Here we give an overview of the method we use to generate the LSF used in analysis, how to apply the LSF and the derived products (eg. RMF) to analysis and a synospsis of the caveats that may effect the fitting of the data. Figure 1 shows the flow of data needed to prepare reference data for use in the analysis of CXO grating data. The starting point for this analysis are calibration derived MARX parameters and input data. These are changed when calibration updates are released. These CIP (Calibration Interface Products) are then used to generate MARX simulations which are used to characterize the HRMA + HETG or HRMA + LETG Line Spread function. These MARX simulations need further processing to convert them into products that the DS can use for analysis and for the convenience of the user. The products derived for the user are 1) the LSF library files, 2) the RMF files, and 3) the encircled energy (EE) library files.
The LSF library files are a one dimensional projection of the MARX events. This projection is done either in the dispersion direction (along the spectrum) or in the cross-dispersion direction (across the spectrum) and the events are binned into 1 m pixels. These profiles are then repackaged into the DS LSF library format and are available in The Grating's PSF Library, http://hea-www.harvard.edu/asclocal/sds/ICD/index.html These library files are currently used by PSF_Model which will return an LSF for a specific energy and zero order position and by PSF_Frac which will integrate the profile to determine the encircled energy fraction for a given radius.
The RMF files are the most familiar files for most X-ray spectral work. These are derived from the MARX 2D LSF data by fitting Gaussians LSF. The fits are then used in RSPgen (GSFC tool) to create RMF's for fitting extracted spectra. These RMF can be currently used with Sherpa or with XSPEC to fit lines in the extracted spectra, to fit models to the spectra or portions of the spectra.
The EE tables created from 1-D products mainly exist for use with the user tool, PSF_Size, to estimate the extent of a feature in the grating data. These tables will be used in future versions of the grating analysis tools (eg. to create shaped extraction regions).