next up previous
Next: WAVELENGTH SOLUTION Up: Calibration of X-ray CCDs Previous: INTRODUCTION

DESCRIPTION OF THE EXPERIMENTAL SETUP

The apparatus used for this experiment consists of an electron impact source, the grating spectrometer (HIREFS), and a CCD chamber (see Fig. 1). An electron impact source (Manson source) operates at 5 kVolts accelerating voltage and emits bremsstsrahlung radiation with energy up to approximately 2 keV. Emitted white light X-rays pass through filter stages (used to adjust the flux) and narrow (5 microns) entrance slit to hit the reflecting focusing optics and the grating of the spectrometer. Dispersed in the vertical direction, an X-ray beam illuminates a CCD in such a way, that every row sees different energy (CCD columns are parallel to the dispersion direction). In the horisontal direction beam is made wide enough to illuminate every column of the CCD.

  
Figure 1: Basic schematic of the calibration setup.

Spectrometer has two focusing mirrors and a variable-space reflecting grating, which allows it to produce a flat image on the CCD placed in the focal plane. The spectrometer produces multiple orders X-rays in any given direction. A tremendous advantage of using a high resistivity X-ray CCD in a combination with a spectrometer is that it allows to separate orders by analyzing an amplitude of the output signal, if an X-ray flux is low enough to avoid photon pile up. To illustrate how it can be done, on Fig. 2 is shown a three-dimentional distribution of the number of counts as a function of the CCD row number (x-axis) and a pulseheight (y-axis). Since dispersion direction of the spectrometer is parallel to the column of the CCD, all the events from the same row are identical energy-wise, and are summed together. Each order of the spectrometer produces a separate ridge in Fig. 2, and, hence, can be analyzed separately. Sharp peaks on the Fig. 2 correspond to carbon and oxygen emission lines of the source and are there due to a contamination of the surface of the source target by hydrocarbons.

  
Figure 2: Pulseheght (y-axis) distribution of the X-ray photons as a function of CCD row number(x-axis).

In a spectroscopic mode an X-ray flux illuminating CCD should be maintained at low levels to allow a separate detection of each photon. In order to get a statistically sufficient amount of data we typically collect 5000 frames in each calibration run. During an experiment CCD temperature was maintained at C to ensure a very low dark current at integration time as long as 7 seconds. Electronics used in this setup allows computer controlled data acquisition with readout noise normally below 3 electrons rms. For some chips noise level dropped below 2 electrons rms.



next up previous
Next: WAVELENGTH SOLUTION Up: Calibration of X-ray CCDs Previous: INTRODUCTION



gyp@space.mit.edu