[IrieConqueror]

Hey guys, I have a tank within 2 liquid fluids with different densities. Inside the tank I put on forced convection (inlet-flow: 50:50 of the 2 fluids), so that they get mixed. After converged solution the density is complete homogeneous (but there should be at least a little deviation at different locations?!).
That made me a little worried, so I'm thinking, that I used the wrong model. To be sure I stoped the forced convection, with the expactation, that the two fluids will get divided again because of the different densities. They didn't.
What kind of setup would you use? I read the whole documentation about multipahse flow but i'm still not sure which model would be the best.
My first idea was to setup a singlefluid flow for the forced convection with a mixture of the two fluids.I did an initialization to fill the tank with 50:50 of the fluids.
My Second idea was to setup multiphase flow with 2 continuous fluids with a fluid buoyancy model (density difference) and, according to the documentation, the mixture model. But the mixture model seems to be "hard to handle". I don't know what value to use for interfacial length scale and why is a drag coefficient required?
Hopefully you can help me out, thanks in advance.

[Gert-Jan]

You should first know it the liquids are miscible (e.g. water and glycerine) or inmiscible (e.g. water and oil)

[ghorrocks]

..... and also the physics you expect to be important. Do you get one phase separating out from another? If so, how? Any chemistry? Are they mixed at the molecular level, or droplets, bubbles, free surface? What fluid movements take place?

[IrieConqueror]

Thanks for the replies.

@Gert-Jan: The liquids are two different motor fuels. I'm not a chemist but i would say that these fluids are partly miscible. For the sake of simplicity i would like to produce both kind of results (miscible and inmiscible). I don't know how to setup partly miscible for this fluids in cfx.

@ghorrocks: "Do you get one phase separating out from another?" If i turn off forced convection then yes, because of the differeces of the densities. Other chemistry issues are not of interest for me. The mixing at molecular level like kinematic diffusion or diffusion because of turbulence are negligible because the mixing is strongly forced convection driven, no droplets or bubbles. By free surface you mean the interaction at the interface between two fluids? I would neglect that too, because of the strongly forced convection? (not sure). Forced convection takes place in sort of tangential flow in a cylindrical tank.

Thanks again

[ghorrocks]

By "partly miscible" do you mean that they do mix, but the rate of mixing is slow, meaning that diffusion is slow in the mixture? If this is the case then it sounds like a multicomponent mixture is what you want, not a multiphase model. Multicomponent mixtures can handle fluids of different properties which interact through diffusion. But note its ability to separate light and heavy fluids by buoyancy is limited - I think you will find if the mass fractions of different parts of the fluid are different then CFX can model the buoyancy effect this causes, but if the mass fractions are intimately mixed then CFX will not separate them out by buoyancy.

[IrieConqueror]

Thanks, yeah exactly. thats what I tried to say. sorry for my bad formulation. I did the multicomponent simulation already. maybe you can take a look at my .ccl

MATERIAL: HeavyOil
Material Group = Constant Property Liquids
Option = Pure Substance
Thermodynamic State = Liquid
PROPERTIES:
Option = General Material
EQUATION OF STATE:
Density = 1000 [kg m^-3]
Molar Mass = 18 [kg kmol^-1]
Option = Value
END
SPECIFIC HEAT CAPACITY:
Option = Value
Specific Heat Capacity = 140 [J kg^-1 K^-1]
Specific Heat Type = Constant Pressure
END
REFERENCE STATE:
Option = Specified Point
Reference Pressure = 1 [atm]
Reference Specific Enthalpy = 0 [J/kg]
Reference Specific Entropy = 0 [J/kg/K]
Reference Temperature = 25 [C]
END
DYNAMIC VISCOSITY:
Dynamic Viscosity = 0.04 [Pa s]
Option = Value
END
THERMAL CONDUCTIVITY:
Option = Value
Thermal Conductivity = 7.96 [W m^-1 K^-1]
END
ABSORPTION COEFFICIENT:
Absorption Coefficient = 1.0 [m^-1]
Option = Value
END
SCATTERING COEFFICIENT:
Option = Value
Scattering Coefficient = 0.0 [m^-1]
END
REFRACTIVE INDEX:
Option = Value
Refractive Index = 1.0 [m m^-1]
END
THERMAL EXPANSIVITY:
Option = Value
Thermal Expansivity = 1.82E-04 [K^-1]
END
END
END
MATERIAL: LightOil
Material Group = Constant Property Liquids
Option = Pure Substance
Thermodynamic State = Liquid
PROPERTIES:
Option = General Material
EQUATION OF STATE:
Density = 700 [kg m^-3]
Molar Mass = 18.02 [kg kmol^-1]
Option = Value
END
SPECIFIC HEAT CAPACITY:
Option = Value
Specific Heat Capacity = 140 [J kg^-1 K^-1]
Specific Heat Type = Constant Pressure
END
REFERENCE STATE:
Option = Specified Point
Reference Pressure = 1 [atm]
Reference Specific Enthalpy = 0.0 [J/kg]
Reference Specific Entropy = 0.0 [J/kg/K]
Reference Temperature = 25 [C]
END
DYNAMIC VISCOSITY:
Dynamic Viscosity = 0.04 [Pa s]
Option = Value
END
THERMAL CONDUCTIVITY:
Option = Value
Thermal Conductivity = 0.6069 [W m^-1 K^-1]
END
ABSORPTION COEFFICIENT:
Absorption Coefficient = 1.0 [m^-1]
Option = Value
END
SCATTERING COEFFICIENT:
Option = Value
Scattering Coefficient = 0.0 [m^-1]
END
REFRACTIVE INDEX:
Option = Value
Refractive Index = 1.0 [m m^-1]
END
THERMAL EXPANSIVITY:
Option = Value
Thermal Expansivity = 2.57E-04 [K^-1]
END
END
END
MATERIAL: Mixture
Material Group = User
Materials List = HeavyOil,LightOil
Option = Variable Composition Mixture
END

ANALYSIS TYPE:
Option = Steady State
EXTERNAL SOLVER COUPLING:
Option = None
END
END
DOMAIN: Fluid
Coord Frame = Coord 0
Domain Type = Fluid
Location = FLUID
BOUNDARY: Out
Boundary Type = OUTLET
Location = AUSTRITT
BOUNDARY CONDITIONS:
FLOW REGIME:
Option = Subsonic
END
MASS AND MOMENTUM:
Mass Flow Rate = 46.18 [kg s^-1]
Mass Flow Rate Area = As Specified
Option = Mass Flow Rate
END
END
END
BOUNDARY: In
Boundary Type = OPENING
Location = In_1,In_2
BOUNDARY CONDITIONS:
COMPONENT: HeavyOil
Mass Fraction = 0.5
Option = Mass Fraction
END
FLOW DIRECTION:
Option = Normal to Boundary Condition
END
FLOW REGIME:
Option = Subsonic
END
MASS AND MOMENTUM:
Option = Opening Pressure and Direction
Relative Pressure = 8 [bar]
END
TURBULENCE:
Option = Medium Intensity and Eddy Viscosity Ratio
END
END
END
DOMAIN MODELS:
BUOYANCY MODEL:
Buoyancy Reference Density = 0 [kg m^-3]
Gravity X Component = 0 [m s^-2]
Gravity Y Component = 0 [m s^-2]
Gravity Z Component = -9.81 [m s^-2]
Option = Buoyant
BUOYANCY REFERENCE LOCATION:
Option = Automatic
END
END
DOMAIN MOTION:
Option = Stationary
END
MESH DEFORMATION:
Option = None
END
REFERENCE PRESSURE:
Reference Pressure = 0 [atm]
END
END
FLUID DEFINITION: Mix
Material = Mixture
Option = Material Library
MORPHOLOGY:
Option = Continuous Fluid
END
END
FLUID MODELS:
COMBUSTION MODEL:
Option = None
END
COMPONENT: HeavyOil
Kinematic Diffusivity = 1e-6 [m^2 s^-1]
Option = Transport Equation
TURBULENT FLUX CLOSURE:
Option = Eddy Diffusivity
Turbulent Schmidt Number = 0.9
END
END
COMPONENT: LightOil
Option = Constraint
END
HEAT TRANSFER MODEL:
Fluid Temperature = 20 [C]
Option = Isothermal
END
THERMAL RADIATION MODEL:
Option = None
END
TURBULENCE MODEL:
Option = SST
BUOYANCY TURBULENCE:
Option = None
END
END
TURBULENT WALL FUNCTIONS:
Option = Automatic
END
END
END

Also it would be great if you could explain your last sentence a little bit more. What does limited mean? In my case, the mass fraction of the fluids are 50% each. because of the forced convection they get intimately mixed i guess. does it mean that the mixing process is calculated right but cfx can't separate them out by buoyancy.

[Gert-Jan]

Not matter what.... I would not use this:

Buoyancy Reference Density = 0 [kg m^-3]

I would set it equal to 700 kg/m3

[ghorrocks]

My comment about buoyancy is that CFX will correctly model buoyant effect where different control volumes have different densities. So density variations which CFX resolves with the mesh will have buoyant effect correctly applied. But fluids which are intimately mixed, so a single control volume contains both heavy and light fluids intimately mixed and they get separated due to buoyancy - CFX cannot model this.

[IrieConqueror]

thanks to both of you.

@ghorrocks: I think I missunderstood you.

If I use the multicomponent model, the fluids are intimately mixed. CFX just calculates one velocity field because I setup only one fluid (the mixture) in the domain. So the multicomponent model, in terms of mixing, is just doing diffusion? This means to me that the differences of the densities and the buoyancy model just supporting diffusion (micro scale) and causing no convection (bigger scale). Am I right? In that case I have to conisder to choose multiphase.

Thanks again!

[ghorrocks]

Am I right? In that case I have to conisder to choose multiphase.

Nope, you are wrong. And you have misunderstood when to choose a multiphase model.

A multicomponent mixture does have one velocity field, but that velocity field can model convection, buoyancy, heat transfer, compressible effects etc etc, in addition to the diffusion of the multicomponent mixture.

You ONLY choose a multiphase model if you have multiple phases. Your fluid is a pair of liquids - they are the same phase, so it is not a multiphase model. It is not actually that simple - multiphase models can also be used for liquids which have distinct boundaries, like oil and water.

You should be aware of the definition of a multiphase flow from the CFX documentation (Modelling Guide): "Multiphase flow refers to the situation where more than one fluid is present. Each fluid may possess its own flow field, or all fluids may share a common flow field. Unlike multicomponent flow, the fluids are not mixed on a microscopic scale in multiphase flow. Rather, they are mixed on a macroscopic scale, with a discernible interface between the fluids. ANSYS CFX includes a variety of multiphase models to enable the simulation of multiple fluid streams, bubbles, droplets, solid particles, and free surface flows."

In your case the mixing is at the microscopic scale, so this definition clearly states you should be using the multicomponent model.