Context Navigation

← Previous Post
Next Post →

Update 10/11

Ran simulations with small and zero magnetic fields, as well as turning off the planet boundary (Parker wind out to bow shock) and lowering the CFL variables to 0.75, 0.1, 0.3. Runtimes below:

Run	Time
B=0	308s
B=1e-10	302s
No planet boundary	17000s (~5 hrs)
CFL decreased	68000s (~18 hrs)

Seems the magnetic field and the incorrect bow shock radius are both playing a role, but bow shock radius is the bigger problem (since we need magnetic field). Some side-by-side comparisons:

B=0, small

Boundary off, B small

Boundary off, CFL low

And alone, without all of the mess:

Some higher-resolution simulations:

2 levels AMR

2 levels AMR, with B field

Full time