Hello Luciano, We quickly discussed this in last week's ET call.
Unfortunately not one present had any experience using Llaman with GRHydro and a non-trivial topology. Oleg Korobkin did some similar work on accretion disks using a Cactus based code with a central cutout (but not Llama based but using an earlier version of Cactus/Carpet based multiblock): https://arxiv.org/abs/1011.3010 You may be a able to reach out to him (at LANL). My own recollection is that hydro_excision is not used very frequently in GRHydro, to it may or may not actually work at all. Note when you say "origin" since this is the Llama grid with a hole in the center, you mean "inner cutout" (ie not r=0 but instead r=3.0)? Yours, Roland > Hi everyone, > > I started testing the Thornburg04nc coordinates (spherical patches without > Cartesian box) from Llama with (a modified version) of GRHydro. I started > with just an atmosphere around a Kerr-Schild metric evolving only hydro. > I'm seeing instabilities near the origin, so I was wondering if someone > already encountered this issue. In particular, I wanted to know if someone > has a good idea of how to treat the inner boundary. Right now, I'm putting > the inner boundary inside the horizon, above the ring singularity, and > applying hydro excision. I've tried a combination of things playing with > those parameters but without success. Has anyone tried a similar thing > evolving hydro? Just want to double check if there's something obvious in > the grid setup I'm missing. > > Below I attach the relevant part of the par file I'm using. Main difference > with the public version of GRHydro is that I'm using Siegel+2017 c2p which > works just fine with the same setup in Cartesian coordinates. > > Thanks a lot!! > > > > > > ###### > > ActiveThorns = "Time MoL" > ActiveThorns = "Coordbase CartGrid3d Boundary StaticConformal" > ActiveThorns = "SymBase ADMBase TmunuBase HydroBase InitBase ADMCoupling > ADMMacros" > ActiveThorns = "IOUtil Formaline" > ActiveThorns = "SpaceMask CoordGauge Constants LocalReduce aeilocalinterp > LoopControl" > ActiveThorns = "Carpet CarpetLib CarpetReduce CarpetRegrid2 " > ActiveThorns = "Coordinates GenericFD Interpolate2 CarpetInterp2 > CarpetInterp" > ActiveThorns = "CarpetIOASCII CarpetIOScalar CarpetIOHDF5 CarpetIOBasic" > > ################################################################################ > # Grid setup and general parameters > ################################################################################ > > # Grid > > CartGrid3D::type = "multipatch" > Coordinates::coordinate_system = "Thornburg04nc" > CartGrid3D::set_coordinate_ranges_on = "all maps" > > Coordinates::h_radial = 0.35 > Coordinates::sphere_inner_radius = 3.0 > Coordinates::sphere_outer_radius = 90.0 > Coordinates::n_angular = 25 > > #Coordinates::radial_stretch = yes > #Coordinates::stretch_rmin_1 = 10.0 > #Coordinates::stretch_rmax_1 = 90.0 > > Coordinates::stagger_patch_boundaries = "no" > Coordinates::stagger_outer_boundaries = "no" > > Interpolate::interpolator_order = 4 > Interpolate2::shift_edges = no > > Coordinates::store_jacobian = yes # I need to > Coordinates::store_inverse_jacobian = yes # Check for runtime errors > #Coordinates::store_jacobian_derivative = yes > Coordinates::store_volume_form = yes > > #GenericFD::jacobian_group = "Coordinates::jacobian" > #GenericFD::jacobian_derivative_group = "Coordinates::jacobian2" > > # General Carpet parameters: > Carpet::enable_all_storage = "no" > Carpet::use_buffer_zones = "yes" > Carpet::use_overlap_zones = "no" > Carpet::schedule_barriers = "no" > #Carpet::processor_topology = "recursive" # recursive breaks > multipatch > > Carpet::check_for_poison = "no" > Carpet::init_3_timelevels = "no" > Carpet::init_fill_timelevels = "yes" > > CarpetLib::poison_new_memory = "yes" > CarpetLib::poison_value = 114 > CarpetLib::check_bboxes = "no" > CarpetLib::interleave_communications = "yes" > CarpetLib::combine_sends = "no" # Erik says this is faster and > does not use more memory than yes > CarpetLib::combine_recompose = "no" # yes is the default, set to no > if we run out of memory during recomposing > #CarpetLib::restriction_order_space = 0 > > > CarpetInterp::tree_search = "yes" > CarpetInterp::check_tree_search = "no" > > # OVERLAP > Coordinates::additional_overlap_size = 3 > > # BOUNDARY ZONES (needs to be set to three) > Coordinates::patch_boundary_size = 3 > Coordinates::outer_boundary_size = 3 > #Carpet::prolongation_order_space = 5 > #Carpet::prolongation_order_time = 2 > Carpet::refinement_centering = "vertex" > > #==================================== > # TmunuBase & HydroBase > #==================================== > TmunuBase::stress_energy_storage = yes > TmunuBase::stress_energy_at_RHS = yes > TmunuBase::timelevels = 1 > TmunuBase::prolongation_type = none > HydroBase::timelevels = 3 > > ############################################################# > # Carpet > ############################################################# > > Carpet::domain_from_multipatch = yes > Carpet::ghost_size = 3 > # not for FD > #Carpet::granularity = 5 # (order+1)*m > #Carpet::granularity_boundary = 1 > #Carpet::poison_new_timelevels = "yes" # does not work with persistent RHS > > #--------CarpetMask-------------------------- > ActiveThorns = "CarpetMask" > CarpetMask::excluded_surface [0] = 1 > CarpetMask::excluded_surface_factor[0] = 1.0 > # > ############################################################# > # Time integration > ############################################################# > > Cactus::terminate = "time" > Cactus::cctk_final_time = 2000000000000 > #Cactus::cctk_itlast = 1 > > Time::timestep_method = "given" > Time::timestep = 0.10 > > MethodOfLines::ODE_method = "RK4" > MethodOfLines::MoL_Intermediate_Steps = 4 > MethodOfLines::MoL_Num_Scratch_Levels = 1 > MethodOfLines::MoL_NaN_Check = no > > > ################################################################################ > # Hydro evolution > ################################################################################ > > SpaceMask::use_mask = "yes" > > #==================================== > # GRHydro > #==================================== > ActiveThorns = "EOS_Omni" > ActiveThorns = "GRHydro" > > HydroBase::evolution_method = "GRHydro" > HydroBase::Bvec_evolution_method = "GRHydro" > # use c++ version of GRHydro > GRHydro::use_cxx_code = "yes" > > GRHydro::method_type = "RSA FV" > GRHydro::riemann_solver = "HLLE" > GRHydro::recon_method = "ppm" > GRHydro::recon_vars = "primitive" > GRHydro::GRHydro_stencil = 3 > GRHydro::bound = "flat" #also tried none > GRHydro::rho_abs_min = 1e-15 > GRHydro::initial_rho_abs_min = 5e-14 > GRHydro::GRHydro_atmo_tolerance = 0.01 > > > # -- settings for 3-param (tabulated) EOS -- > grhydro::grhydro_eos_type = "General" > grhydro::grhydro_eos_table = "nuc_eos" > EOS_Omni::nuceos_read_table = "yes" > EOS_Omni::nuceos_table_name = > "/home/d/dsiegel/lcombi/repository/EOS_tables/etk_bns_evolution/sfho/Hempel_SFHoEOS_rho222_temp180_ye60_version_1.1_20120817.h5" > GRHydro::GRHydro_eos_rf_prec = 1e-10 > hydrobase::Y_e_evolution_method = "GRHydro" > hydrobase::temperature_evolution_method = "GRHydro" > > GRHydro::reconstruct_temper = "yes" # more robust for > reconstruction > GRHydro::GRHydro_hot_atmo_temp = 0.012 #8.62e-06=1e5K > GRHydro::GRHydro_hot_atmo_Y_e = 0.5 > grhydro::grhydro_y_e_min = 0.012 > grhydro::grhydro_y_e_max = 0.599 > > > GRHydro::Grhydro_MaxNumConstrainedVars = 34 #33 > GRHydro::GRHydro_MaxNumEvolvedVars = 10 > > # -- settings for con2prim -- > GRHydro::GRHydro_c2pMHD_method = 1 > GRHydro::GRHydro_c2pMHD_method_backup = 5 > GRHydro::GRHydro_use_c2pMHD_method_backup = "yes" > GRHydro::GRHydro_c2p_grace_radius = 7.5 > grhydro::grhydro_c2p_warnlevel = 1 #1 > grhydro::grhydro_c2p_warn_from_reflevel = 5 #5 # 6 > > # --- EOS Omni --- > eos_omni::poly_gamma = 1.33333333 > #eos_omni::poly_gamma_initial = 1.333333333 > eos_omni::poly_k = 0.3605 #145.877704 #100 > eos_omni::gl_gamma = 1.33333333 > eos_omni::gl_k = 0.3605 #145.877704 #100 > > > > ################################################################################ > # Spacetime evolution > ################################################################################ > > ADMBase::metric_type = "physical" > ADMBase::evolution_method = "static" > ADMBase::lapse_evolution_method = "static" > ADMBase::shift_evolution_method = "static" > ADMBase::dtlapse_evolution_method= "static" > ADMBase::dtshift_evolution_method= "static" > > ADMBase::metric_timelevels = 3 > ADMBase::lapse_timelevels = 3 > ADMBase::shift_timelevels = 3 > ADMMacros::spatial_order = 4 > > #==================================== > # Apparent horizon & excision > #==================================== > ActiveThorns = "hydro_initexcision" > hydro_initexcision::hydro_initexcision = "yes" > hydro_initexcision::hydro_initexcision_type = "sphere" > hydro_initexcision::hydro_initexcision_coordinate_length = 4.2 # should be > slightly larger than a_BH = spin_BH*mass_BH > > # --- Settings for apparent horizon --- > ActiveThorns = "AHFinderDirect SetMask_SphericalSurface" > AHFinderDirect::N_horizons = 1 > AHFinderDirect::find_after_individual_time[1] = 0 > AHFinderDirect::find_every = 20000 > AHFinderDirect::output_h_every = 20000 > AHFinderDirect::initial_guess_method[1] = > "Kerr/Kerr-Schild" > AHFinderDirect::initial_guess__Kerr_KerrSchild__x_posn[1] = 0.0 > AHFinderDirect::initial_guess__Kerr_KerrSchild__y_posn[1] = 0.0 > AHFinderDirect::initial_guess__Kerr_KerrSchild__z_posn[1] = 0.0 > AHFinderDirect::initial_guess__Kerr_KerrSchild__mass[1] = 3.0 # adjust to > settings for TorusID > AHFinderDirect::initial_guess__Kerr_KerrSchild__spin[1] = 0.8 # adjust to > settings for TorusID > AHFinderDirect::geometry_interpolator_name = "Lagrange > polynomial interpolation" > AHFinderDirect::geometry_interpolator_pars = "order=4" > AHFinderDirect::surface_interpolator_name = "Lagrange > polynomial interpolation" > AHFinderDirect::surface_interpolator_pars = "order=4" > AHFinderDirect::mask_buffer_thickness = 0 > AHFinderDirect::mask_radius_multiplier = 0.8 > AHFinderDirect::mask_radius_offset = 0 > AHFinderDirect::max_allowable_horizon_radius[1] = 10 > AHFinderDirect::max_allowable_Theta_growth_iterations = 10 > AHFinderDirect::max_allowable_Theta_nonshrink_iterations = 10 > AHFinderDirect::max_Newton_iterations__initial = 50 > AHFinderDirect::max_Newton_iterations__subsequent = 50 > AHFinderDirect::which_surface_to_store_info_by_name [1] = "apparent > horizon" > > ################################################################################ > # Initial data > ################################################################################ > > InitBase::initial_data_setup_method = "init_some_levels" > ADMBase::initial_data = "exact" #TorusID" > ADMBase::initial_lapse = "exact" #TorusID" > ADMBase::initial_shift = "exact" #TorusID" > ADMBase::initial_dtlapse = "zero" > ADMBase::initial_dtshift = "zero" > > ActiveThorns = "Exact" > Exact::exact_model = "Kerr/Kerr-Schild" > Exact::Kerr_KerrSchild__mass = 3.0 > Exact::Kerr_KerrSchild__parabolic = "yes" > > > > > --- > *Dr. Luciano Combi* > Postdoctoral Researcher > Perimeter Institute for Theoretical Physics > CITA National Fellow (U. of Guelph) > --- -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://pgp.mit.edu .
pgpV77m141fRe.pgp
Description: OpenPGP digital signature
_______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users