Open Hub
Open Hub

 Settings |  Report Duplicate

2
I Use This!
Inactive

Commits : Listings

Analyzed 1 day ago. based on code collected 1 day ago.
Sep 05, 2024 — Sep 05, 2025
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
using convention that the sign of the beta coefficients for the explicit laplacian calculation are defined when the coefficients are created; not when they are harmonically averaged to faces. will use this convention in thermal_conduct.f90 as well.
ajnonaka
More... almost 18 years ago
setting temp_new = temp_old within the react_base function rather than in advance
ajnonaka
More... almost 18 years ago
adv_bc_level_array and ell_bc_level_array need to be defined with one extra component due to my previous checkin
ajnonaka
More... almost 18 years ago
Change the algorithm so that we don't in fact update base state temperature in react_base, instead the "out" value is equal to the "in" value.
almgren
More... almost 18 years ago
for the adv bc for pressure, set to EXT_DIR instead of FOEXTRAP (the original functionality before my previous commit). Introduced a new type of bc, associated with neumann_comp, which always does FOEXTRAP for adv bcs and BC_NEU for elliptic bcs
ajnonaka
More... almost 18 years ago
for the adv bc for pressure, set to FOEXTRAP instead of EXT_DIR. This allows me to use the setbc function on multifabs of p0 for the thermal conduction
ajnonaka
More... almost 18 years ago
starting to add p0 stuff in; fixed some incorrect comments
ajnonaka
More... almost 18 years ago
Lots of changes. I leverage upon the put_beta_on_faces function in thermal_conduct.f90, which allows me to remove all the setup_""_op functions in make_explicit_thermal, which streamlines the code greatly. These changes will also make adding the p_0 term very simple in the future. The only functional changes are last-bit changes due to order of operations differences for adding and multiplying numbers.
ajnonaka
More... almost 18 years ago
in the put_beta_on_faces functions, I didn't like the inconsistent lower array location pass-in references. I fixed them to all be lo(1)-1:,lo(2)-1:, etc. instead of just -1:,-1:, and adjusted the function accordingly. No actual change in code functionality, just cleaned up the multi-grid case to make it more clear.
ajnonaka
More... almost 18 years ago
in the setup_""_op functions, I didn't like the inconsistent lower array location pass-in references. I fixed them to all be lo(1)-1:,lo(2)-1:, etc. instead of just -1:,-1:, and adjusted the function accordingly. No actual change in code functionality, just cleaned up the multi-grid case to make it more clear.
ajnonaka
More... almost 18 years ago
the amrvis plots now print "tfromh" or "tfromH" depending on use_big_h
ajnonaka
More... almost 18 years ago
cleaned up make_S function. The velocity isn't needed as an input.
ajnonaka
More... almost 18 years ago
Hopefully this is a fix for how we handle hgrhs when doing pressure iterations. What was there before was just wrong, and only didn't blow up before because we were running with init_iter = 1.
almgren
More... almost 18 years ago
Slight re-ordering of terms for my convenience.
almgren
More... almost 18 years ago
A few more setvals just to be careful...
almgren
More... almost 18 years ago
We were double counting the reaction terms in make_S because they were still in "thermal" from a previous sub-step plus being added explicitly in make_S.
almgren
More... almost 18 years ago
The detonations in test2 don't occur if you comment out add_react_to_thermal in the second half of the algorithm, so I commented this out until we figure out how to do it right.
ajnonaka
More... almost 18 years ago
Use a call to cell_to_edge every time in the algorithm we want a 1-d cell-centered quantity evaluated on the edges -- this should make sure the edge states are always consistent.
almgren
More... almost 18 years ago
Don't need dependence on fill_3d_module.
almgren
More... almost 18 years ago
Spacing...
almgren
More... almost 18 years ago
Remove put_1d_beta_on_edges routine.
almgren
More... almost 18 years ago
Add cell_to_edge generic function which does the averaging onto edges -- want this to be consistent everywhere. Note that there is now min/max limiting when using the higher-order formula.
almgren
More... almost 18 years ago
if use_thermal_diffusion = F, we compute the temperature from rho, H and stores it in s2. If use_thermal_diffusion = T, temperature is computed inside thermal_conduct
ajnonaka
More... almost 18 years ago
Added the omegadot terms to the temperature sources in the second half of the algorithm. Also, all newly created multifabs are initialized to zero now.
ajnonaka
More... almost 18 years ago
an additional dS/dt timestep constraint, originally used in nova, has been implemented. So far it works wonderfully in controlling the stability of the 1D flame problem.
ajnonaka
More... almost 18 years ago
simplified the interfaces to the estdt and firstdt subroutines and greatly simplified the calling sequence of these functions as it relates to the initial projection, divu iterations, init iterations, and main step. In particular, I eliminated all redudant calls to firstdt and estdt and provided additional print statements.
ajnonaka
More... almost 18 years ago
dSdt = dSource/dt is now stored and written to checkpoints instead of Source_nm1. John suggested doing it this way looking ahead to AMR. dSdt is now computed after the final call to make_S in advance, and dSdt is passed into extrap_to_halftime
ajnonaka
More... almost 18 years ago
output formatting; also the print statements in checkpoint actually correspond to what's happening now
ajnonaka
More... almost 18 years ago
need rho_omegadot2 in the checkpoint file now due to the way it is used to comptue the temperature source terms in advance.f90
ajnonaka
More... almost 18 years ago
using little h formulation now inputs file
ajnonaka
More... almost 18 years ago