Observations when using GRHydro_Avec

Issue #1988 closed
jmahlmann created an issue

Dear Sir or Madam, I am currently working on problems including the evolution of the vector potential. With this I encountered some problems in setting up my simulations. Please allow me to state some of my observations:

  • The atmosphere reset routines are not scheduled all in their GRhydro_Avec version (e.g. the initial reset in CCTK_Initial)

  • Within the atmosphere reset routines, it is not made use of the GRHydro_Avec routines of variable conversion. (prim2conAM not used but the magnetic field ones)

  • The subroutine GRHydro_SqrtSpatialDeterminant is not always scheduled before the GRHydro_BvecfromAvec. Therefore NaNs are produced in GRHydro_BvecfromAvec.

I do have further problems in GRHydro_CalcUpdate at the moment, however, I do not know yet where they lie. Please contact me, if there is interest in reviewing the GRHydro_Avec implementations. As I will use this option extensively, I will definetely be working further on this.

My best Jens Mahlmann

Keyword: GRHydro_AVec

Comments (5)

  1. anonymous
    • removed comment

    I find that in GRHydro_CalcUpdate segmentation faults occur as it is tried to assign values to Bconsrhs even though it is not allocated due to the evolution of Avec when GRHydro_Avec is used.

  2. Roland Haas
    • removed comment

    Hello Jens,

    the AVec evolution scheme is not used very often and it is quite possible that it does not always work as expected. You are very welcome to propose patches / pull requests for it to help us make the ET better.

    You may also want to have a look at IllinoisGRMHD (also in the ET) which has its own version of AVec evolution.

    Yours, Roland

  3. Log in to comment