Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.stecf.org/conferences/sardinia/node7.html
Дата изменения: Mon Jun 12 19:04:27 2006 Дата индексирования: Mon Oct 1 20:50:43 2012 Кодировка: Поисковые слова: п п п п п п п п п п р п р п р п р п р п р п р п р п р п р п |
Peter Bizenberger, Rainer Lenzen, Heinrich Bellemann, Clemens Storz, Andreas Tusche, Karl Wagner Max-Planck-Institut für Astronomie Königstuhl 17, 69117 Heidelberg, Germany
Reiner Hofmann Max-Planck-Institut für Extraterrestrische Physik Gießenbachstr., 85748 Garching, Germany
VLT instrumentation, ALADDIN InSb detector, NIR imaging, NIR spectroscopy, re-imaging optics, IR optics
In order to provide an double-sampling over the wavelength band from 1 - 5 m, the imaging scale is designed to be variable. Three discrete image scales (13, 27 and 55 mas/pixel) sample the wavelengths of interest. These optics are designed to illuminate a 1024 x 1024 array with 27 m pixel pitch without vignetting. A fourth set of camera optics is available for a larger pixel scale of about 110 mas/pixel which leads to a field of view of about 73 arcsec in diameter. For each scale (except the slowest camera) the camera systems are split in two wavelength regions, one for 1.0 - 2.5 m and a second for 2.0 - 5.0 m. Thus, there are seven imaging systems in total, all of which can be used within the K-band (Figure 1). The complete imaging optics are realized as a re-imaging system. Inside the cryostat, just behind the plane-parallel CaF2 entrance window, lies the instrument focal plane. A number of field-limiting masks, slits, coronographic masks or special stripe masks for Wollaston applications can be inserted here. The beam is then collimated by an achromatic LiF/BaF2 doublet giving a well collimated beam and a pupil image all over the wavelength region. Residual chromatic errors are compensated by the individual camera systems. Even though the whole optical system is achromatic, the detector position in the back focal plane of the optics can be remotely adjusted during operation.
CONICA can take 40 filters (1 inch diameter), mounted close to the pupil position. A set of standard broad-band filters, as well as 15 individual narrow band filters are available. In addition, there is a set of blocking filters for the Fabry-Perot. All filters lie behind the pupil, so there is no shift of the pupil image due to different filter thickness. All filters are tilted by 6 degrees to avoid ghost images.
A cryogenic Fabry-Perot can be inserted into the collimated beam, giving spectroscopic imaging for all available imaging scales at a spectral resolution R 1800 within the K-band.
At present, three grisms of medium resolution R 350 - 2400 per pixel allow long-slit spectroscopy with diffraction limited spatial resolution. An additional grism is on order to cover the whole range of resolution for all wavelengths. Two of the grisms are direct ruled KRS5 for operation at longer wavelengths (up to 5 m). The remaining two grisms are replica grisms for the wavelength range from 1.0 - 2.5 m (Table 1)
For polarimetric applications two Wollaston prisms and four wire-grid analysers are mounted. Both types of analyzers are suitable for the whole wavelength range.
In front of the cooled camera, a tunable atmospheric dispersion compensator (TADC) at ambient temperature corrects the chromatic dispersion of the atmosphere for higher airmasses. Particularly for the J band a correction is essential, since the dispersion at an airmass of 2 is at the order of the diffraction limited FWHM at this wavelengths. For longer wavelengths, this effect is negligible, allowing the TADC to operate at ambient temperature.
All these analyzing components have been tested in an instrument very similar to CONICA. Since autumn 1997, an infrared camera called Omega Cass (Lenzen et al. 1998a) has been available at the 3.5 m telescope on Calar Alto in Spain. Omega Cass is equipped with the same capabilities as CONICA, including the adaptive optic system ALFA (Hippler et al. 1998). The only difference to CONICA is the smaller telescope and the reduced operating wavelength range of 1.0 - 2.5 m of the HgCdTe detector array.
Since CONICA will be sensitive up to the thermal NIR region of 5.0 m the whole imaging and analyzing optics are cooled below 80 K (except the TADC). The radiation shield and the internal cryogenic structure are thermally coupled to the first stage of a two-stage Gifford-McMahon closed cycle cooler. As the environmental temperature of the detector array can determine instrumental sensitivity (particularly for spectral applications), special care has been taken to reach temperatures below 80 K for all surfaces which are directly seen by the detector. Finite element analysis has been performed to check the expected temperature distributions.
The array itself is thermally coupled to the second stage of the closed cycle cooler. The operation temperature can be stabilized at any value between 30 K and 40 K. The optimum operating temperature for the InSb detector is expected to be 35 K. The typical cool down process has been tested and stabilized conditions can be reached within less than 24 hours. This relatively short time is achieved by floating a tube system surrounding the radiation shield with liquid nitrogen. CONICA uses a cold shutter for integration times shorter than the frame read time of about 50 ms and for seeing selection in case bad seeing disrupts the NAOS image quality during long integrations. For the shutter drive, a modified magnetic head drive of a hard disk unit is used. This system works at LN2 temperatures and allows beam switching within a few milliseconds and repeat rates of more than 10 Hz. It is driven by newly-developed electronics which are controlled by one command bit and produce current pulses to accelerate and decelerate the shutter arm. Figure 2 shows a schematic of the dewar and mounting flange. Figure 3 is an image of the first integration of the dewar and mounting flange with the head of our workshop as a scale.
Table 2 shows the calculated limiting magnitudes of point sources for 3 in one hour integration time, assuming a Strehl ratio of 70 % down to a wavelength of 2.0 m and a total efficiency of 65 % including telescope, optics, filter and detector. NAOS is calculated with 100 % throughput since no other numbers are known yet. The R = 50 row in the table represents narrow band imaging. R = 500 is an indicator for the detection limits in spectroscopy mode, but these numbers vary for other resolutions. The R = 1800 is for the Fabry-Perot mode which is only available in the K band.
We are grateful to the ESO staff members who have been involved in developing components for CONICA.