ACIS Flight Gain Tables for FP Temperature -90C

Last updated August 30, 1999
cxccal@cfa.harvard.edu

Other releases:
Obtained during Chandra OAC phase using HRMA Contamination Measurement data and ACIS door source data. The focal plane temperature during this phase is -90c (nominal = -120c).
The gain table was derived using ACIS door source and FCM measurements obtained with a warmer focal plane temperature (-90c) than will be used during the GO phase (-120c). We expect small, secular gain drifts when the focal plane is cooled to -120c, near the end of the OAC phase of operation. The gain will be reevaluated at at that temperature using the next-in-line mode calibration source.

The gains were derived from linear, least-squares fits to measurements of gaussian centroids for 2-5 X-ray features in the PHA spectra (G02346). These features and their energies are tabulated on the webpage. The slopes and zero points are given for each node of each chip. The 1 sigma errors on the coefficients, assuming unit weights, are tabulated, as are the standard deviations about the means for each fit. In addition, we measured the FWHM for each line (G02346 and G0), but only the Mn Ka and Mn Kbeta values are reliable (the remaining lines being too faint to measure reliable widths).

We present an example spectrum with features indicated. The spectra, linear fits, residuals, and FWHM measurements can be viewed by clicking on "Gain plots, Postscript format" and "Spectra and fit data, HTML format".

Because the FCM did not uniformly illuminate the focal plane, the gains are not all measured to the same precision. For example the gains for S0, S1, S5, and I0 are less reliable than the others due to low flux levels. Order of magnitude variations in X-ray surface brightness are seen across the focal plane. In all cases where points are omitted or the residuals are large, the fluxes were low.

You may also find that the residuals in the low light level data are often not random. We tend to measure the centroid of the Mn Ka feature 2-3 ADU higher than the Mn Kbeta feature (or the converse). This will affect chip 5 (S1) more than the remaining devices, as we were able to measure reliable centroids only for these features in chip 5. The gains are least reliable at low energies, where fits are weighted by the Mn Ka escape feature and the Si Ka fluorescence feature. These features are down in flux by factors of 70-100 compared to Mn Ka feature, and their centroids are often difficult if not impossible to measure. The residual plots should be a guide to the reliability of the gains as a function of energy, but errors on the order of several ADU at low energies would not be surprising.

OBSID62742, OBSID62743

Lines measured:
Line Energy
[keV]
Mn Ka 5.89875
Mn Kb 6.49045
Mn Ka esc. 4.15877
Si Ka 1.73998
Au L 9.713

  • Spectra and fit data, HTML format: G02346, G0
  • Gain plots, Postscript format
  • ASCII table of gain values

    • OBSID = 62742, 62743
    Chip Node Gain [ADU/eV] Error [ADU/eV] Int. [ADU] Error [ADU] Std.Dev. [ADU]
    0 0 0.2599 0.0002 0.3 1.0 0.71
    *1 0.2514 0.0005 -1.6 4.0 1.10
    *2 0.2612 0.0006 -9.1 4.5 1.22
    3 0.2536 0.0004 -3.2 2.4 1.67
    1 0 0.2528 0.0002 0.0 1.2 0.98
    1 0.2490 0.0003 -5.7 1.7 1.16
    2 0.2507 0.0001 -3.2 0.5 0.40
    3 0.2526 0.0003 3.5 1.8 1.52
    2 0 0.2599 0.0001 -1.8 1.0 0.69
    1 0.2629 0.0008 3.6 5.2 4.32
    2 0.2640 0.0004 5.5 2.6 1.84
    3 0.2597 0.0002 0.5 1.3 1.05
    3 0 0.2544 0.0002 -0.4 1.2 0.95
    1 0.2525 0.0001 2.5 0.9 0.72
    2 0.2491 0.0002 0.7 1.1 0.92
    3 0.2544 0.0004 2.8 2.3 1.91
    4 0 0.2725 0.0012 -20.8 7.2 5.96
    1 0.2657 0.0003 0.3 1.6 1.35
    2 0.2585 0.0008 -17.8 5.0 4.09
    3 0.2559 0.0004 -2.3 2.5 2.09
    OBSID = 62742, 62743
    Chip Node Gain [ADU/eV] Error [ADU/eV]Int [ADU] Error [ADU] Std.Dev. [ADU]
    5 *0 0.2316 Inf 14.5 Inf 0.00
    *1 0.2199 Inf -15.1 Inf 0.00
    *2 0.2230 Inf -7.0 Inf 0.00
    *3 0.2276 Inf 2.0 Inf 0.00
    6 0 0.2518 0.0001 0.7 0.8 0.65
    1 0.2605 0.0001 2.1 0.9 0.72
    2 0.2535 0.0001 2.4 0.4 0.35
    3 0.2602 0.0002 1.3 1.1 0.88
    7 0 0.2226 0.0005 -6.0 3.0 2.48
    1 0.2138 0.0007 -9.5 4.4 3.64
    2 0.2158 0.0006 -9.1 3.6 2.97
    3 0.2243 0.0008 -10.8 4.8 3.93
    8 0 0.2314 0.0003 -1.9 1.6 1.33
    1 0.2260 0.0001 -0.5 0.6 0.46
    2 0.2280 0.0003 -2.0 1.9 1.54
    3 0.2374 0.0004 -2.6 2.7 2.21
    9 0 0.2538 0.0001 -3.1 0.5 0.38
    1 0.2560 0.0004 -9.3 2.4 1.97
    2 0.2573 0.0005 -1.4 3.0 2.46
    *30.2203 Inf 11.1 Inf 0.00

    Analysis Notes:
  • * 2-3 features measured
  • "Inf" error indicates two features measured
  • "Std.Dev" denotes standard deviation about the mean residual
  • All errors assume unit weights
    Software:

    DOWNLOAD these files if you want to use them, as they will look strange in your browser window due to some necessary HTML tags inside the programs.

  • l1merge.bat Reads in level 1 evt files and calls either l1merge or l1merge_2. Note: saves memory by deleting evt files after pulling out the necessary data.
  • l1merge.pro Takes evt data from l1merge.bat, sums spectra from 3 files to improve signal-to-noise, plots spectra, and fits gaussians to peaks: MnKa, MnKb, MnKa escape, SiKa, AuM, and MnL+FeL.
  • l1merge_2.pro Same as l1merge, except only sums 2 files and fits AuL instead of AuM and MnL+FeL.
  • mergeplot.bat, mergeplot.pro Reads in data files generated by l1merge/l1merge_2 and plots offset vs. energy and FWHM vs. energy; fits lines to offset plots for gain measurements.
    Back to the ACIS Calibration Page