Add option to disable double-flux approach. #78
Conversation
|
I'm getting some pretty strange results on the example cases, so this might not be ready to merge in. While part of the advantage of the double-flux approach is improved shock-capturing, I would expect the combination of the fifth-order WENO scheme and the HLLC Riemann solver to handle discontinuities like these even without it. 
I may need to set up some Sod shock tube test cases to investigate this further. |
TimothyEDawson
force-pushed
the
double-flux-toggle
branch
from
c69933c to
37d6902
Compare
|
Alright, realized I made a tiny mistake where the fluxes ended up using energy where they should have been using enthalpy! Now everything checks out. 
Additionally, I added a Sod shock tube case to the examples which is set up for comparing different schemes! 
This is ready to merge in any time. |
TimothyEDawson
force-pushed
the
double-flux-toggle
branch
from
37d6902 to
3808e8e
Compare
Initializing a `Combustor` with `use_double_flux=False` will now skip the double-flux variable calculations. Temperature will be computed from the internal energy instead of having pressure computed from the double-flux variables, and a single inviscid flux is computed at the cell faces. Added a Sod shock tube example script for testing scheme options.
TimothyEDawson
force-pushed
the
double-flux-toggle
branch
from
3808e8e to
7b19880
Compare
jjhans
left a comment
jjhans
left a comment
There was a problem hiding this comment.
Looks good! I'll try running the validation case again with these new changes
2 participants
Initializing a
Combustorwithuse_double_flux=Falsewill now skip the double-flux variable calculations. Temperature will be computed from the internal energy instead of having pressure computed from the double-flux variables, and a single inviscid flux is computed at the cell faces.