[Previous] | [Session 66] | [Next]
J. L. Matteson (UCSD), P. F. Dowkontt (Wash. U.), F. Duttweiler, W.A. Heindl (UCSD), P. L. Hink (Wash. U.), G.L. Huszar, E. Kalemci, P.C. Leblanc, R.E. Rothschild, R.T. Skelton (UCSD), K.R. Slavis (Wash. U.), E.A. Stephan (UCSD)
Coded mask imagers are appropriate for important objectives of high energy X-ray astronomy, e.g., gamma- ray burst localization, all-sky monitors and surveys, and deep surveys of limited regions. We report results from a coded mask imager developed to establish the proof-of-concept for this technique with CZT detectors. The detector is 2 mm thick with orthogonal crossed strip readout and an advanced electrode design to improve the energy resolution. Each detector face has 22 strip electrodes, and the strip pitch and pixel size are 500 microns. ASIC readout is used and the energy resolution varies from 3 to 6 keV FWHM over the 14 to 184 keV keV range. A coded mask with 2 x 2 cycles of a 23 x 23 MURA pattern (860 micron unit cell) was built from 600 micron thick tantalum to provide good X-ray modulation up to 200 keV. The detector, mask, and a tiny Gd-153 source of 41 keV X-rays were positioned with a spacing that caused the mask cells in the shadowgram to have a projected size of 1300 microns at the detector. Multiple detector positions were used to measure the shadowgram of a full mask cycle and this was recorded with 100 percent modulation transfer by the detector, due to its factor of 2.6 oversampling of the mask unit cell, and very high strip-to-strip selectivity and spatial accuracy. Deconvolution of the shadowgram produced a correlation image in which the source was detected as a 76-sigma peak with the correct FWHM and base diameter. Off-source image pixels had gaussian fluctuations that agree closely with the measurement statistics. Off-source image defects such as might be produced by systematic effects were too small to be seen and limited to <0.5 percent of the source peak. These results were obtained with the "raw" shadowgram and image; no "flat fielding" corrections were used.
The author(s) of this abstract have provided an email address for comments about the abstract: jmatteson@ucsd.edu