Phantom v1.3.0 released

[Daniel Price]

Dear Phantom users,

I am happy to announce a new stable release of phantom, v1.3.0. Full release notes attached below, and here:

https://bitbucket.org/danielprice/phantom/wiki/releasenotes

This is a minor update, but it’s been a while so there’s quite a bit in it.

The stable release has also been pulled to the stable branch (see https://bitbucket.org/danielprice/phantom/wiki/stable).

Best wishes,

Daniel

Monash Centre for Astrophysics (MoCA)
School of Physics and Astronomy
10 College Walk, Clayton Campus,
Monash University, VIC 3800
Australia
http://users.monash.edu.au/~dprice/

———————————————
Release notes
v1.3.0 - 22 Feb 2019 - 4d45cb3
———————————————
Physics:
• Multigrain dust simulations with multiple large grains now possible (Mentiplay). This complements the multigrain method used for small grains, but simulating small and large grain populations simultaneously is not yet fully functional
• Further updates to dust growth algorithms (Vericel)
• Much improved wind injection routines (Price, Siess)
• Improvements to Roche lobe injection module (Worpel)
• Injection modules can now provide an additional timestep constraint where needed
• One fluid dust uses method of Ballabio et al. (2018) to prevent negative dust fractions
• can now set a maximum density after which the simulation will end, also dtmax will dynamically decrease/increase if density increases too rapidly (Wurster)
• removed obsolete and unused etamhd fixed resistivity variable (Wurster)
• reduced timestep from physical viscosity force by factor 0.4: this has been found to lead to much better convergence of disc simulations that use this method (Nixon)

Bugs:
• bug fix with momentum conservation in two fluid dust-gas drag when ISOTHERMAL=yes
• array bounds error in analysis_tde fixed
• bugfix in read options for externbinary module

Tests:
• test for momentum and energy conservation in two fluid dust-gas drag
• code performance is now checked nightly against a suite of benchmarks
• sends error code to system if a fatal error happens (Pinte)
• added check on the conservation of angular momentum with dust/gas

Setup:
• Binary disc setup uses Farris et al. (2014) locally isothermal equation of state for discs around more than one star
• Disc setup routine modularised and made more general (Mentiplay)
• gwdisc setup now allows disc inclination (Pereira et al. 2019)
• setup_star given fairly major restructure so logic is clearer; more cleanly split interactive from non-interactive parts
• Flyby setup updated with the following roll angle convention: incl=0 => prograde orbit (disc and perturber anti-clockwise; incl=180 => retrograde orbit (disc anti- and perturber clockwise). See Cuello et al. 2019
• minor fixes to dustyshock and dustywave setups (Hutchison)
• binary_w in setup_disc is now 270 degrees by default
• asteroidwind setup added
• added option to setup a settled dusty disc, working with both one and 2 fluid (Dipierro)

Build:
• version number and git sha now written to dump file headers
• memory is now allocated at runtime for main arrays in Phantom (Chan). This avoids the need to recompile with MAXP= when you change the particle number. Only applies to main phantom binary at present, not to phantomsetup.
• many compiler warnings fixed
• cleanup of evolve module
• obsolete preprocessor flags -DSORT_RADIUS_INIT and -DDUSTFRAC deleted
• you can now supply JOBNAME=<your-job-name> when making job scripts with make qscript, otherwise it continues to choose delightful random words

Analysis:
• Multigrain post-processing works properly with MCFOST
• phantomevcompare will not duplicate data when merging files
• further integration with MCFOST
• analysis disc planet prints the effective tilt between the inner and outer disc (Nealon)
• disc analysis now defaults to sorting particles by cylindrical radius - this should fix any discrepancies that may have been occurring. Deliberately made it very hard not to chose this option (Nealon)
• disc analysis now returns the total angular momentum components as well (Nealon)
• precession files: these can now be made even if the first file input is not the first file of the simulation (Nealon)
• utils_disc now handles an eccentric disc - bins are defined by semi-major axis, not by radius (Nealon)
• analysis_dustydisc - Added check Ltot!=0 to prevent NaNs in the output (Ragusa)
• moddump_extenddisc implemented to extend an existing disc simulation in radius (Nealon)
• disc scale height now calculated from particle positions but works perfectly with a warped disc (Nealon)

Other:
• phantom outputs helpful error message if .setup file is given on command line instead of .in file
———————————————