1: \begin{abstract}

2: 

3: High-accuracy numerical relativity simulations of binary neutron star

4: mergers are a necessary ingredient for constructing gravitational

5: waveform templates to analyze and interpret observations of compact

6: object mergers. Numerical convergence in the post-merger phase of such

7: simulations is challenging to achieve with many modern codes. In this

8: paper, we study two ways of improving the convergence properties of

9: binary neutron star merger simulations within the

10: Baumgarte-Shapiro-Shibata-Nakamura formulation of Einstein's

11: equations. We show that discontinuities in a particular constraint

12: damping scheme in this formulation can destroy the post-merger

13: convergence of the simulation. A continuous prescription, in contrast,

14: ensures convergence until late times. We additionally study the impact

15: of the equation of state parametrization on the pre- and post-merger

16: convergence properties of the simulations. In particular, we compare

17: results for a piecewise polytropic parametrization, which is commonly

18: used in merger simulations but suffers unphysical discontinuities in

19: the sound speed, with results using a ``generalized" piecewise

20: polytropic parametrization, which was designed to ensure both

21: continuity and differentiability of the equation of state. We report

22: on the differences in the gravitational waves and any spurious

23: pre-merger heating, depending on which equation of state

24: parametrization is used.

25: \end{abstract}

26: