Discussion Forum

Hello,

I am trying to model the settling of solids in a settling pond with a mixture model.

The original model was 3D and the pond is a large structure and when we got the error we were unable to pin down exactly where the issue so we decided to down size and try a simple 1mx1m tank in 2D. I am still getting the same error.

The error reads singular matrix, "xyz" void equations...(the coordinates near the inlet and outlets).

It looks like a simple straight forward model... not sure what is causing teh problem.

Any help would be greatly appreciated!

Thanks,
Preethi

ps: unable to attach the model (file extension error!)... so ill try and provide as many details as possible.

The original 3D pond model:
parameters defined in the global definitions sections:
u_in 0.725m/s
phid_in 0.015

Geometry is a hexahedron with
upper surface LXB= 130X40
lower surface LXB= 90X2.4
Depth =8m

two inlet pipes 0.6m in dia, 4m in length at approx 0.6m depth from top and one outlet pipe (same dimensions as inlet pipes) at same hieght

mixture properties:
continuous phase: rho-1000kg/m3, viscocity 0.001pa.s
solid dispersed phase- rho 1600kg/m3, dia 6e-2

Boundary conditions: wall function at all boundaries except inlet/outlet

Inlets: Vel in x direction- u_in, specific turbulent length and intensity (default values), dispersed phase concentration - phid_in
Outlet: pressure no viscous stress p0=0, No dispersed phase flux (although tried a couple of other options like dispersed phase outlet & dispersed phase conce. with no useful result- leads to same resuls)

Gravity : -9.82 (also tried -gconst) in z direction

Initial conditions: all zeros. (although tried adding different numeral values values ( and phid_init and u_init and specifying in parameters) for u_in and phid_in )

Mesh: Two Tethed meshes, finer one for the corners and the inlet and outlet areas and slightly coarser one for the general body of pond

Solver: Stationary solver (default settings) : seggregated solver