Learning Forum

[chuminhkhue]

Hi, everybody.
I'm doing in ansys fluent 2023R2.
-  Steady-state for 2D model.

- Multiphase model: Eulerian.

- Primary phase material: Newtonian.

- Second phase material: non-Newtonian.

-  Scheme: phase coupled simple
-  Mesh size: 5e-3mm.
-  Diameter for secondary phase: 8e-3mm.
-  Momentum is second-order upwind.
Boundary condition: rectagular (0.1 mm and 20mm ), velocity inlet is 0.5m/s. zero presure at outlet

My model cannot converge when using Schiller-Naumann model for Drag Coefficient and saffman-mei for Lift Coefficient. If Lift Coefficient and Drag Coefficient are constant, then it can converge with the default criteria.


I am looking forward help from experts and knowledgeable users
Thank you so much!!!

[Rob]

Have a look at the flow field - what's going on? Multiphase models rarely run well in steady state as there are nearly always some transient features in the flow. How are the monitors looking?

[chuminhkhue]

Because of some proposals, I am trying to represent it using a steady model.
Monitor looks like this.

[Rob]

Point monitors from the report definitions - ie how the flow is behaving in the domain. 

[chuminhkhue]

Sorry for my misunderstand your mean.
Monitors from the report definitions like this.
1. the pressure for the entire domain.

2.drag force and lift force on wall.


3. viscosity in secondary phase.

[Rob]

No worries. You may find you want single point values in various locations to get a better idea of what is going on. 

Of the above, the Lift plot looks most diagnostic as the scale is reasonable (you can alter the scale and number type in the plots). That suggests the solution isn't stable, and that something is changing with each iteration. You now need to review more data points, mass reports and a mix of contour & vector plots at intervals to decide if the result is good enough to answer the question the model was built to solve. 

[chuminhkhue]

 

1. scaled residuals 


2. I observe various points along the centerline
a) at a third

b) at the center

c) at two-thirds

d) static pressure contour

I have observed. but I can’t achieve convergence be achieved. could you give me some advice for my case?
thanks and best regards,

 

[Rob]

Change the scale on the xy plots - there may well be some noise that's being masked by the nonsense values from the first 50-100 iterations. That may well explain what's going on. 

[chuminhkhue]

 

 

Thank you for your support. Here is the zoomed graph

a) at a third

b) at the center

c) at two-thirds

2.drag force and lift force on wall.

 

[Rob]

So the solution is still changing, and may have some transient features. Run on and see what happens. Also save some contours etc to see if the flow field changes.