CALDB Public Release Notes Version 3.0.4 Effective Date: 2005-06-14T13:00:00 To be installed in standard data processing (SDP) only; New files will be released in CalDB 3.1.0 later in June of 2005. Summary of Changes: A. ACIS split thresholds file installation Filenames: acisD1999-11-01evtspltN0002.fits acisD2005-07-01evtspltN0002.fits Location: $CALDB/data/chandra/acis/bcf/evtsplt/ Up until now, CIAO and SDP tool "acis_process_events" has run with a split threshold set literally at the value of "13" by the "spthresh" parameter in the parm file. The existing CalDB file acisD1996-11-01evtspltN0001.fits was never used in either SDP or in CIAO. With DS 7.6.0 and CIAO 3.2.2, scheduled for release in June 2005, this will change, and the split thressholds will be read from the EVTSPLT files in the CalDB. However, the format of the EVTSPLT file has changed slightly, so that the N0001 file needs to be replaced with a new file, specifically of the format specified for version N0002. See the Technical Details section for the format description. Motivating this change, is that in the near future after the installation of DS 7.6.0, the parameter blocks for CC-MODE ACIS observations will specify/require a split threshold value of "16" rather than "13" for all of the active front-illuminated chips. Hence setting the value to "13" be default is no longer acceptable at least for CC-Mode. This change in the CC-mode split threshold is simultaneous with and required by the new CC-mode bias algorithm being introduced in processing with DS7.6.0. Pipes/tools affected: ACIS L1 and CIAO acis_process_events Threads affected: "Create a New Level=2 Event File" http://cxc.harvard.edu/ciao/threads/createL2/ and related acis_process_events threads. This is a MINOR CHANGE only: the appropriate split threshold value will be read from the CalDB for each chip and node, rather than using the default "SPTHRESH" parameter value. B. HRC GAPLOOKUP table files for HRC-I and HRC-S Filenames: hrciD1999-07-22gaplookupN0001.fits hrcsD1999-07-22gaplookupN0001.fits Location: $CALDB/data/chandra/hrc/bcf/gaplookup/ >From F. Primini, in a memorandum entitled "Specifications for improved degap corrections in hrc_process_events," dated 12 April 2005: "The HRC event positioning algorithm results in periodic "gaps" in positions, which are corrected by applying a 5th-order polynomial correction to coarse event positions (see Section 6 of the ARD for the HRC Level 1 Pipeline). The polynomial coefficients were determined from flat-field data taken pre-launch. For HRC-S, analysis of flight data has revealed inaccuracies in the current "degap" solution which contribute to the observed non-linear dispersionrelation in the LETG+HRC-S (Kashyap, Drake, and Chung, 2004, Proc. SPIE 5488-71). Independently, M. Juda has demonstrated inaccuracies in the HRC-I degap solution as well, albeit for a limited range in position near theaim-point. Reviews of the degap status for both HRC-I and HRC-S were held on Feb. 17 and Feb. 23, 2005. [snip] It was recommended that we replace the HRC-S degap solution with simple look-up tables which incorporated both the current corrections plus the additional corrections determined by Kashyap, Drake, and Chung. Similar look-up tables would be used for HRC-I, derived entirely from the current HRC-I degap solution." At this time, look-up tables have been derived for both HRC-I and HRC-S, which have been derived entirely from the current HRC-I and HRC-S degap solutions. That is, the additional corrections for HRC-S from Kashyap, Drake,and Chung 2004 have NOT been added to the proposed HRC-S gap lookup table. "hrc_process_events" for both DS 7.6.0 and CIAO 3.2.2 have been modified to include the new degap algorithm, which requires the new look-up table format in the CalDB to be used. Hence the new files are required for the installation of DS 7.6.0 in processing.For HRC-S, an update to its look-up table is forthcoming, but has not yet been released. Pipes/Tools affected: HRC L1 processing (hrc_process_events) in DS 7.6.2 (TBD release date) CIAO hrc_process_events (CIAO 3.2.2 patch and later versions) Threads affected: HRC-I Degap Correction http://cxc-test.harvard.edu/ciao/threads/hrci_degap/ HRC-S Degap Correction http://cxc-test.harvard.edu/ciao/threads/hrcs_degap/ Note: The change in the threads above is that the new GAPLOOKUP files will be read into hrc_process_events rather than the now-obsolete "DEGAP" files. This is a data-neutral change; there is no change to the results at this time. II. Technical Details A. ACIS split threshold files There are two extensions in the EVTSPLT N0002 files, the first is for Timed Event mode, and the second is to be selected for Continuous Clocking mode: dramen-40: dmlist acisD2005-07-01evtspltN0002.fits block ------------------------------------------------------------------------------ Dataset: acisD2005-07-01evtspltN0002.fits ------------------------------------------------------------------------------ Block Name Type Dimensions ------------------------------------------------------------------------------ Block 1: PRIMARY Null Block 2: AXAF_EVTSPLT Table 3 cols x 10 row Block 3: AXAF_EVTSPLT Table 3 cols x 10 row The column specification includes the CCD number (CCD_ID), the event threshold values for each node of a given CCD, and the corresponding split threshold values. The event thresholds in these files are NOT used in acis_process_events; they are included for completeness in these CalDB files. dramen-41: dmlist acisD2005-07-01evtspltN0002.fits cols ----------------------------------------------------------------------- Columns for Table Block AXAF_EVTSPLT ----------------------------------------------------------------------- ColNo Name Unit Type Range 1 CCD_ID Int2 - 2 EVT_THR[4] adu Int2(4) - 3 SPL_THR[4] adu Int2(4) - The first or TE mode extension of both files is exactly as follows: dramen-42: dmlist acisD2005-07-01evtspltN0002.fits data ----------------------------------------------------------------------- Data for Table Block AXAF_EVTSPLT ----------------------------------------------------------------------- ROW CCD_ID EVT_THR[4] SPL_THR[4] 1 0 [38 38 38 38] [13 13 13 13] 2 1 [38 38 38 38] [13 13 13 13] 3 2 [38 38 38 38] [13 13 13 13] 4 3 [38 38 38 38] [13 13 13 13] 5 4 [38 38 38 38] [13 13 13 13] 6 5 [20 20 20 20] [13 13 13 13] 7 6 [38 38 38 38] [13 13 13 13] 8 7 [20 20 20 20] [13 13 13 13] 9 8 [38 38 38 38] [13 13 13 13] 10 9 [38 38 38 38] [13 13 13 13] These same data are in the second or CC-mode extension of file acisD1996-11-01evtspltN0002.fits, the active CC-mode extension for data taken before the parameter block change. The second extension of the acisD2005-07-01evtspltN0002.fits (for after the CC-mode parameter block change) has the following data: dramen-206: dmlist acisD2005-07-01evtspltN0002.fits"[3]" data ----------------------------------------------------------------- Data for Table Block AXAF_EVTSPLT ----------------------------------------------------------------- ROW CCD_ID EVT_THR[4] SPL_THR[4] 1 0 [38 38 38 38] [16 16 16 16] 2 1 [38 38 38 38] [16 16 16 16] 3 2 [38 38 38 38] [16 16 16 16] 4 3 [38 38 38 38] [16 16 16 16] 5 4 [38 38 38 38] [16 16 16 16] 6 5 [20 20 20 20] [13 13 13 13] 7 6 [38 38 38 38] [16 16 16 16] 8 7 [20 20 20 20] [13 13 13 13] 9 8 [38 38 38 38] [16 16 16 16] 10 9 [38 38 38 38] [16 16 16 16] where it is clear that the split thresholds for the FI chips are all "16", and for the BI's they are still "13" as prescribed in REASONS for CHANGE above. B. HRC GAPLOOKUP Tables See the DEGAP Calibration web page for the HRC, at http://cxc.harvard.edu/cal/Hrc/degap.html for any technical details on the current calibration. Note that for this CalDB upgrade, there is NO CHANGE in the calibration per se, but only a change in the format of the CalDB file for future adjustments to be made more accurately.