http://en.wikipedia.org/wiki/Very_Large_Telescope#Interferometry_and_the_VLTI wrote:
<<Although the four 8.2-metre Unit Telescopes can be combined in the Very Large Telescope Interferometer (VLTI), they are mostly used for individual observations and are only available for interferometric observations for a limited number of nights every year. But the four smaller 1.8-metre ATs are available and dedicated to interferometry to allow the VLTI to operate every night.
The top part of each AT is a round enclosure, made from two sets of three segments, which open and close. Its job is to protect the delicate 1.8-metre telescope from the desert conditions. The enclosure is supported by the boxy transporter section, which also contains electronics cabinets, liquid cooling systems, air-conditioning units, power supplies, and more. During astronomical observations the enclosure and transporter are mechanically isolated from the telescope, to ensure that no vibrations compromise the data collected. The transporter section runs on tracks, so the ATs can be moved to 30 different observing locations.
Because the Unit Telescopes are used most of the time independently, they are used in the interferometric mode mostly during bright time (that is, close to Full Moon). At other times, interferometry is done using 1.8 meter Auxiliary Telescopes (ATs), which are dedicated to full-time interferometric measurements. The first observations using a pair of ATs were conducted in February 2005, and all the four ATs have now been commissioned. For interferometric observations on the brightest objects, there is little benefit in using 8 meter telescopes rather than 1.8 meter telescopes.
In its interferometric operating mode, the light from the telescopes is reflected off mirrors and directed through tunnels to a central beam combining laboratory. The VLTI is intended to achieve an effective angular resolution of 0.002 arcsecond at a wavelength of 2 µm. This is comparable to the resolution achieved using other arrays such as the Navy Prototype Optical Interferometer and the CHARA array. Unlike many earlier optical and infrared interferometers, the AMBER instrument on VLTI was initially designed to perform coherent integration (which requires signal-to-noise greater than one in each atmospheric coherence time). Using the big telescopes and coherent integration, the faintest object the VLTI can observe is magnitude 7 in the near infrared for broadband observations, similar to many other near infrared / optical interferometers without fringe tracking.
Because of the many mirrors involved in the optical train, about 95 percent of the light is lost before reaching the instruments at a wavelength of 1 µm, 90 percent at 2 µm and 75 percent at 10 µm. This refers to reflection off 32 surfaces including the Coudé train, the star separator, the main delay line, beam compressor and feeding optics. Additionally, the interferometric technique is such that it is very efficient only for objects that are small enough that all their light is concentrated. For instance, an object with a relatively low surface brightness such as the moon cannot be observed, because its light is too diluted. Only targets which are at temperatures of more than 1,000°C have a surface brightness high enough to be observed in the mid-infrared, and objects must be at several thousands of degrees Celsius for near-infrared observations using the VLTI. This includes most of the stars in the solar neighborhood and many extragalactic objects such as bright active galactic nuclei, but this sensitivity limit rules out interferometric observations of most solar-system objects.
The first two instruments at the VLTI were VINCI (a test instrument used to set-up the system, now decommissioned) and MIDI, which only allow two telescopes to be used at any one time. With the installation of the three-telescope AMBER closure-phase instrument in 2005, the first imaging observations from the VLTI are expected soon. Deployment of The Phase Referenced Imaging and Microarcsecond Astrometry (PRIMA) instrument started 2008 with the aim to allow phase-referenced measurements in either an astrometric two-beam mode or as a fringe-tracker successor to VINCI, operated concurrent with one of the other instruments. As of March 2008, the VLTI had already led to the publication of 89 peer-reviewed publications and had published a
first-ever image of the inner structure of the mysterious Eta Carinae.>>
3 ATs
