AAS Meeting #193 - Austin, Texas, January 1999
Session 66. X-Ray/Gamma-Ray Detectors
Display, Friday, January 8, 1999, 9:20am-6:30pm, Exhibit Hall 1

[Previous] | [Session 66] | [Next]


[66.06] Absolute Calibration of AXAF - Transfer Standard Solid State Detectors

E.M. Kellogg, R. Edgar, W. McDermott, S Serej (SAO), J. Auerhammer, F. Scholze, G. Ulm (PTB-Berlin, Germany)

The objective of this work is to calibrate the effective area of AXAF as close as possible to 1%, to permit accurate absolute x-ray flux observations. We report in this paper on our work in calibrating transfer standard detectors used for prelaunch x-ray calibration of AXAF. We discuss the transfer standard high purity germanium detector, referred to as the SSD. The absolute calibration of SSD quantum efficiency uses two approaches. The first is calibration at about twenty or thirty energies, each having a bandwidth of ~5 eV, over the range 400-1700 eV, using the PTB cryogenic electrical substitution radiometer as a ~1% standard for input flux to the SSD. The second is a broad-band calibration from ~3-6 keV using the continuum spectrum of synchrotron radiation from the BESSY storage ring. In the latter case, we rely on an absolute prediction of the incident flux from Schwinger's formula, using accurately measured values of the storage ring magnetic field, electron beam energy and current, source-to-detector distance, detector aperture size, and beam emissivity. This prediction is accurate to ~0.5%. We construct a model of the detector response, and fit the pulse height spectra from the narrow-band beam. This model contains an energy scale, main peak with a fano factor, incomplete charge collection shelf and tails. Continuing the work of P.~Barnes et al.~(Proc SPIE 1998), we develop a detector response matrix for use in the XSPEC fitting program, based on the detector model, by fitting the pulse height spectra from each narrow band energy, obtaining detector model parameters, and deadtime corrected counting rates. The incident flux rate is given by the response of a set of secondary standard photodiodes in the storage ring for the narrow band case. The ratio of detected to observed rates gives the QE of the detector at the measurement energies, in the energy range where the detector's windows, nominally 2000 A Al, 12000 A parylene and ~20000 A ice, dominate the QE. We will fit this to a Henke x-ray absorption model and incorporate the result in the final detector model. Using the response matrices, we then fit the pulse height spectra from the broad-band beam, giving the measured broad-band QE.

This work is supported by NASA contract NAS8-40224.


[Previous] | [Session 66] | [Next]