Energy Scale of XIS Taken with a Window Option
Updated 2010-05-17
The XIS team has updated the energy scale assignment tool, xispi, in Version 12 of Suzaku FTOOLS.
This revision incorporates an update of the CTI correction formula that is important for users analyzing XIS data taken with a (1/4 or 1/8) window option. All such users should install Version 12 of Suzaku FTOOLS, and reprocess the data with the new version of xispi. Although it is a good practice to always use the latest calibration files through the Suzaku CALDB, this update of xispi is independent of CALDB update and is not tied to specific versions of calibration files.
Explanation of the Change
We have noticed a discrepancy in the energy scale of window option data, compared with the full window data. For observations with SCI off, the 1/4 window mode data are offset by about 40 eV at 6 keV for the FI CCD, and 60 eV for the BI CCD (=XIS1).
This discrepancy has been resolved by considering the different types of charge transfer. Radiation damage has created charge traps, leading to charge transfer inefficiency (CTI), reducing the pulse height that is out for the same amount of charge originally created by an X-ray photon. However, CTI per transfer is different between fast charge transfers and slow charge transfers. Because the numbers of fast and slow charge transfers are different between the full window mode and the window mode, this leads to differences in energy scales.
The new xispi incorporates a new formula that takes the two types of charge transfer fully into account. With this, the energy scale of the FI and BI data in 1/4 window mode with SCI-off are in agreement to within +/-20 eV at 6 keV with that of the full window mode.
Similarly for SCI-on data, the energy scale of the 1/4 window mote data has been confirmed to agree well with that of the full window mode.
If you have any questions concerning Suzaku, visit the Feedback form.