Learning Forum

[dray19]

Hello,

I am looking at the fluid dynamics in a square microchannel for 0.1<=Re<=1000. This involves determining the hydrodynamic entrance length and the friction factor for both the developing and developed regions.

I have conducted a mesh independent study for several Re>100 and validated the results with previous researchers experimental and analytical work

When I did the mesh independent study (7 meshes) for Re=0.1,1,10, inlet pressure increases by 2% (~0.01, 0.1, 1 Pa respectively) with each mesh refinement. The other parameters (velocity, entrance length) seem to constant expect entrance length changed in last few meshes. I increased the number of elements in each direction by 25% to give approximately 2x the number of elements. 

Should I keep increasing the mesh? Am I missing something else? I have read (forums) that fluent should be able to handle creep flow but maybe it can't.

Any help is greatly appreciated.

Attached are the results for Re=0.1 mesh study

Model Information:

Microchannel: 100um x 100um x 210um 

Model Domain: 50um x 50um x 210um (symmetry applied)

CFD: Ansys Fluent 19.1

Double Precision

Solver - Pressure-Based, Steady State

Model – Laminar

Residual – 10E-6: Continuity, Velocity in X,Y,Z Direction

Methods:

Scheme: SIMPLE

Spatial Discretization

Gradient: Least Squares Cell Based

Pressure: Second Order

Momentum: Second Order Upwind

 

[Raef.Kobeissi]

hi,

how did the pressure inlet increase by 2% ? This is a boundary condition and should remain constant regardless of the mesh, can you send a picture that shows this increase.

Regards

[Karthik R]

Hello,

Could you please elaborate your meshing strategy near the wall? How is your boundary layer mesh? How does it change with each successive refinement?

Thank you.

Best Regards,

Karthik 

[dray19]

Raef,

I am using a Velocity-Inlet and Pressure-Outlet boundary conditions. If I apply a Pressure-Inlet B.C, the velocity profile is fully developed at the inlet and I can't determine entrance length or developing friction factor.

[dray19]

Kremella,

I set the number of elements for each direction (ex: 15 in x, 15 in y, 65 in z-axial direction) and apply a bias factor to have smaller elements at the walls and in the developing flow regime. For refining, I increased the number of elements in each direction by 25%

 

 

[Karthik R]

Hello,

Please take a look at this discussion on the student community. This might help you understand mesh independence better.

https://forum.ansys.com/forums/topic/convergence-of-results-in-the-study-of-mesh-independence/

Thank you.

Best Regards,

Karthik