Specifying model parameters and other options

RNS has several options which allow you to specify model parameters and choose from different output formats. A model is defined uniquely by specifying two parameters - one will always be the central energy density and the other can be one of the following: mass, rest mass, angular velocity, angular momentum, or the ratio of the polar coordinate radius to the coordinate equatorial radius (axes ratio, see definition of coordinates later in the text) .

The parameters are specified using the following flags:

-e central energy density in ${\rm gr/cm^3}$

-r axes ratio

-m mass in ${\rm M}_{\odot}$

-z rest mass in ${\rm M}_{\odot}$

-o angular velocity in ${\rm 10^4 s^{-1}}$

-j angular momentum in ${\rm GM_{\odot}^2/c}$

Note that if a polytropic star is requested, dimensionless units should be used. Consult the tables given in Cook et. al 1994 to find suitable values for these parameters.

The code is written so as to directly construct a model when the axes ratio is specified (which is therefore the fastest option). If one specifies a different parameter e.g. mass, the code constructs several (usually more than ten) models by varying the axes ratio, until it finds a model, for which the chosen parameter is within some allowed tolerance of the specified value. The default tolerance is 10^-4 (relative error) and can be changed by using the flag

-b tolerance

A smaller tolerance means more models will be constructed.

A model is constructed by iteratively solving the field equations and the hydrostatic equilibrium equation, until the coordinate equatorial radius changes by less than a specified relative accuracy. The default accuracy is 10^-5and this can be changed by using the flag

-a accuracy

Convergence is monitored by printing out the relative difference in the coordinate equatorial radius from one iteration to the next. This print-out can be suppressed by using the flag

-d 0

In rare cases, such as for unstable models of very stiff equations of state, the iteration may not, at first, converge. Such cases can easily be fixed by using a relaxation factor in the iteration. A factor of 0.8 usually makes the iteration convergent (the default is 1.0, which amounts to no relaxation). It can be specified using

-c relaxation factor

Specifying the above parameters is not sufficient to get the program started. One also needs to select the task that is to be performed.