Skip to content

icl-rocketry/injectorCFDModelling

Repository files navigation

injectorCFDModelling

WIP CFD model of injectors using openFOAM 2006
Currently using interPhaseChangeFoam, from modified cavitating bullet tutorial This uses incompressible fluid model, as seen in CIRA paper Instructions for each case are in the foam.foam file of that folder

Literature:
in drive

Github did not process Eulerian case, it is also on drive

notes

mesh

  • roughly 98% hexahedral, 2% prisms

  • generated by:

    • creating a wedge using blockmesh
    • discarding the plate modelled in cad using SHM from the wedge
    • collapsing the faces on the wedge knife edge using collapseEdges
    • manually changing the patch types for side1 & side2 to cyclical
  • so far I've used a cyclical BC, but it's possible wedge would be better

  • generated by:

    • creating a wedge using blockmesh
    • dimensions were taken from CIRA paper for Waxman case to compare results.
  • mesh was updated to use wedge BC

  • when finding the cells to discard, I've been using an axisymmetric wedge made in Fusion360 of the correct size etc, but i believe it would work just as well using an stl of the whole injector plate

  • cell refinement near the injector is created using grading in the x direction

  • may have to use multiple blocks & merge them to create mesh similar to UoT, with defined refinement around each injector hole. Doing this by hand sounds mind numbing if running a decent no. of cases, a matlab or python script could automate it somewhat

turbulence

  • k-omega SST turbulence was used as boundary layer was well meshed and it proviedes the most reliable results available with RANS approach.

cavitation

  • currently SchnerrSauer coeffs are exactly as seen in the cavitating bullet tutorial
  • can calculate these using Rayleigh–Plesset equation (see paper in literature for more details
  • n_0 has to be found by testing convergence of simulation. It affects mostly where does cavitation start.

current models

  • k-omega SST model was used to model turbulence
  • cavitation does not match with predictions in CIRA paper, probably further refinement of mesh near intlet to injector have to be done
  • to extract mass flow rate velocity field has to be integrated over section of mesh in Paraview

About

WIP CFD model of injectors using openFOAM

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages