includingthe L iftForce For both granular and nongranular flows. or bubble mainly due to velocity gradients in the primaryphase flow field.• Select userdefined to use a userdefined function for the drag function (see the separate UDF Manual for details). so there is no reason to include it.544). or bubbles.9 by default..543 and ess in Equation 23. • If you want to temporarily ignore the interaction between two phases. you need to specify the coefficients of restitution for collisions between particles (els in Equation 23. Perform the following steps: 1. All restitution coefficients are equal to 0. droplet.58) on the secondary phase particles. you will also specify the restitution coefficient for collisions between particles of the same phase. the lift force is insignificant compared to the drag force. it is possible to include the effect of lift forces (F~lift in Equation 23. droplets. if the phases separate quickly). If the lift force is significant (e. For each pair of phases. Click the Collisionstab to display the Restitution Coefficient inputs. but the FLUENT model assumes that the particle diameter is much smaller than the interparticle spacing. In most cases. select none. Specifyingthe Restitution Coefficients(G ranular Flow Only) For granular flows.
2.To include the effect of lift forces. turn on the Virtual Mass option in the Phase Interaction panel. To include the effect of the virtual mass force.g. Click the Lift tab to display the Lift Coefficient inputs. The virtual mass effect will be included for all secondary phases. and then specify the value in the text field. Note that. • Select none (the default) to ignore the effect of lift forces. Including the V irtualM ass Force For both granular and nongranular flows. The virtual mass effect is significant when the secondary phase densi ty is much smaller than the primary phase density (e. For each pair of phases. it is possible to include the “virtual mass force” (F~vm in Equation 23. .59) that is present when a secondary phase accelerates relative to the primary phase. or bubble are due mainly to velocity gradients in the primaryphase flow field. for a transie nt bubble column). you will not specify lift coefficients for pairs consisting of two secondary phases. • Select constant to specify a consta nt lift coefficient. select the appropriate specification method from the corresponding dropdown list. droplet. perform the following steps: 1. lift coefficients are specified only for pairs consisting of a secondary phase and the primary phase. • Select userdefined to use a userdefined function for the lift coefficient (see the separate UDF Manual for details). it is not possible to enable it just for a particular phase.. since the lift forces for a particle.
10: Turbulence Models in the Viscous Model panel (Figure 23. Figure 23.12.3 ModelingTurbulence If you are using the Eulerian model to solve a turbule nt flow.4: The Viscous Model Panel for an Eulerian Multiphase Calculation .12.5.4). you will need to choose one of turbulence models described in Section 23.23.12.
model. the interphase momentum.10: Turbulence Models for details about each): • Select Mixture to use the mixture turbulence model. . you can enable them using the multiphaseoptions command in the define/models/viscous/m ultiphaseturbulence/ text menu. This model is ap. This option will enable you to modify µt in the k. see the separate UDF Manual.12: Steps in Using a Turbulence Model. Select either kepsilon or Reynolds Stress under Model. 3. a userdefined function can be used to customize the turbule nt viscosity for each phase. k. In most cases these terms can be neglected. indicate the desired multiphase turbulence model (see Section 23.5.turbulence model for each phase. Under kepsilon Multiphase Model or RSM Multiphase Model.M ultiphaseTurbulentViscosity If you are using the k. Customizingthe k. For more information. This is the default model. If you want to include any of these source terms. under UserDefined Functions. and sources are not included in the calculation. and then continue the calculation with these terms included.plicable when there is clearly one primary continuous phase and the rest are dispersed dilute secondary phases. Select the desired kepsilon Model or RSM Multiphase Model and any other related parameters. In the Viscous Model panel. • Select Per Phase to use a k. 2.The procedure is as follows: 1. If you are looking for additional accuracy.multiphase turbulence model. • Select Dispersed to use the dispersed turbulence model. Note that the inclusion of these terms can slow down convergence noticeabl y. Including Source Terms By default. as described for singlephase calculations in Section 12. select the appropriate userdefined function in the Turbulent Viscosity dropdown list. you may want to compute a solution first without these sources. This model is appropriate when the turbulence transfer among the phases plays a domina nt role.
Note the following regarding the available choices: gunn is frequently used for Eulerian multiphase simulations involving a granular phase. none allows you to ignore the effects of heat transfer between the two phases userdefined allows you to impleme nt a correlation reflecting a model of your choice.. after you have turned on the energy equation in the Energy panel. Define −→Phases.5).5: The Phase Interaction Panel for Heat Transfer 2.g. Click the Interaction.23. through a userdefined function.. Select the desired correlation for the Heat Transfer Coefficient. ure 23. button to open the Phase Interaction panel (e. .12.12..12.. you will need to visit the Phase Interaction panel. ranzmarshall is frequently used for Eulerian multiphase simulations not involving a granular phase. Fig Figure 23. 1. Click on the Heat tab in the Phase Interaction panel.. 3.4 Including HeatTransferEffects To define heat transfer in a multiphase Eulerian simulation.
and can use it in conjunction with the energy multiphase equations and available multiphase turbulence models.. for a pressure inlet or intake fan).5. you can only define one of the phases as a compressible ideal gas (i.1.8: Defining Multiphase Boundary Conditions for more information on defining conditions for a mass flow inlet in Eulerian multiphase calculations. Set the appropriate thermal boundary conditions. • You can define only one compressible fluid phase. See Section 9. and Section 23. See Chapter 7: Boundary Conditions for more information on boundary conditions. i Note that if you read a case file that was set up in a version of FLUENT previous to 6.5 Modeling Flows Compressible You can model compressible multiphase flows.4. See Section 23. and for the mixture on some boundaries.8: Eulerian Model for more information on specifying boundary conditions for a Eulerian multiphase calculation. you will need to redefine the conditions at the mass flow inlets. you will need to specify mass flow or mass flux for each individual phase.e.9: Description of Heat Transfer for more information on heat transfer in the framework of a Eulerian multiphase simulation. you can select the idealgas for the density in the Materials panel of only one phase’s material). FLUENT will use the specified value for temperature at that boundary as total temperature for the compressible phase. See Section 23.g. You will specify the thermal boundary conditions for each individual phase on most boundaries. • If you specify the total pressure at a boundary (e. When using the Eulerian multiphase model for a compressible flow. and as static temperature for the other phases (which are incompressible). . 23. note the following: • While you can specify both compressible gas phases and compressible liquid phases.6: Compressible Flows for more information about compressible flows.12. • For each mass flow inlet..9. There is no limitation on using compressible liquids using userdefined functions.9.
13 Setting Up the Wet Steam M odel Once you have enabled either of the densitybased solvers in FLUENT.15. 23. . Solution settings and strategies for the wet steam model can be found in Section 23.13.6: Wet Steam Model Theory) by opening the Multiphase panel and selecting the Wet Steam option..1 Using UserDefinedThermo dynamicWet Steam Properties FLUENT allows you to use your own property functions and data with the wet steam model. This is achieved with userdefined wet steam property functions (UDWSPF). Define −→ −→Multiphase.1: ModelSpecific Variables .1: The Multiphase Model Panel with the Wet Steam Model Activated This section includes information about using your own property functions and data with the wet steam model..23. Postprocessing variables are described in Section 23.13.14. you can activate the wet steam model (see Section 23. These userdefined functions are written in the C programming language and there is a certain programming format that must be used so that you can build a successful library that can be loaded into the FLUENT code. Models Figure 23.5: Wet Steam Model.
5. Load your newly created UDWSPF library using the text user interface. DEFINE ON DEMAND) to be able to use the compiled UDFs utili ty. 2. Turn on the wet steam model. Create a C source code file that conforms to the format defined in this section. as described in the list that follows. define −→ models −→ multiphase −→ wetsteam −→loadunloaduserdefine dwetsteamlibrary 7. Run your calculation. 4. the UDWSPF are available for use with the densi tybased solver only.e. • The code must include at least one of the UDF’s DEFINE functions (i. 23. . • Any values that are passed to the solver by the UDWSPF or returned by the solver to the UDWSPF are assumed to be in SI units.23. define −→ models −→ multiphase −→ wetsteam −→compileuserdefinedwe tsteamfunctions 6.13 Setting Up the Wet Steam M odel The following is the procedure for using the userdefined wet steam property functions (UDWSPF): 1. These functions are the mechanism by which your thermodynamic property data is transferred to the FLUENT solver. Compile your UDWSPF C functions and build a shared library file using the text user interface.13. Start FLUENT and set up your case file in the usual way.2 W riting UserDefinedWet Steam Property Functions (UDWSPF) the Creating a UDWSPF C function library is reasonably straig htforward: • The code must contain the udf. Therefore. Define the wet steam equation of state and all related thermodynamic and trans port property equations. 3. i Note that the UDWSPF can only be used when the wet steam model is activated. • You must use the principle set of userdefined wet steam property functions in your UDWSPF library.h file inclusion directive at the beginning of the source code. This allows the definitions for DEFINE macros and other FLUENT functions to be accessible during the compilation process.
M odelingM ultiphase Flows The following lists the userdefined wet steam property function names and argume nts. which takes on temperature in K and vapor density in kg/m 3 and returns specific heat at consta nt volume in J/kg/K. Function inputs from the FLUENT solver consist of one or more of the following variables: T = temperature (K ). real rho) This is the vapor specific heat at consta nt pressure. • real wetst sv(real T. real rho) This is the vapor specific entropy. real T) This is the saturated temperature function. • real wetst eosP(real rho. as well as a short description of their functions. • real wetst hv(real T. • real wetst cvv(real T. which takes on temperature in K and vapor density in kg/m 3 and returns specific heat at consta nt pressure in J/kg/K. which takes on pressure in Pa and temperature in K and returns vapor density in kg/m 3 . You use it to initialize wet steam model constants or your own model consta nts. which takes on pressure in Pa and a starting guess temperature in K and returns saturation temperature in K.real rho) This is the vapor specific enthalpy. . • real wetst cpv(real T. real T) This is the equation of state. which takes on temperature in K and vapor density in kg/m 3 and returns specific entropy in J/Kg/K. real rho) This is the vapor specific heat at consta nt volume. which takes on temperature in K and returns saturation pressure in Pa. and ρ = vaporphase density (kg/m 3 ). • real wetst satP(real T) This is the saturated pressure function. It returns nothing. • real wetst eosRHO(real P. • real wetst satT(real P. P = pressure (P a). which takes on temperature in K and vapor density in kg/m 3 and returns specific enthalpy in J/Kg. which takes on vapor density in kg/m 3 and Temperature in K and returns pressure in Pa. • void wetst init(Domain *domain) This will be called when you load the UDWSPF. real T) This is the equation of state.
• real wetst surft(real T) This is the liquid surface tension. • real wetst cpl(real T) This is the saturated liquid specific heat at consta nt pressure. • real wetst ktl(real T) This is the liquid thermal conductivi ty. which takes on temperature in K and returns liquid specific heat in J/kg/K. which takes on temperature in K and returns liquid density in kg/m 3 . which takes on temperature in K and vapor density in kg/m 3 and returns viscosity in kg/m/s. which takes on temperature in K and vapor density in kg/m 3 and returns thermal conductivity in W/m/K. real rho) This is the vapor thermal conductivi ty.23. which takes on Temperature in K and returns dynamic viscosity in kg/m/s. • real wetst ktv(real T. • real wetst rhol(real T) This is the saturated liquid density. which takes on Temperature in K and returns surface tension N/m. .13 Setting Up the Wet Steam M odel • real wetst muv(real T. which takes on temperature in K and returns thermal conductivity in W/m/K. real rho) This is the vapor dynamic viscosity. • real wetst mul(real T) This is the liquid dynamic viscosity.
liquid specific heat*/ wetst_mul. Microsoft Visual C ++.3 C om piling Your UDW SPF and Building a Shared L ibrary File This section presents the steps you will need to follow to compile your UDWSPF C code and build a shared library file. /*sat. /*vapor entropy*/ wetst_cpv. /*vapor isochoric specific heat*/ wetst_muv. /*sat. /*sat. i To use the UDWSPF you will need to first build the UDWSPF library by compiling your UDWSPF C code and then loading the library into the FLUENT code. v6. Most UNIX operating systems provide a C compiler as a standard feature. The libraries must. If you are using a PC. . The structure is of type WS Functions and its name is WetSteamFunctionLis t. /*vapor isobaric specific heat*/ wetst_cvv.. This process requires the use of a C compiler. you will need to ensure that a C ++ compiler is installed before you can proceed (e. /*Saturation temperature*/ wetst_eosP. /*vapor enthalpy*/ wetst_sv. /*vapor dynamic viscosity*/ wetst_ktv.g. /*vapor thermal conductivity*/ wetst_rhol. liquid thermal conductivity*/ wetst_surft /*liquid surface tension*/ }. liquid density*/ wetst_cpl. Internally. /*initialization function*/ wetst_satP. liquid viscosity*/ wetst_ktl.0 or higher). therefore. The UDWSPF shared library is built in the same way that the FLUENT executable itself is built.13.M odelingM ultiphase Flows At the end of the code you must define a structure of type WS Functions whose members are pointers to the principle functions listed previously. This shared library is then loaded into FLUENT (either at runtime or automatically when a case file is read) by a process called dynamic loading. as well as to the particular version of the FLUENT executable being run. UDF_EXPORT WS_Functions WetSteamFunctionList = { wetst_init. /*sat. 23. a script called Makefile is used to invoke the system C compiler to build an object code library that contains the native machine language translation of your higherlevel C source code. /*Saturation pressure*/ wetst_satT. /*equation of state*/ wetst_eosRHO. /*equation of state*/ wetst_hv. The object libraries are specific to the computer architecture being used.
– Enter the compiled UDWSPF library name. FLUENT will then start compiling the UDWSPF C code and put it in the appropriate architecture director y. when the computer’s operating system level changes. – Enter the C source file names. However. then a directory called mywetsteam will be created and it will contain the library libudf. The general procedure for compiling UDWSPF C code is as follows: • Place the UDWSPF C code in your working directory (i. or when the job is run on a different type of computer. where your case file resides). however. • Read your case file into FLUENT. If you do not have an extra header file. If. • You can now compile your UDWSPF C code and build a shared library file using the commands provided in the text command interface (TUI): – Select the define/models/multiphase/wetsteam menu item define −→ models −→ multiphase −→wetsteam – Select the compileuserdefinedwetsteamfunctions option. then press <Enter> when prompted. – Enter the header file names..e.23.13 Setting Up the Wet Steam M odel be rebuilt any time FLUENT is upgraded. you type a new library name such as mywetsteam. – Continue on with the procedure when prompted. • Launch FLUENT. i Ideally you should place all of your functions into a single file. you can split them into separate files if desired. and this directory will contain library file called libudf.. For example. The name given here is the name of the directory where the shared library (e. if applicable. libudf) will reside.g. if you hit <Enter> then a directory should exist with the name libudf. .
... define −→ models −→ multiphase −→wetsteam • Select the loadunloaduserdefinedwetsteamlibrary option and follow the procedure when prompted. perform the following steps: • Go to the define/models/multiphase/wetsteam menu item in the text user interface..M odelingM ultiphase Flows 23. If the loading of the UDWSPF library is successful. Library "libudf/lnx86/2d/libudf.4 Loadingthe UDW SPFShared LibraryFile To load the UDWSPF library.. . Initializing user defined material properties..so" opened Setting material properties to WetSteam.13. you will see a message similar to the following: Opening userdefined wet steam library "libudf".
5 UDW SPFExample This section describe a simple UDWSPF.150 .h" /*Global Constants for this model*/ real ws_TPP = 338. Drake book (1972) Use idealgas EOS with Steam properties to model wet steam condensation in low pressure nozzle Author: L.h" #include "stdarg.Reynolds tables (1979) Liquid Sat. /********************** ************************ ************************/ /* User Defined Wet Steam Properties: EOS : Ideal Gas Eq. You can use this example as a the basis for your own UDWSPF code. For approximate calculations at low pressure. Vapor Sat.0 .01 . Zori Date : Jan. real cpg = 1882. Eckert & R.h" #include "stdio. 29 2004 */ /********************** ************************ ************************/ #include "udf.C.13 Setting Up the Wet Steam M odel 23.h" #include "ctype.23.13. the simple idealgas equation of state and consta nt isobaric specific heat is assumed and used. real ws_aaa = 0. Line: E./* Cpvapor at lowpressure region*/ DEFINE_ON_DEMAND(I_do_n othing) { /* This is a dummy function to allow us to use */ /* the Compiled UDFs utility */ } . The properties at the saturated liquid line and the saturated vapor line used in this example are similar to the one used by FLUENT. Line : W.
21868400E04 if (T > ws_Tc) T = ws_Tc .M odelingM ultiphase Flows void wetst_init(Domain *domain) { /* You must initialize these material property constants. return psat.97210000E01 real a3 = 1. .50 .0)*SUM./*Critical Pressure */ mw_f = 18.52065800E03 real a8 = 5. psat = ws_Pc *exp(pratio) .09409899E03 real a6 = 4.0. . real F .1.00 .016 ./*vapor Gas Const*/ } real wetst_satP(real T) { real psat .ws_TPP) SUM = a1 + F*(a2+ F*(a3+ ./*fluid droplet molecular weight (water) */ Rgas_v = 461. . */ ws_Pc = 22089000.286 . F*(a4+ F*(a5+ F*(a6+ F*(a7+ F*a8)))))) .41924200 real a2 = 2. ..15528600E01 real a4 = 8. pratio = (ws_Tc/T . . F = ws_aaa*(T .39993000Ereal a7 = 03 2. real pratio. ./*Critical Temp.68563500Ereal a5 = 03 1. /*Pa */ } . . real SUM=0. . real a1 = 7. they will be used in the wet steam model in fluent */ ws_Tc = 647.
1*T . ++i) { if (T > ws_Tc) T = ws_Tc0. real dT. T = T + dT.5. . P = rho* Rgas_v * T .dPdT. i<25. for (i=0. int i .p1 .1 . dP = P . dTM = 0. dT = dP/dPdT .p1. real T) { real tsat .1) dPdT = (p2p1)/0.13 Setting Up the Wet Steam M odel real wetst_satT(real P. return P.dTM. /*Pa */ } . dTA. . /*K */ } real wetst_eosP(real rho.dP. dTA = fabs(dT). if (fabs(dT) < TEMP_eps*T) break. real T) { real P . if (dTA > dTM) dT=dT*dTM/dTA . p1= wetst_satP(T) p2= wetst_satP(T+0. } tsat = T.23. return tsat.p2.
cv = wetst_cpv(T. cp = cpg . return rho. real T) { real rho . h = T* wetst_cpv(T. real rho) { real cp.Rgas_v . rho = P/(Rgas_v * T) . /* (J/Kg) */ } . /* (J/Kg/K) */ } real wetst_cvv(real T.rho) . return h.real rho) { real h.rho) . /* (J/Kg/K) */ } real wetst_hv(real T. real rho) { real cv.M odelingM ultiphase Flows real wetst_eosRHO(real P. return cv. return cp. /*kg/m3 */ } real wetst_cpv(real T.
/* (Kg/m/s) */ } real wetst_ktv(real T. return ktv. real TDatum=288. return muv. /* (J/Kg/K) */ } real wetst_muv(real T.rho)*log( T/TDatum)+ Rgas_v*log(PDatum/ (Rgas_v*T*rho)). real rho) { real muv . real rho) { real s . return s.0242 . ktv=0. s=wetst_cpv(T. real rho) { real ktv .15.7894e05 .13 Setting Up the Wet Steam M odel real wetst_sv(real T. real PDatum=1.23.01325e5. muv=1. /* W/m/K */ } .
D[0] = 3. . .6 .8243852E+02 .0 real D[8] . . i/3. .M odelingM ultiphase Flows real wetst_rhol(real T) { real rhol.0) } rhol = rhoc*(1. return rhol.T/ws_Tc).96724 . if (T > ws_Tc) T = ws_Tc .6711257 . real real real real a1= 36571. a3= 2. D[7] = 5. a2= 555.ii < 8. D[4] = 2.0+SUM). D[2] = 2.0 . SUM += D[ii] * pow((1. D[6] = 1.6122557E+03 . D[3] = 8. D[1] = 2. D[5] = 2.0002765E+03 .3350520E+02 . /* (Kg/m3) */ } real wetst_cpl(real T) { real cpl. real SUM = 0.217 . a4= . for(ii=0.2265240E+02 . .8512396E+01 .++ii) { i = ii+1 .0 int ii int i real rhoc = 317.8297674E+03 .
23.13 Setting Up the Wet Steam M odel 0. real a6= 5.14336E09.00778551. . real a5= 1.00561e05.
26258E07. if (T > ws_Tc) T = ws_Tc . real a7= 7. real a4= 1. real a6= 1. return mul.M odelingM ultiphase Flows if (T > ws_Tc) T = ws_Tc .48603E05. /* (J/Kg/K) */ } real wetst_mul(real T) { real mul . mul = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*(a6+ T*a7))))) . real a5= 2.14092E17.00729561.55638E13. real a1= 1. real a3= 4. real a6= 1. real a1= 0. real a3= 1. real a2= 0. real a5= 2.17633.13969E10.31689E07. cpl = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*a6)))) .00791645. if (T > ws_Tc) T = ws_Tc . ktl = a1 + T*(a2+ T*(a3+ T*(a4+ T*(a5+ T*a6)))) .92145E13.47590E10. /* (Kg/m/s) */ } real wetst_ktl(real T) { real ktl . return cpl. /* W/m/K */ } .16604E05 . return ktl.530784. real a4= 1. real a2= 0.
/*vapor isochoric specific heat*/ wetst_muv. /*sat. Tr = T/ws_Tc . /*Saturation pressure*/ wetst_satT. sigma = 0.612 . /*sat.13 Setting Up the Wet Steam M odel real wetst_surft(real T) { real sigma . liquid specific heat*/ wetst_mul. 82. /*vapor isobaric specific heat*/ wetst_cvv. /********************** ************************ ************************/ . liquid viscosity*/ wetst_ktl. /*vapor enthalpy*/ wetst_sv.928 ./* N/m */ } Tr*(a2+ Tr*(a3+ /* do not change the order of the function list */ UDF_EXPORT WS_Functions WetSteamFunctionList = { wetst_init. /*sat. /*vapor dynamic viscosity*/ wetst_ktv. if (T > ws_Tc) T = ws_Tc . 256. /*sat. real real real real real Tr a1= a2= a3= a4= .23. /*equation of state*/ wetst_hv. /*equation of state*/ wetst_eosRHO. 75. /*vapor entropy*/ wetst_cpv.27 . liquid density*/ wetst_cpl.001*(a1 + Tr*a4))) .889 . 95. return sigma . /*initialization function*/ wetst_satP. /*vapor thermal conductivity*/ wetst_rhol. /*Saturation temperature*/ wetst_eosP. liquid thermal conductivity*/ wetst_surft /*liquid surface tension*/ }.
14.0). If the zero of the relative pressure field is in a region where the pressure variations are small. less roundoff will occur than if the variations occur in a field of large nonzero values. You can patch an initial volume fraction for each secondary phase using the Patch panel. Solution strategies for the VOF.14.4.. You can move this location by specifying a new Reference Pressure Location in the Operating Conditions panel. and 23. Thus in systems containing air and water. mixture.3.14.1 Setting InitialVolum e Fractions Once you have initialized the flow (as described in Section 25.. the reference pressure location is the center of the cell at or closest to the point (0. Define −→Operating Conditions. Initialize If the region in which you want to patch the volume fraction is defined as a separate cell zone. Setting the Reference Pressure Location The site of the reference pressure can be moved to a location that will result in less roundoff in the pressure calculation. the gas phase.0.g. See Section 25. it is importa nt that the reference pressure location be in the portion of the domain filled with air rather than that filled with water.14. if you have a gas phase and one or more liquid phases). . By default. for a steadystate simulation. 23. The position that you choose should be in a region that will always contain the least dense of the fluids (e.2: Patching Values in Selected Cells for details. you can simply patch the value there..2 VO F M odel Several recommendations for improving the accuracy and convergence of the VOF solution are presented here. respectively. for example.14.. given the same velocity distribution.2. Otherwise.M odelingM ultiphase Flows 23. this distribution will serve as the initial condition at t = 0. 23.14 Solution Strategies for MultiphaseModeling 23. and Eulerian models are provided in Sections 23. Solve −→ −→Patch. you can define the initial distribution of the phases.14: Initializing the Solution). This is because variations in the static pressure are larger in a more dense fluid than in a less dense fluid.14. For a transie nt simulation.. you can create a cell “register” that contains the appropriate cells and patch the value in the register. setting an initial distribution can provide added stability in the early stages of the calculation.
.. .14 SolutionStrategiesfor M ultiphase Modeling Pressure Interpolation Scheme For all VOF calculations. Controls Discretization Scheme Selection for the Im plicitand ExplicitFormulations When the implicit scheme is used. you should use the bodyforceweighted pressure interpolation scheme or the PRESTO! scheme. one of which is Allow selection of all applicable discretization schemes? [no] to which you will respond yes. Solve −→ −→Solution. the available options for Volume Fraction Discretization are • First Order Upwind • Second Order upwind • Modified HRIC • QUICK When the explicit scheme is used. the available options for Volume Fraction Discretization are • GeoReconstruct • CICSAM • Modified HRIC • QUICK When using the explicit scheme. as it gives a sharper interface than the modified HRIC scheme. First Order Upwind.23. and DonorAcceptor can be made available under Volume Fraction Discretization by using the following text command: solve −→ set −→expert You will be asked a series of questions. Second Order upwind. i You are encouraged to use the CICSAM scheme.
.8 for pressure is recommended for improved stability with the PISO scheme. The PISO scheme is recommended for transie nt calculations in general. an underrelaxation factor of 0. You can generally increase the underrelaxation factors for all variables to 1 and expect stability and a rapid rate of convergence (in the form of few iterations required per time step). Unde rRelaxationfor the SteadyState Formulation If you are using the steadystate implicit VOF scheme. the underrelaxation factors will need to be decreased if the solution exhibits unstable.3 M ixture odel M Setting the Unde rRelaxationFactor for the Slip Velocity You should begin the mixture calculation with a low underrelaxation factor for the slip velocity.2 or less is recommended.14.tum trans port equations may reduce the stability of the solution compared to cases using firstorder discretization. it is recommended to use a loworder varia nt of RhieChow face flux interpolation. using a highorder discretization scheme for the momen. enter yes. In such situations.M odelingM ultiphase Flows i In VOF modeling. A value of 0. If the solution shows good convergence behavior. For calculations on tetrahedral or triangular meshes.5 for improved stabili ty.elocity Couplingand UnderRelaxationfor the Timedependent V Formulations Another change that you should make to the solver settings is in the pressurevelocity coupling scheme and underrelaxation factors that you use.2 and 0. you can increase this value graduall y. Using PISO allows for increased values on all underrelaxation factors. Pressure. which can be turned on using the text command: solve −→ set −→numerics When asked to disable high order RhieChow flux?[no].7–0. divergent behavior with the underrelaxation factors set to 1. the underrelaxation factors for all variables should be set to values between 0.. Controls As with any FLUENT simulation. Solve −→ −→Solution. . without a loss of solution stability. Reducing the time step is another way to improve the stabili ty. 23.
4 EulerianM odel Calculatingan InitialSolution To improve convergence behavior. and may make it even more difficult for the flow to converge. Once you have obtained an initial solution for the primary phase. turn the volume fraction equations back on and continue the calculation for all phases. deselect Volume Fraction and Slip Velocity in the Equations list. i You should not try to use a singlephase solution obtained without the mixture or Eulerian model as a starting point for an Eulerian multiphase calculation.. 23. complete the setup. You can then enable the Eulerian model.14 SolutionStrategiesfor M ultiphase Modeling Calculatingan InitialSolution For some cases (e. deselect Volume Fraction in the Equations list in the Solution Controls panel. Controls In the Solution Controls panel. It is recommended that you set the value of the volume fraction close to the value of the volume fraction at the inlet. you may be able to obtain a solution more quickly if you compute an initial solution without solving the volume fraction and slip velocity equations. you may want to compute an initial solution before solving the complete Eulerian multiphase model.. and continue the calculation using the mixturemodel solution as a starting point. Once a converged flow field is obtained. There are three methods you can use to obtain an initial solution for an Eulerian multiphase calculation: • Set up and solve the problem using the mixture model (with slip velocities) instead of the Eulerian model.14.23. but compute the flow for only the primary phase. you can temporarily disable these equations and compute an initial solution. • Use the mass flow inlet boundary condition to initialize the flow conditions. turn the Volume Fraction and Slip Velocity equations back on again. Doing so will not improve convergence. . Solve −→ −→Solution. cyclone separation). You can then compute the initial flow field. and compute the mixture solution.. To do this. Once you have set up the mixture model. • Set up the Eulerian multiphase calculation as usual.g.
and Max Cycles. define these forces appropriatel y. the following two field variables will show up in the inflow.5 Wet Steam M odel BoundaryConditions. indicating zero droplets entering the domain. • Liquid Mass Fraction (or the wetness factor) In general. See Section 23. Once the solution without these forces starts to converge.Initialization. outflow boundary panels.22.2: Defining the Phases for the Eulerian Model for more information about granular phases. PostSweeps.M odelingM ultiphase Flows Tem porari IgnoringL iftand V irtualM ass Forces ly If you are planning to include the effects of lift and/or virtual mass forces in a steadystate Eulerian multiphase simulation.13: Setting Up the Wet Steam Model). and continue the calculation.2 and 25. and in the Solution Initialization and Patch panels. Using WCycle Multigrid For problems involving a packedbed granular phase with very small particle sizes (on the order of 10 µm).6: Wet Steam Model Theory and Section 23. . for dry steam entering flow boundaries the wetness factor is zero. 23. When you are choosing the values for these parameters.6. • Log10(Droplets Per Unit Volume) In general this value is set to zero.3 for details about multigrid cycles. you may need to use higher values for PreSweeps.14.12. Under Fixed Cycle Parameters in the Multigrid Controls panel.. you should also increase the Verbosity to 1 in order to monitor the AMG performance. to make sure that the pressure equation is solved to a desired level of convergence within the AMG solver during each global iteration. and Sections 25. you can interrupt the calculation. i. convergence can be obtained by using the Wcycle multigrid for the pressure.e. Patching and When you use the wet steam model (described in Section 23. you can often reduce stability problems that sometimes occur in the early stages of the calculation by temporarily ignoring the action of the lift and the virtual mass forces.
When the wetness factor is greater than 0.. Once you have obtained a proper initial solution. Lower the underrelaxation factor for the wet steam equation below the current set value. . Solution Strategies for the Wet Steam Model If you face convergence difficulties while solving wet steam flow. you can do so using the text user interface.. turn the Wet Steam equation back on again and compute the mixture solution. 2. define −→ models −→ multiphase −→ wetsteam −→ set −→ maxliquidmassfraction i Note that the maximum wetness factor should not be set beyond 0.. To adjust the temperature limits. Solve −→ Controls −→Solution. then try the following solver settings: 1. the maximum wetness factor has been set to a default value of 0. Once a converged flow field is obtained.1. In the Solution Controls panel. However.. Solve for an initial solution with no condensation. you can adjust this limit to whatever is permissible for your data. If you use your own steam property functions. try to initially lower the CFL value and use firstorder discretization schemes for the solution. To turn condensation on or off. When doing so. it is common that this limit will be reached.1.2 since the present model assumes a low wetness factor. the solution tends to be less stable due to the large source terms in the trans port equations.. In general. which corresponds to the fact that most nozzle and turbine flows will have a wetness factor less that 0. during the convergence process. Thus. you are preventing condensation from taking place while still computing the flow based on steam properties.14 SolutionStrategiesfor M ultiphase Modeling Solution Limitsfor the Wet Steam Model When you activate the wet steam model for the first time. go to the Solution Controls panel. deselect Wet Steam in the Equations list. The underrelaxation factor can be found in the Solution Controls panel. but eventually the wetness factor will drop below the value of 0.1.. Solve −→ −→Limits. in cases where the limit must be adjusted. go to the Solution Limits panel.23. Controls The default maximum wetness factor or liquid mass fraction (β) is set to 0. If you are still unable to obtain a converged solution. turn on the condensation. Solve −→ Controls −→Solution.1.1. a message is displayed indicating that the Minimum Static Temperature should be adjusted to 273 K since the accuracy of the builtin steam data is not guara nteed below a value of 273 K.
some additional field functions will be available for postprocessing.15. categories) represent mixture quantities. Chapters 28 and 29 explain how to generate graphics displays and reports of data. Thermal quantities will be available only for calculations that include the energy equation. See Chapter 30: Field Function Definitions for a complete list of field functions and their definitions.2: Displaying Velocity Vectors • Section 23. VO F Model For VOF calculations you can generate graphical plots or alphanumeric reports of the following additional item: • Volume fraction (in the Phases. and Velocity... The variables that are not phase specific are available (e.15.15..5: Reporting Flow Rates 23. Information about these postprocessing topics is provided in the following subsections: • Section 23.M odelingM ultiphase Flows 23.15. .1: ModelSpecific Variables • Section 23. as listed in this section... as appropriate for the general multiphase model and specific options that you are using. You can also report flow rates for individual phases for all three models.15..4: Reporting Forces on Walls • Section 23. and display velocity vectors for the individual phases in a mixture or Eulerian calculation.1 ModelSpecific Variables When you use one of the general multiphase models.g.3: Reporting Fluxes • Section 23.15 Postprocessing for MultiphaseModeling Each of the three general multiphase models provides a number of additional field functions that you can plot or report.. Most field functions that are available in single phase calculations will be available for either the mixture or each individual phase. variables in the Pressure.15. category) This item is available for each phase.
23.15 Postprocessing for M ultiphase odeling M
M ixture Model
For calculations with the mixture model, you can generate graphical plots or alpha numeric reports of the following additional items: • Diameter (in the Properties... category) This item is available only for secondary phases. • Volume fraction (in the Phases... category) This item is available for each phase. The variables that are not phase specific are available (e.g., variables in the Pressure... category) represent mixture quantities. Thermal quantities will be available only for calculations that include the energy equation.
EulerianModel
For Eulerian multiphase calculations you can generate graphical plots or alpha numeric reports of the following additional items: • Diameter (in the Properties... category) This item is available only for secondary phases. • Granular Conductivity (in the Properties... category) This item is available only for granular phases. • Granular Pressure (in the Granular Pressure... category) This item is available only for granular phases. • Granular Temperature (in the Granular Temperature... category) This item is available only for granular phases. • Volume fraction (in the Phases... category) This item is available for each phase. The availability of turbulence quantities will depend on which multiphase turbulence model you used in the calculation. Thermal quantities will be available (on a perphase basis) only for calculations that include the energy equation.
M odelingM ultiphase Flows
Multiphase Species Transport
For calculations using species trans port with either of the multiphase models, you can generate graphical plots or alphanumeric reports of the following additional items: • Mass Fraction of speciesn (in the Species... category) This item is available for each species. • Mole Fraction of speciesn (in the Species... category) This item is available for each species. • Molar Concentration of speciesn (in the Species... category) This item is available for each species. • Lam. Diff Coeff of speciesn (in the Species... category) This item is available for each species. • Eff. Diff. Coeff. of speciesn (in the Species... category) This item is available for each species. • Enthalpy of speciesn (in the Species... category) This item is available for each species. • Relative Humidity (in the Species... category). • Turbulent Rate of Reactionn (in the Reactions... category) This item is available for each species. • Rate of Reaction (in the Reactions... category). • Mass Transfer Rate n (in the Phase Interaction... category) This item is available for each mass transfer mechanism that you defined. Thermal quantities will be available only for calculations that include the energy equation.
23.15 Postprocessing for M ultiphase odeling M
Wet Steam Model
FLUENT provides a wide range of postprocessing information related to the wet steam model. The wet steam related items can be found in Wet Steam.... category of the variable selection dropdown list that appears in the postprocessing panels. • Liquid Mass Fraction • Liquid Mass Generation Rate • Log10(Droplets Per Unit Volume) • Log10(Droplets Nucleation Rate) • Steam Density (GasPhase) • Liquid Density (LiquidPhase) • Mixture Density • Saturation Ratio • Saturation Pressure • Saturation Temperature • Subcooled Vapor Temperature • Droplet Surface Tension • Droplet Critical Radius (microns) • Droplet Average Radius (microns) • Droplet Growth Rate (microns/s)
Report −→Forces. To do so. Note that you can color vectors by values of any available variable.. To display the mixture velocity ~vm (relevant for mixture model calculations only). and mixture as the Phase.. it is possible to display velocity vectors for the individual phases using the Vectors panel. or select the name of a phase to report fluxes just for that phase. and then select the desired phase in the Phase dropdown list. you will be able to specify the individual phase for which you want to compute the forces. You can also choose Relative Velocity to display the phase velocity relative to a moving reference frame. Select the name of the desired phase in the Phase dropdown list on the left side of the panel.2 Displaying Velocity Vectors For mixture and Eulerian calculations.M odelingM ultiphase Flows 23.. Report −→Fluxes. To display the velocity of a particular phase. Select mixture in the Phase dropdown list at the bottom of the panel to report fluxes for the mixture. you will be able to specify whether the report is for the mixture or for an individual phase.15. select Velocity in the Vectors Of dropdown list. select Velocity (or Relative Velocity for the mixture velocity relative to a moving reference frame). .15.3 Reporting Fluxes When you use the Flux Reports panel to compute fluxes through boundaries.4 Reporting Forces on Walls For Eulerian calculations. Display −→Vectors. make the appropriate selections in the Color by and following Phase dropdown lists.. 23.. for any phase you defined. 23. when you use the Force Reports panel to compute forces or moments on wall boundaries.15..
2330244 zone 3 (pressureoutlet): 9.15. An example is shown below. FLUENT will list each zone.7560663 zone 11 (spiralvelinlet): 0. followed by the mass flow rate through that zone for the specified phase. /report> mf (mixture water air) domain id/name [mixture] air zone 10 (spiralpressoutlet): 1.6150589 zone 8 (spiralwall): 0 zone 1 (walls): 0 zone 4 (velocityinlet): 4.15 Postprocessing for M ultiphase odeling M 23.5 Reporting Flow Rates You can obtain a report of mass flow rate for each phase (and the mixture) through each flow boundary using the report/massflow text command: report −→massflow When you specify the phase of interest (the mixture or an individual phase).9132133 net massflow: 5.23.4608185 .
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.