ACIS Calibration Issues in Progress

08/11/2002 L. David has written a memo covering the specifics of the QE degredation reported on by N. Schulz. Analysis of ACIS Data Affected by the Low Energy QE Degradation

                Norbert Schulz  -  04/03/02
                nss@space.mit.edu

Dear User,

We are currently looking into an apparent degradation in the low energy detection efficiency of the ACIS instrument. The QE reduction appears to affect the counting rates of all detectors. Some degradation with time, for example due to a deposition of contaminants on the CCDs or the optical blocking filter, was expected. However the rate of degradation appears to be somewhat higher than projected from pre-launch studies. We are currently investigating the effect and possible solutions.

The degradation is not yet severe above 0.7 keV, but may in some cases influence the spectral analysis at low energies, i.e. below 1 keV, most likely below 0.5 keV. Above 1 keV we do not observe such a degradation. The effect increases towards lower energies and spectral bins near the carbon K edge (284 eV) are most affected. Also, since this is a time dependent effect, data from the early phases of the the mission are significantly less likely to be affected. So far we cannot give more quantitative information other than we face changes near the oxygen K edge of the order of 20% over a two year period.

We will keep the users informed about further progress on this matter and we will post more quantitative information as it becomes available.

                Alexey Vikhlinin  -  03/01/02
                alexey@head-cfa.harvard.edu

We have discovered that ACIS-S3 spectra may contain spurious features near the energies 1.6 and 2.6 keV. This problem is present only in the spectra extracted in the PI channels from observations performed after Jan 29, 2000 (at the focal plane temperature -120 C) and processed with the gain tables released in August 2001. The PHA spectra are not affected. We believe that the features are the result of abrupt changes in the slope of the PHA-Energy relation at these energies. CXC is working on the new gain tables that will eliminate the problem. In the meantime the users are urged to exercise caution in interpreting weak emission and absorption features near 1.6 and 2.6 keV.

• XMM/Chandra cross calibration presentation presented by Steve Snowden

• Calibration Review, Oct 30-31, 2001
  

                Richard J. Edgar  -  12/01/01
                edgar@head-cfa.harvard.edu

The response of the Back Illuminated (BI) chips is being reworked.

• We expect to release new BI FITS Encoded Function files (FEFs, from which the mkrmf tool makes response matrices) in July 2001. These include fits to the monoenergetic pulse-height distributions including up to ten Gaussian components (vice 3 in the CALDB 2.0 release), which allows us to model the low pulse-height tails of the distributions. Fits to astrophysical data are much improved at low energies, and fit parameters such as temperatures, absorbing column densities, etc. are more physical with the new products.
• S3 products for -120 C will be the highest priority, followed closely by S3 products for -110 C, and then S1 matrices at both temperatures.
• Several bugs in the process are being investigated as of this writing.

The January 2000 (CALDB 2.0) release of the BI matrices has systematic problems with its energy scale and line widths. See discussions here for the energy scale, and here for the line widths.

Quantum Efficiency. There is some evidence (ref: Herman Marshall's page http://space.mit.edu/ASC/calib/letg_acis/ck_cal.html) that the S3/S2 sensitivity ratio is larger than we have advertised by 10-20% in the energy regime below 1 or 1.2 keV. We are investigating this by several methods, including flight grating spectra of multiple AGN and QSO targets, re-analysis of low-energy ground calibration data, and fitting of spectra from other flight targets (notably hot clusters of galaxies). It's difficult to disentangle this effect from the (to date poorly modelled) asymmetries in the line shape function at low energies. Re-analysis of ground calibration data will be required to decide which (if either) of the QE curves is the correct one. This is an ongoing topic of debate within the calibration group and the PI team, and may not be resolved until late 2001 (or later).

Front Illuminated (FI) chip response, and CTI correction

• A Charge-Transfer-Inefficiency (CTI) corrector is being developed by the CXC data system. This will allow the user to correct, on the ground, data taken with ACIS, to recover a portion of the pre-flight performance of the instrument. New response products will be needed to use with these corrected data.
• Starting July 2001, an intensive effort will be underway to produce response products for both CTI-corrected and uncorrected data from the FI chips of ACIS. We expect to release these products in late 2001.
• The released (CALDB 2.0) response products model lines in ACIS as Gaussians which are substantially narrower than the data in some cases. We have posted a caveat message about this here: http://asc.harvard.edu/ciao2.1/caveats/specres.html

Also in work:

• Make a Bad Pixel list for -110 C
• Rework web page to make it easier to find descriptions of CALDB products and files.

Here are the viewgraphs presented to the Chandra Users' Committee on 2001-JUN-29. http:CUC_2001_06_29.ps

Last modified: 11/19/03

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.