PIN flare events after 28 July 2007
Update on Dec 9, 2008
Due to an increase in the low eneergy noise, PIN data taken since 2008 September 1 are processed using a new threshold value. There is no need for the users to re-process their data in this case.
One of the 64 PINs of the HXD started showing flare-like events around the low energy threshold on 28 July, 2007. In response, the HXD team has decided to (1) keep the on-board settings the same; but (2) raise the software lower discriminator threshold in ground processing.
However, some Guest Observers have already received data using a version of the processing pipeline/calibration databse before the threshold has been changed.
1. Reprocessing
These noise events in observations taken after 28 July, 2007, are not excluded in the processing pipline v2.0.6.13 (users can check this version by looking up the PROCVER keyword in their FITS files). Observations taken before 28 July, 2007 are unaffected.
We hope to reprocess the affected data in due course.
In the meantime, users with such data can reprocess their own data, starting from the cleaned HXD event files. This requires the latest version of the Suzaku FTOOLS (HEAsoft v6.3 or later). The reprocessing steps are:
- Re-run hxdgrade
- Re-select valid PIN events
(It is also possible to reprocess the data using unfiltered HXD event files, as the earlier version of this document stated. However, that is not necessary.)
First, users need To re-run hxdgrade on the cleaned event file to flag these noise event as bad, using the updated calibration file ae_hxd_pinthr_20070822.fits. If the user is accessing the latest version of the Suzaku CALDB (2007-09-15 or later), hxdgrade should automatically select the correct file by specifying hxdgrade_pinthres_fname=CALCB.
Once this is done, the resulting event file should be screened using the criterion of DET_TYPE==1. This can be done in xselect using the following set of commands.
xsel> filter column xsel> "DET_TYPE=1:1" xsel> extract event xsel> seve event
2. Response Matrices
Due to changes in bias voltages and the threshold (described above), users must now choose PIN response matrices that are appropriate for the epoch of observation. Note that this supercedes the earlier recommendation to select only those PIN units with the bias voltage of 500V.
| Epoch | File(s) |
|---|---|
| 2005.8.17 -- 2006.5.13 | ae_hxd_pinXXXXXe1_20080129.rsp (XXXXX=xinom,hxnom,flat) |
| 2006.5.13 -- 2006.10.2 | ae_hxd_pinXXXXXe2_20080129.rsp (XXXXX=xinom,hxnom,flat) |
| 2006.10.2 -- 2007.7.28 | ae_hxd_pinXXXXXe3_20080129.rsp (XXXXX=xinom,hxnom,flat) |
| 2007.7.28 -- 2008.8.31 | ae_hxd_pinXXXXXe4_20080129.rsp (XXXXX=xinom,hxnom,flat) |
| 2008.9.1 -- ** | ae_hxd_pinXXXXXe5_20080716.rsp (XXXXX=xinom,hxnom,flat) |
These files are available from the Suzaku CALDB (2007-09-15 version or later).
The non X-ray background files, to be made available shortly, also corresponds to all PIN units and appropriate bias voltage/threshold settings.
If you have any questions concerning Suzaku, visit the Feedback form.