|
|
|
| News | Issues | Miscellaneous |
| Basics | History |
May 2002: An error in some pipeline reprocessing of HRC-S/NONE data has been identified. Only level 2 files processed before before July 2001 are affected. Update: As of 19 August 2002, all affected data have been reprocessed.
Data from the HRC go through several levels of processing. In
level 1 processing, the results of filters evaluated on orbit and
tests performed during automatic processing
are recorded in a status flag for each event. The flag is 32 bits
long, with each bit representing the result of a test. Definitions
for each status bit can be found here. In
level 2 processing, this status flag is
used to filter out "bad" events. Here are the default status flags
used in standard processing:
Some of the HRC filtering tests require that coefficients be supplied to set limits on acceptable events. The coefficients currently in use by Chandra automatic data processing can be found here.
In addition to filtering on the status flag, event lists are also filtered on the good time intervals (GTIs), and, for HRC-S data, PHA=0:300.
The standard filtering done by pipeline processing software as described above is considered to be the optimal solution by the CXC calibration and HRC IPI teams. Custom filtering on status=0 is strongly discouraged.
Data from the HRC have a subset of events that are misplaced. Empirically, the fine position ratio and normalized central amplifier ratio define a hyperbolic locus of points where well-placed or "good" events will lie. The following images show these hyperbolas for the u and v axes.
U-Axis |
V-Axis |
 |
 |
In the H-test, this function is used to filter events. The bulk of the events that fail the H-test are from charged particles and thus part of the background. However, a small percentage of them derive from the source and generate ghosting and false jet-like features. For a brief discussion of displaced source events, please see Steve Murray's memo (March 2000). For more comprehensive information on event screening and the H-test, consult this SPIE paper by Murray et al, 2000.
The H-test is used to flag HRC-S data in level 1 processing, but events which fail are not removed from the level 2 event list. For more information on filtering HRC-S/LETG data, see LETG/HRC-S Background Filtering with CIAO.
For the HRC-I, however, events failing the H-test are removed from the level 2 event list. The effects of H-test filtering on HRC-I source and background events have been investigated by the HRC CXC Calibration team. For imaging, it is probably best to use the level 2 event list. On the other hand, for photometry, you may want to custom filter the level 1 event list, not excluding events which fail the H-test. This can be done by following the Create a New Level=2 Event File CIAO thread for the HRC-I, but filtering on status=xxxxxx00xxxx0xxx00000000x00000xx, where the last two xs indicate that either outcome of the H-test is acceptable.
As an example, the following images show different types of processing on an HRC-I calibration observation of ArLac. On the left is the level 1 image. The central image has been filtered like level 2 data, but without excluding events that fail the H-test. On the right is the level 2 image.
Level 1 |
Like Level 2, But No H-test |
Level 2 |
 |
 |
 |
The quality of HRC data has evolved over time due to improvements of Calibration Database (CALDB) Products and the Standard Pipeline Processing. As an example of how the revisions have improved image quality, we display an original HRC-I observation of ArLac done in October 1999 along with three reprocessings, shown in chronological order from left to right. The top row shows the level 1 images, and the bottom row shows the level 2 images. Note the improvement of the filtered (i.e. level 2) images with time.
# 1 - Nov 99 | # 2 - Aug 00 | # 3 - Dec 00 | # 4 - Apr 01 | |
Level 1 |
 |
 |
 |
|
Level 2 |
 |
 |
 |
|
Note that the numbers (1-4) above the images indicate the number of times this observation has been reprocessed, not the pipeline or CALDB version numbers used in reprocessing. To determine the pipeline and CALDB versions used in processing your observation, look for the ASCDSVER and CALDBVER keywords in the header of the data file. Release notes for each pipeline revision are available grouped by instrument, so one can track processing changes affecting HRC data products. CALDB release notes are also available.
The HRC IPI Team has developed a software tool (revised Nov 2001), screen_evt1, to allow users to screen their own data manually. See "HRC Event Screening" for more information about the tool, including how to download and run the code.
Last modified: 09/27/12
 |
The Chandra X-Ray
Center (CXC) is operated for NASA by the Smithsonian Astrophysical Observatory. 60 Garden Street, Cambridge, MA 02138 USA. Email: cxcweb@head.cfa.harvard.edu Smithsonian Institution, Copyright © 1998-2004. All rights reserved. |