RESIK & Diogeness NEWS

Data gaps due to missing telemetry: ~ 33 h.

Note, that the link to the catalogue page  is now: http://www.cbk.pan.wroc.pl/resik_catalogue.htm.

RESIK Dispersion – Simple Approach

Please note update of the  weekly 15, 2003 RESIK and RHESSI Observations of the Thermal Component of Flare X-ray Emission - the plot just has been added showing RHESSI and RESIK energy spectra plotted together.

For purposes of line identification it is of primary interest to know the relation between the wavelength of the crystal-diffracted radiation and the spectral bin No. where it is recorded. RESIK spectrometer blocks have been pre-calibrated in this respect at Rutherford Appleton Laboratory (RAL) in mid 2000 and the so-called linearity tests of the detectors have been made at Mullard Space Science Laboratory (MSSL). The results of the calibrations are stored on files and await for the interpretation. The interpretation appears to be a large task, for which we do not have at present enough human resources in Wroclaw. However, the knowledge of dispersion is a necessity, so we have made an attempt to estimate it based on the observed positions of identified emission lines and their respective wavelengths known from theory or earlier measurements.

In Figure 1, we present a portion of the „construction” scheme showing the crystal-detector geometry.

Figure 1. Section of the construction scheme (dr. Charlie Brown, NRL) showing relative orientation of the bent crystal wafer, the detector Be window aperture, the anode, hidden bins and crystal-illuminated portions of the anode for RESIK channel #2.

It is seen from the scheme, that the crystal-diffracted (or „reflected”) X-rays illuminate the position-sensitive detector through the Be entrance window over the length of ~9 cm (green bar). Solar X-ray spectrum is formed over the length of a blue portion of the anode only; „hidden” bins are not directly illuminated from the Sun. Owing to the special construction, the position of X-ray photon absorbed in the Ar-Xe detector gas can be determined from the ratio of charge measured on „both ends” of detector’s wedge-and-wedge position encoding system. The accuracy of photon arrival position determination is ~ 0.3 mm, therefore ~ 500 position bins can be resolved over the length of the anode.

In order to secure the maximum possible resolution of the 8-bit position readout system, a special so-called position lookup table is used, stored in the spectrometer’s EPROM memory. The lookup table is arranged so to decrease the accuracy of position readout in the hidden bin portions, increasing in return, the accuracy of the readout in the portion of anode sensitive to crystal-reflected radiation. The characteristic of the adopted position encoding (or „bin compression”) scheme is shown in Figure 2.

Figure 2.  Electronic „bin compression” pattern stored in the spectrometer memory lookup table. The 45 (hidden) end position bins are compressed to 10 bins (with indexes: 0 – 9 and 246 – 255). Saved (70) bins increase resolution in the central portion of the detector where the spectrum is recorded, pushing the spectral bin size down towards the intrinsic detector’s position resolution.

The adopted compression scheme stretches the spectral dispersion by a factor of 1.42 in the central portion of the detector. This effect will be taken into account in the forthcoming analysis of the ground calibration data.
For purposes of the present prompt analysis of RESIK spectra, we decided to use a simple linear approximation for the spectral dispersion. Derived parameters (see Table 1) describing the wavelength – bin No. dependence have been determined for a specific flare observed early in the mission on 22 April 2002 at 00:10 UT. For this flare (M4.4, S06W12), the dependence of wavelength λ [Å] and the (electronic) bin No p_b for the first order reflection is:

λ =b_0i + d_0ip_b

Where d_0i represents (assumed linear) dispersion in [Å]/bin in a particular spectral channel „i”. b_0i represents the shortwavelength limit in each channel. Values of coefficients have been determined (by JS and KJHP – Ken Phillips ) from (pairs of) the observed positions of stronger identified emission lines and their known theoretical wavelengths. Estimated dispersion values are roughly inversely proportional to the bent radii of the crystals (as expected).
Depending on the position of the source (flare), the lines „move” along the dispersion of the spectrometer. The calibration of this effect is in progress.

Prepared by: Janusz Sylwester js@cbk.pan.wroc.pl and Barbara Sylwester bs@cbk.pan.wroc.pl

Page made on 2  May 2003 by:  Anna Kepa ak@cbk.pan.wroc.pl and Jarek Bakala jb@cbk.pan.wroc.pl