[helen]

Hello everyone,

i am trying to simulate the mixing of two solutions in a Cross junction using ANSYS CFX, every solution has its properties. what i want to ask is how to take dispersion in consideration in ANSYS CFX.
if i define one of this solution as a dispersed fluid i have to define the mean diameter, what would be the mean diameter for a liquid that have almost the same density as water?
i will be very thankful for a reply.

[ghorrocks]

Before you jump into questions about multiphase flow - is this flow actually a multiphase flow? A multiphase flow model is not much use if it is not a multiphase flow.

What are your solutions? Is it something dissolved in a liquid (for instance salt in water)? Or a dispersion of solids (a slurry)? Or maybe an emulsion? Or air bubbles?

What happens when the solutions mix? Do they diffuse into each other? Or chemically react? Or some other mixing?

[helen]

first of all thank you for your reply.
well i am not sure if it could be considered as a multi phase flow, i have already done a simulation using Ansys CFX by definig both liquids as continuis fluid, and i am not sure if it was right.
the two solutions are water and salted water they enter the cross junction from two adjacent pipes and i want to know how the concentration at both outlets distributed.

[ghorrocks]

Salt in water normally dissolves. This means there is only one phase - so it is not a multiphase flow. If this is the case then forget about a multiphase model.

A dissolved substance in a liquid can be modelled two ways:
* Additional variable - this is very simple, easy and does not add much to the complexity of the model. It can model convection and diffusion, but cannot model the effect of the dissolved substance on the physical properties of the liquid. If the physical properties of the liquid don't change significantly then this approach is the one to use.

* Multicomponent mixture - this is a little more complex. It can do everything the additional variable approach can do but it can also model the change of physical properties on the liquid. It can also model chemistry if that is important.

[helen]

do you mean that i have to define the salted water as a multicomponent mixture (Water and Salt) and define its properties in CFX Pre?
my problem is that in the experiments the salted water was mixed with methanol in order to get a density as the density of water.
if i defined an addition varaible that have the same properties as this solution (Methanol, salt and water) , what should i define in ANSYS CFX to take difusion and convection in consideration?
i am realy thankful for your help

[ghorrocks]

So what I am saying is:

* It is not a multiphase flow, so don't use a multiphase model.

* You have two main options available for modelling this flow:

** Additional variable - simple but does not model changes in physical properties
** Multicomponent mixture - more complex but can handle changes in physical properties.

If you are new to this area I would start with the additional variable approach. Add an additional variable in CFX-Pre and define it to have convection and diffusion. You will need to set the diffusion constant.

[helen]

Hello Glenn,

in the last time i tried to read more about additional variable, because before i had no idea what does it mean.

i defined an additional variable (Variable type: specific; Tensor type:scalar), since the salted water comes from inlet 1 and clean water comes from inlet 2 (inlet 1 and inlet 2 are adjacent). i defined the value of the additional variable to be 1 for inlet1 and 0 for inlet 2.
i used the transport equation to solve the additional variable, and defined the kinematic diffusivity.
i am not sure if i need to define something more to get good results, because by using these definitions i couldn't get results that match the results of the labor experiemnts.
i really apprciate your help and i will be very thankful if you can help me with this problem

[ghorrocks]

Your flow appears to be affected by the density difference between the water and salty water. Have you linked the fluid density into the salt density? If you have not done this then you have missed what appears to be a fundamental driving force in the flow.

[helen]

the salty water in the experiemnts as i told you before is already mixed with methanol to get a density of almost 1000 kg/m³. and by the simulation i used the buoyancy model.
should i define the density for the additional variable, and if so how can i do that?

thank you in advance

[ghorrocks]

When water is mixed with water/salt/methanol, what happens? If the salt goes one way and the methanol goes another way you will have a density difference and a flow will be generated. If the salt and methanol stay together then there is no density difference and no buoyancy driven flow is generated.

Do you know which option occurs?

[helen]

in the experiemnts, the methanol and salt stay together and they flow after some time at the buttom of the pipe because the solution density is alittle bit higher then the density of water.

[ghorrocks]

I see, so it sounds like the density of the water/methanol/salt is significant. In that case you have to make the density a function of the additional variable so the density variation is generated.

[helen]

m sorry! but how to do that by using additional variable?
where can i define the density?

[ghorrocks]

You define the density in the material properties.

Have a look at the CFX reference manual for the naming syntax for variables.