Calibration status and caveats
- On-axis EEF
To evaluate the imaging capability and model the encircled-energy fractions (EEFs), we analyzed the XIS data of SS Cyg taken on 2005 November 5. The object was chosen because it is a moderately bright, point-like source and the pile-up effect of the CCD cameras is insignificant. Figure 1 shows a comparison of thus-obtained EEFs of the four XRT and the XISSIM simulations. For all integration radii from 1' - 6', the systematic error (the difference between the observation and the simulation) is less than 20%. Note that no error on the attitude control is included in the simulations. The information on calibrations of the EEFs at the inner-most region (<1') is not currently available for the reason that they are affected not only by the calibration error of the XRT but also by the attitude wobbling and the pile-up. Thus users who attempt to extract spectra from fairly small regions should pay attention to these issues. In addition, outside 5' the PSF contains a large calibration error (up to ~1000%) though it is not easily recognized in the plots of the EEFs. Thus the ARF files build with the XISSIMARFGEN would contain large systematic errors in the case that spectra were accumulated from some specific portions of the PSF.
Figure 1: Encircled-energy fractions (EEFs) of XRT, derived from SS Cyg observations (black) and the XISSIM simulations assuming an incident X-ray energy of 4.5 keV (red).
- Off-axis EEF
The PSF depends on position in the focal plane: the further off-axis becomes, the more distorted the PSF gets. We investigated the EEF by ray-tracing simulations and found that reproducibility of the EEFs is as good as that for the on-axis observations.
Related links
- Suzaku memo 2006-34 "The X-ray Telescope onboard Suzaku"