[Tyson97]

Hi all,

I'm running a simulation on a model based off of medical images. When I run the Carreu model for the fluid's viscosity, it appears that the majority of the flow escapes from the mesh as seen in the attached JPEG.

This same mesh is able to handle and deliver reasonably accurate results for a Newtonian flow simulation where the flow stays within the mesh boundary. What do you think could be the issue?

Apologies, I would have posted this in the Meshing forum but the problem has only been noticed when I changed the viscosity inputs on Fluent.

[vinerm]

The flow is not outside the mesh; just the size of vectors is too large, which is a display for clarity. For non-Newtonian fluids, it is very important that turbulence model is appropriate. If the flow is laminar, then, start with 0 velocities everywhere and use first-order discretization for first 200 iterations and then change to second or higher order. Final solution must be with higher order scheme, at least second order, for non-Newtonian fluids.