Introduction Optical calibration of the DAWN framing cameras G. Abraham,G. Kovacs, B. Nagy Department of Mechatronics, Optics and Engineering Informatics Budapest University of Technology and Economics NASA has initiated the DAWN mission in the beginning of the new millennium. The spacecraft shall be the first in the history of space research to visit two planetary bodies in one mission. The asteroid Vesta and the dwarf planet Ceres are both located in the asteroid belt between Mars and Jupiter. DAWN s mission is to orbit these large rocks in space and gather as many data of them as possible. For this reason several instruments are placed upon the spacecraft such as framing cameras, gamma ray and neutron spectrometer and visible/infrared (VIR) spectrometer. Fig.1. The DAWN spacecraft The DAWN spacecraft has two redundant Framing Cameras (FC) on board for scientific imaging and navigation purposes. These functional eyes of the spacecraft will support spacecraft maneuvers, orbit insertion and maintenance at the target asteroids. For in-flight calibration, star fields, clusters, and solar system objects will be imaged. Fig.2 The 3D modell of the DAWN cameras
To achieve high-resolution images the FC s have a four lens telecentric optical design with filters for spectral band imaging (polychromatic clear filter) and seven filters from the blue wavelengths up to the near infrared band and a CCD detector (Fig.2.) The flight units of Framing Cameras were calibrated at the Max Planck Institute for Solar System Research (MPS) in Lindau, Germany. The calibration campaign and image data analysis was supported by optical engineers from the Budapest University of Technology and Economics, Hungary. Methods The optical calibration of the FC focused on the imaging properties of the integrated device, including the lens system, baffle, filters, the CCD and the electronics. Several combinations of CCD exposures and spectral bands were tested to gain as much information as possible on the resolution, contrast, homogeneity and irradiation properties. The measurements were carried out in four main categories. Fig.3. DAWN framing camera in the vacuum chamber PSF: the actual focal position and the image resolution was determined by the Point Spread Function (PSF) measurement. MTF: The Modulation Transfer Function (MTF) measurement shows the contrast ratio of the FC. Different spatial resolutions on a USAF target were imaged by the optical system up to the CCD s Nyquist frequency. The contrast values at the various spatial frequencies provided the modulation transfer function which had to be in line with the optical specifications. RAD: The cameras spectral response and the whole system s quantum efficiency was determined by the RAD tests. The exit slit of a high resolution monochromator was imaged by a collimator and the camera to the CCD.
FLAT: An integrating sphere was used to test the light energy distribution over the detectors surface homogeneously. PSF The goal of this test was to determine the PSF (point spread function) of the FC for all filters at certain specified field angles. The measurements were carried out by simulating different focal positions, so we could establish the best focus position of the camera for all filters. This was critical, because of the wide spectral range, and the method of chromatic correction, involving different filter thicknesses. (Fig.4.) Fig.4. Modelling the PSF A target with a 0.035 mm pinhole was positioned at the focus of a reflective collimator (F=1750mm) providing a 11.6X reduction on the image plane (Fig.5.) The target war translated along the optical axis, and several images were shot to find the minimum on- and off-axis spot. The measurement used sub-pixel camera movements to cope with the CCD aliasing effects. Rotator Dawn FC Collimator Pinhole Condenser Halogen lamp Fig.5. PSF setup
MTF For the evaluation of the in-focus image quality, the MTF test provides the best data showing the image contrast at various resolutions. The two limiting factors for the MTF are the imaging optical system, and the CCD device. These two parameters should be in coherence to achieve the sharpest pictures. In our test the Dawn FC optical axis was aligned to the optical axis of the collimator. The USAF 1951 target (Fig.6.) was mounted in the target holder of the collimator. A halogen source with a diffuser was used to illuminate the target. The target was imaged to the CCD array of the FC. The tip-tilt table inside the vacuum chamber provided the pointing for the various field positions. The translators of the collimator target holder were used to simulate the defocusing effects for the camera. The collimator target translators were also used for movements to avoid the aliasing effects on the CCD. Fig.6. MTF target and results The measurement results (Fig.6.) showed close match with the calculations based on the design, approx. 40% contrast at the Nyquist frequency at all bandpass channels but the blue. The test was also performed at various defocus positions, as cross-check of the PSF results. RAD The RAD measurements determined the absolute spectral response of the DAWN Framing Camera for all channels. This data is essential for accurate photometric evaluation of the collected images. The test setup was the following: a diffuser target was positioned at the focal plane of the collimator. The energy of a halogen light source was collected at the entrance slit of the monochromator, and the selected wavelength band - at the exit slit of the monochromator - was focused to the diffuser target with a collector lens. The FC was aligned to the optical axis of the collimator, and the image of the monochromator exit slit (on the diffuser) was re-imaged to the FC CCD detector. (Fig.7.)
Fig.7. RAD measurement and results (Quantum efficiency) The output energy of the monochromator was measured by a PTB calibrated photodiode at several places in the optical path. The photodiode was positioned between the monochromator and the diffuser. Diode readouts are recorded at every 2 nm. Images of the FC were recorded at different wavelength positions of the monochromator, and at all filter positions. This way we collected a dataset along the full spectrum, including the transmission and blocking ranges and calculated the quantum efficiency of the CCD. FLAT The purpose of the flat field measurements was to determine the field uniformity across the active area of the CCD, and to separate high and low spatial frequency non-uniformities for all spectral channels. It was also important to find out the source of the non-uniformities (filt er transmission, particles on filters / CCD, pixel gain variation etc.). The results of this measurement serve as the basis of the image reduction pipeline. Fig.8. FLAT measurement with Ulbricht s sphere and results For the generation of the uniform illumination, a large integrating sphere was used (Fig.8.) This sphere was used previously for the calibration of the
Rosetta/Osiris camera. The integrating sphere was positioned close to the vacuum chamber window. The FC and the integrating sphere were aligned to the same optical axis, FC was looking at the back surface of the sphere. The integrating sphere was illuminated by halogen light sources. During the measurements, the light intensity was recorded by calibration diodes in the UV, VIS and NIR range. More than 500 images were recorded at different filter positions, and at different exposure times. Additional images were recorded for the check of absence of vignetting and to determine the in- and out-of-field stray light performance. At the end of the calibration sequence the integration sphere illumination was cross checked using a calibrated spectroradiometer. Conclusions The optical calibration of the FCs demonstrated that the design requirements are fulfilled by the cameras performance. The measurement data analysis showed that the focal position is well defined, the contrast ratios are as required. The FLAT and illumination energy measurements provide the necessary data for the image data reduction pipeline.