Cutaway schematic of SALT, showing the key elements of the segmented primary mirror, which can rotate in azimuth, but is at a fixed elevation of 53 ◦ , and the Tracker (which 

Context in source publication

Context 1

... with an optical equivalent of the Arecibo radio telescope, in which the segmented spherical primary mirror is held at a fixed elevation (37 ◦ ZD for SALT so that both Magellanic Clouds are accessible) but with selectable azimuth. Nevertheless, the primary is stationary during observing, and the target field is recorded via instrumentation on a Tracker unit that moves across the primary’s wide focal surface ( figure 1). However, by the time the SALT construction team formed, it was clear that there were a number of design enhancements required to the HET original. Principal among these was the need to improve the performance of the spherical aberration corrector (SAC), that is central to the optical design. O’Donoghue (2000) produced a radically better SAC design than that of HET, simultaneously delivering much improved image quality, access to a wider field of view and larger effective collecting area (as a result of the increased pupil size). The SALT SAC also gave a much larger back-focal distance and, combined with the use of carbon composites in the moving mass of the Tracker, allowed for an increased mass budget and easier access. Furthermore, the presence of the SAC’s additional mirrors in the optical path requires the use of advanced multi-layer coatings to give the improved UV performance (down to 320nm) that is an unusual feature amongst very large telescopes. Finally, the building design is substantially different to that of HET, with much more aggressive heat control (a cooled glycol system operates throughout) and louvres (for natural ventilation while observing) to deliver an image quality of 0.7 arcsecs (FWHM). 3.1. Imaging Camera, SALTICAM The first light (and principle telescope commissioning) instrument has been SALT’s imaging camera, SALTICAM (O’Donoghue et al. (2003), Buckley et al. (2006)). Based on a pair of mosaiced E2V 4096x2048x15 μ frame-transfer CCDs, SALTICAM acts as both an efficient acquisition camera (it covers SALT’s 8 arcmin diameter field of view) and science instrument. Equipped with a range of broad and narrow band filters, SALTICAM’s CCDs have high throughput down to ∼ 320nm. When operated in frame-transfer mode SALTICAM can provide 2.5s time-resolved images, or even faster (approx 60ms) when a special 20 arcsec wide occulting mask is introduced (“slot” mode). To date, this has been applied to observations of relatively bright (V < 20) targets, as SALTICAM’s autoguider has not yet been installed (and hence precludes deep imaging exposures). 3.2. Imaging Spectrograph, RSS The principal spectroscopic instrument on SALT is RSS (the Robert Stobie Spectrograph, named after the former SAAO Director, one of SALT’s original driving forces. Formerly known as PFIS (the Prime Focus and Imaging Spectrograph), this low to intermediate dispersion spectrograph was designed and constructed by the University of Wisconsin (UW, Nordsieck et al. (2003)), together with Rutgers University (mechanical structure and Fabry-Perot optical subsystem) and SAAO (CCD detectors). Intended as a true workhorse instrument, RSS has been designed to offer a wide range of modes that exploit SALT’s very good UV performance. These include long and multi-slit capabilities, plus tunable Fabry-Perot imaging and spectropolarimetry (table 1). RSS was installed on SALT in late 2005, and most of its observing modes have been exercised. The poor image quality of the telescope precluded comprehensive testing of its wide field capabilities, but the major problem exhibited by RSS was in its poor UV/blue throughput. Eventually traced to an optical flaw (degraded lens coupling fluid in the multiplets), RSS was removed from SALT in late 2006 and its optics are currently under repair (see Buckley et al. (2008) for more details). It is expected that a substantially improved RSS will be restored onto SALT in early 2009. Väisänen et al (these proceedings) present an example of RSS science that was accomplished during its initial commissioning. The original mechanical and optical design of RSS allowed for its extension into the near-IR band. This was to be accomplished with a dichroic which allows light beyond 900nm to enter the NIR arm. At present a folding flat is in place, as the NIR arm was not funded as part of SALT’s suite of first generation instruments. However, the design of RSS/NIR is now underway at UW, with the aim of extending RSS’s wavelength range to ∼ 1.7 μ , the precise value being a function of the extent of cooling that can be achieved. The intention is to offer a similar range of operating modes as in the optical, and to complete the RSS/NIR arm by 2011. 3.3. Fibre-fed High Resolution Spectrograph, HRS The first-generation suite is completed by a high resolution spectrograph, HRS. In order to reach resolving powers of up to 65,000 (as is needed for precision radial velocity work), the spectrograph is housed in a vacuum tank to provide extreme stability against temperature and pressure variations. Such a scale dictates that HRS be located in the 40 Charles, Buckley & O’Donoghue spectrograph room underneath the main telescope observing floor, where fibres from the Tracker feed light from the single target plus sky region. The dual-beam design which is now under construction (at Durham University’s Centre for Astronomical Instrumentation) uses an R4 echelle, after which a dichroic splits the spectrum into blue and red arms, each of which has its own VPH cross-disperser and camera giving a range of resolving powers (from 16,000 to 65,000, depending on the use of image slicers; see table 1). HRS is expected to enter commissioning in early 2010. It was only in late 2005 that simultaneous imaging over the 8 arcmin field of view was obtained, leading to the realization that there was a field-dependent image quality (IQ) problem. This manifests itself as a focus gradient, plus time-dependent effects that appear associated with the instrument rotator angle and temperature. A detailed IQ study through much of 2006 showed that the source of this problem lay not with the instruments (SALTICAM, RSS) or the primary mirror array, but with the SAC. A full report on this can be found in O’Donoghue et al. (2008), who established that (i) the last pair of mirrors in the SAC are mis-aligned with respect to the optical axis of the telescope, and (ii) there are significant mechanical stresses transmitted into the SAC via the Tracker interface due to thermal effects and instrument rotation. Since high quality (0.85 arcsec) images have been obtained within SALT images, there is no reason to doubt the optical quality of the individual SAC mirrors, nor the overall SAC optical design. A redesign of the SAC-Tracker interface is currently underway, with the aim to install the new interface and realign the SAC mirrors in early 2009. Despite the extended commissioning period of SALT, the total project timescale actu- ally compares favourably with comparably sized projects, and, as has been demonstrated in related presentations at MEARIM, limited science operations have already begun. Furthermore, the original HET design paradigm is in the process of being taken to a new level of performance as a result of the modifications that were made for SALT. Indeed, an extended version of the SALT SAC design is at the heart of a new instrument currently under development for HET. And the last year has seen additional partners join SALT (AMNH, the American Museum of Natural History in New York, and IUCAA, the Inter-University Centre for Astronomy and Astrophysics in Pune, India) bringing new investment into the SALT Foundation, and demonstrating confidence in the direction of the project. All of the SALT partners recognise the enormous importance of promoting and stimulating science education in South Africa (see Whitelock, these proceedings), and SALT provides a superb icon for such ...