next previous
Up: BeppoSAX Low-Energy Concentrator Spectrometer


3 Data selection and processing

  In order to study the properties of the LECS background, data were extracted from 350 BeppoSAX Final Observation Tapes, corresponding to 170 individual observations. These data were processed using the standard LECS processing pipeline described below and cleaned event lists for the intervals when the instrument was observing both the dark Earth and the sky in the nominal operating configuration extracted. Observations with <100 counts were excluded. This left total sky and dark Earth exposures of 2847 ks and 498.6 ks, respectively.

Table 1 gives the contents of the LECS telemetry for each detected event. The standard LECS data processing package (SAXLEDAS 1.8.0) performs the following steps to produce cleaned, linearized event lists:

1.
Conversion of RAWX and RAWY into linear coordinates referred to as DETX and DETY using the algorithm given in Parmar et al. (1997b).
2.
Determination of the overall instrument gain (which is temperature dependent), using the mean energy of the two calibration sources as a reference.
3.
Conversion of Pulse height channels to Pulse Invariant (PI) channels which have a fixed relation to energy. This conversion takes into account the overall gain and a FOV position dependent gain correction.
4.
Selection of intervals with nominal instrument settings (such as high voltage settings), and separation into events that are >4$^\circ$from the Earth's limb (taken to be good on-target intervals), and those where the dark Earth occults the FOV.
5.
Removal of background events using the VETO and BL signals. The VETO signal is the ratio between the light measured by the central PMT anode and the sum of that measured by the surrounding anodes. It therefore has a high value for on-axis X-ray sources and lower values for events occurring further off-axis. In SAXLEDAS 1.8.0 a sliding cell VETO window is applied to the data. The BL signal is a measure of the duration of the light flash produced following absorption in the gas cell. It depends on the penetration depth of an event into the cell, which is energy dependent. In SAXLEDAS 1.8.0 events are selected if their PI and BL values are within a set of 62 pre-defined boxes.

The standard LECS data analysis technique is to extract counts within a radius of 8' (corresponding to 35 RAW pixels; see Fig. 1) from the source centroid using the cleaned event lists produced by SAXLEDAS. This large radius ensures that 95% of the 0.28 keV X-rays are included in the analysis. For absorbed, or very faint sources, extraction radii of 4$^\prime$ or 6$^\prime$ may be appropriate. The BeppoSAX pointing accuracy is such that most target sources are located within $\sim$2' of the nominal position of RAWX, RAWY = 131, 124.


  
Table 1: LECS telemetry contents

\begin{tabular}
{lll}
\hline\noalign{\smallskip}
Parameter & Range & Comment \\ ...
 ... & $0-255$\space & Event burstlength \\ \noalign{\smallskip}
\hline\end{tabular}


next previous
Up: BeppoSAX Low-Energy Concentrator Spectrometer

Copyright The European Southern Observatory (ESO)