RJ4d test

  • Hash: 177f0c9
  • Branch: develepment-bc

Bifrost

Intial conditions

$$\rho_L$$ $$v_{x,L}$$ $$v_{y,L}$$ $$v_{z,L}$$ $$P_L$$ $$B_{y,L}$$ $$B_{z,L}$$ $$\rho_R$$ $$v_{x,R}$$ $$v_{y,R}$$ $$v_{z,R}$$ $$P_R$$ $$B_{y,R}$$ $$B_{z,R}$$
1 0 0 0 1 0 0 0.3 0 0 1 0.2 1 0

Using the stagger2/bifrost solver, with a end time of t=0.2 and timestep of 0.01 produces the following plots

Intial conditions

Intitial conditions are the same as stated from the before. But now using ramses solver instead

v p rho

For the iout=16

The corresponding animations are

v_a p_a rho_a

Comparing them to the Fig 15 (Athena: A new code for astrophysical MHD, Stone et al)

Changes to the conditions

Will change all the bifrost parameters with a factor of 10. Since it is already established that it should look identical in all the axis'. I will only animate in the $$x$$-axis.

And the following animations v_a_rho p_a_rho rho_a_rho

See from the above animations that the only parameter worth studying closer for the time being is the change in velocity.

v_collage v_collage100 v_collage01

Observe that the parameters stabilize when the time progresses. But that for the $$1/10$$ density bifrost parameter, it will be slightly higher values then the original. And that most of the deviations stems from the intersection between the left and right side of the domain.

Ramses

Using the exact same procedure as above. But now with the solver=ramses/hlld instead. Where the original slope parameter is 3.5. We get the following

anim anim anim

Where we see that the change of slope has very litte effect on the domain. Except for the intersection between the left and right. But interestingly for a slope parameter of one the simulation crashes after one timestep.