There were three primary goals for the High Resolution Camera (HRC) measurements conducted at the XRCF in Huntsville. First, and foremost was the determination of the combined response of the mirror/detector assembly to X-rays as a function of energy. Second was the measurement of the count rate linearity and third was the generation of a map of the point-spread-function (PSF).
Because the optics are a grazing incidence system and the critical angle of incidence is energy dependant, the reflecting area of the mirror is also energy dependant. This energy dependant area is combined with the total quantum efficiency of the detector (micro-channel plate (MCP) and UVIS) to produce what is referred to as the effective area.
Great pains were taken to sample a wide range of energies and locations on the detector in order to characterize the spatial uniformity of the HRC response as well as the energy dependance. As a result the bulk of the testing performed at the XRCF was aimed at characterizing the total effective area of the combined HRC/HRMA assembly.
One issue of concern, due to the limited energy resolution of the HRC, was the spectral content of the X-ray beam. It was important to make sure that the ratio of the flux at the energy of interest to that of all other contaminating lines and the continuum remain high so that the interpretation of the data would be unequivocal. To this end, we employed the two monochrometers available, the Double Crystal Monochrometer (DCM) and the High-Resolution Erect-Field Spectrometer(HIREFS).
HIREFS coverage spanned the range from 0.400 keV to 2.000 eV, while the DCM covered from 1.500 keV to 10.000 keV. Observations of three Electron Impact Point Sources (EIPS), C, B and Be, at energies of 0.277, 0.183 and 0.109 keV respectively, provided the extreme low energy coverage unavailable with the monochrometers.
Because we expect the response curve to be fairly complex due to the many features from both the HRMA and the HRC, great care was taken to measure the system response above and below all known absorption edges. In all, the effective area was measured on-axis at over 70 energies. Most of these measurements were conducted slightly out of focus so as to allow a high flux which reduced the time needed to obtain good photon statistics. This defocus condition also protected the HRC against over exposure.
One complication of the testing process was that for energies between 0.400 and 1.000 keV, i.e. while using the HIREFS source, we expected moderate spectral contamination due to higher orders. In other words, the 0.400 keV line measurement would also have non-trivial flux at 0.800, 1.200, 1.600 and 2.000 keV. In anticipation of this we planned our testing such that for each desired energy between 0.400 and 1.000 keV we also measured the flux at all the higher order energies. Fortunately, these energies often coincided with those at which we had planned to sample due the edge structure.
In the end our goal is to produce a three dimensional (two spatial and one of energy) response function. To that end, 5 fiducial energies were sampled at over 30 locations on the detector. All told over 800 measurements of the effective area were performed on the HRC detector alone. By combining these off-axis measurements with Sub-Assembly ``flat field" results and the HRC/HRMA on-axis data mentioned above, we can produce our effective area ``cube".
As the analysis of calibration data proceeds, we will continue to refine our results by including higher order effects such as the incidence angle of the incoming beam relative to the pore pitch.
The second critical component of the calibration process which could only be performed at the XRCF was the measurement of the count rate linearity of the system. This measures how quickly a very localized area of the MCP refreshes after detecting a photon. By sampling at a wide range of fluxes we determined the level at which the response begins to turn over, i.e. we are pouring in photons faster than the detector can recover. An analogue in optical CCD observing is the level at which the CCD begins to saturate.
The XRCF facility was critical to this process because it combined an extremely small image spot size with a high flux, something that would not be possible in the lab where the simultaneous collecting and focusing performed by the HRMA would not be present. Measurements were performed at 0.277 and 2.560 keV, at 7 flux levels ranging from 1 to 50 counts per second in the focal plane. This type of information will be crucial for observers in planning what measurements of bright sources will be feasible with the HRC, as well as interpreting results from serendipitous sources in the field.
The final primary goal of the HRC phase of the testing at the XRCF was the generation of a preliminary map of the PSF. Although the 1-g loading of the mirror and the finite focus distance does distort the image in ways that will not occur on orbit, these measurements do provide a useful cross-check with the results from the earlier Phase I (HRMA only) testing that occured at the XRCF.
Although, relatively few tests were performed explicitly to measure the PSF - just over 70- many of the effective area measurements at large off-axis angles will be incorporated into the analysis. This is possible because during the planning stages we realized that the expected FWHM of the images at large off-axis angles (>8 arcmin) would allow the effective area tests at these angles to be performed in-focus, while still allowing a high flux. For the on-axis PSF tests, the FAM was used to "dither" the image to prevent excessive exposure of any region of the MCP.
Similar to the effective area measurement strategy, we sampled the PSF at many locations with a set of fiducial energies. These will then be combined with the HRMA results to produce the final PSF map.
Secondary goals at the XRCF included a set of measurements to verify other performance characteristics of the HRC. These included measuring the spatial linearity of the system by moving the detector to 140 discrete locations in the focal plane while holding the mirror fixed. The dwell locations measured from the data are then compared with those predicted from the motor encoder values. This measurement confirms not only the spatial linearity of the detector but also the degapping and de-dithering algorithms both of which will be critical for processing of flight data.
Also tested at the XRCF were our ability to perform a ``flight focus" using the on-board HRC shutters, as well as mapping out the final architecture (seams between MCP's and location of the HESF) of HRC-S detector, and providing supporting data for the analysis of the angular response of the MCP pores.
Our efforts calibrating the HRC have been towards producing a small number of very detailed products for the user (effective area ``cube", PSF map and limiting count rate). Our aim is twofold. On the one hand, we hope to make the task of determining the feasibility of an observing program straightforward and requiring only a small amount of data. On the other, we wish to provide easy and direct access to the wealth of information gleaned from our testing and calibrating efforts.
R. Hank Donnelly