Spritz code: Improving the robustness of GR Magnetohydrodynamics

General relativistic magnetohydrodynamic (GRMHD) simulations are an indispensable tool to obtain a detailed physical understanding of binary neutron star (BNS) mergers. Flux-preserving GRMHD codes numerically evolve a set of conservative equations based on conserved variables which then need to be converted back to the primitive physical variables. The corresponding conservative-to-primitive (c2p) recovery procedure is a crucial aspect at the core of any GRMHD code. In this talk, I will discuss a completely new c2p recovery scheme for ideal MHD. Using both stand-alone tests and by performing demanding three-dimensional GRMHD tests, we have demonstrated robustness, accuracy, and efficiency over the whole relevant parameter space. This includes the critical case of very strong magnetizations. I will also talk about technical aspects of implementing the scheme in our new GRMHD code Spritz, such as the requirements for the EOS framework.