"First Light" in the NTT R
Total Page:16
File Type:pdf, Size:1020Kb
No. 56 - June 1989 "First Light" in the NTT R. WILSON, ESO On the night 22-23 March 1989, "First Light" was achieved in the NlT in the sense that the first direct images of astronomical objects were produced with the optics in an optimized state near the zenith position. The fact that the quality of these im- ages turned out to be the best ever ' recorded in ground-based astronomy certainly made it the most memorable event in the ESO career of the author and his colleagues whose work on the NTT had led to this remarkable success. But this result did not drop out of the sky! Many years of hard work by a small group of dedicated colleagues were necessary to achieve it. The principal new technology feature of the N7T was its Active Optics control system, the background and principles of which were described in The Messenger No. 53 (September 1988), page 1. The i other new technology feature of funda- mental importance for the optical quality was the design of the building. Although its basic concept is the same as that of the MMT (a building rotating with the alt- az mounted telescope), the new I features were crucial, particularly the opening of the telescope area on both sides to allow maximum natural ventila- tion. These two new technology features, together with the excellent work of Carl Zeiss in figuring the optics, were the key to the image quality achieved at First Light. The Basic ("Passive") Alignment of the Optics This false-colour photo shows one of the best CCD frames obtained with the ESO NTduring the night of "first light" on March 23, 1989. The fieldjs near the centre of the bright southern The final stage for the NTT globular cluster Omega Centauri and the frame size is -47 x 47 arcseconds. The measured started in November 1988. The diameter of the stellar images (at half of the maximum intensity) is only 0.33 arcseconds. mechanics of the telescope had been Exposure time: 10 seconds; North is up and East is to the left. See page 3 for more details. assembled under the supervision of - Optical identification of OPT, the line the mean of many measurements as Gerardo Ihle. To him and his colleagues defined as the optical axis of the tele- about 12,800 nm or 4.5 arcsec with zero in Chile we owe, also for much help scope and which is perpendicular to correction (central position), of the x-y later, a deep sense of gratitude. The ALT and traverses the intersection movement of M2. Using the calibration optical alignment procedure and the point of ALT and AZ. value given by Bernard Delabre from the handling of all the elements involved, - Reduction of the angular error be- optical design of the telescope, it could required the definition and procurement tween OPT and AZ and the intersec- be deduced that the MI cell had to be of a considerable number of mounting tion (coplanarity) error in the knot tilted by about 2 arcmin to correct this and handling devices. This was planned point of AZ relative to ALT. large coma. This is a measure of the and defined by Francis Franza and Paul - Adjustment of M3 bearing axis to be extreme sensitivity of passive tele- Giordano last autumn and manufac- coincident in angle and position with scopes to decentring coma, since this is tured with the help of Rainer Grote. OPT. a quite small angle and well within the Francis and Paul then spent 6 weeks - Mount M2, centre it to OPT, set it mechanical tolerances of the telescope. from the beginning of November, with- perpendicular to OPT and align the A maximum of 2,5 mm had to be out intermission, preparing this material focus movement to OPT. machined off the flexion bar feet holding for the following alignment mission. This - Remove M 1 dummy (counterweight), the M 1 cell in order to achieve this tilt. included the mounting of the prime mount M 1 cell (without M 1) and The biggest problem was to interpret mirror supports and their testing with centre M 1 cell to OPT. the physical orientation of the coma in Lothar Noethe. As an example of the - Insert M3, adjust its tilt in both planes the telescope: an error here would have sort of auxiliary material involved, one and place its reflecting surface on the meant the whole operation of lowering can cite the temporary "bridge" over M 3 intersection of ALT with OPT. the cell and machining the flexion bars which serves for its mounting but also - Mount M 1, adjust tilt of the M 1 cell to would have to be repeated! It was there- as support for one of the sighting tele- eliminate decentring coma on a fore with a feeling of great relief that we scopes used in the alignment. natural star ("Technical First Light"). saw the confirmation of the orientation Then followed a second mission from The last step above is the final one in when the first image analysis showed 10 January, also of 6 weeks without any the passive alignment. In fact, it could that the coma had been reduced to less intermission, for the basic (passive) not be completed at that stage as the than one tenth of its original value. alignment of the optics of the NTT, image analyser (our off-line test system This formally completed the passive carried out by Francis Franza, Paul Gior- called "ANTARES") was not yet avail- adjustment of the telescope. It remained dano and Ray Wilson. The procedure for able at that time with a CCD detector. At to optimize its performance in the zenith this alignment was laid down by the that stage, therefore, no attempt was by the dc active optics correction, i.e. above on the basis of a standard proce- made to adjust the M 1 cell but about ?h the correction of residual fixed errors. dure used by them on a number of La of the considerable coma (which was Silla telescopes, notably the 3.6 m, the normal and expected at this stage with- The Night of "'First Light" Danish 1.54 m, the CAT and, in its final out adjustment of the MI cell) was stages, for the MPlA 2.2 m. The NTT corrected by a visual inspection of the Apart from one effect, the image anal- alignment procedure was a more refined defocused image and by using the full ysis had brought no surprises. This one and complete form, taking account of its range available of the M2 x-y movement effect was a residual of axi-symmetric alt-az mount and double Nasmyth for fine coma correction. "Technical (spherical) aberration which was some focus. If one considers certain inevitable First Light" was celebrated on 15 Febru- five times larger than the maximum we problems of control of the telescope, ary with an image on TV of Jupiter, with had expected. Variations of this value this alignment programme, comprising modest seeing and at a zenith distance could be explained by a hot-air bubble 56 individual technical steps, rep- of about 50". effect on the primary, but a fixed re- resented the absolute maximum achiev- sidual of about 0.5 arcsec remained. able in 6 weeks of continuous work. In The Final Effort: Trimming the This exceeded the passive specifica- fact, without the absolutely excellent NTT Optics to the Limit tion. Because its correction absorbed a support from the La Silla colleagues in much larger part of our total active dy- electronics, mechanics and optics, it The final stage of the setting up and namic range than we expected, it would not have been possible in the optimization of the optics near the proved most convenient to correct us- time. In particular, the continuous zenith was accomplished by the above ing the "natural mode", a system of superb support of the mechanical work- team and Lothar Noethe in a 16-day modal formulation suggested by Noethe shop must be emphasized as playing a mission ending on 23 March. Lothar and Schneermann for the VLT. In this key role in the completion of the heavy played the central role as the person way, this aberration as well as the other work load. This general procedure in- responsible for the entire software and fairly small residuals were all reduced to cluded the aluminization of the three image analysis side, following up the values within one tenth of an arcsec, mirrors, carried out under the supervi- work on the 1 m experiment and the including the residue of coma. This re- sion of Paul Giordano who has many acceptance of the optics at the quired the averaging of ten or more years of experience in this field. Of manufacturers, Carl Zeiss. measurements. We were helped by the course, the biggest task here was the The first essential step was to get the fact that, during the nights leading up to primary aluminization, above all be- image analysis working. This was rapid- "First Light" on 22-23 March, the exter- cause of the delicate handling proce- ly achieved with no significant prob- nal seeing was extremely good. dure involved. lems. Gaetano Andreoni of La Silla be- On 20 March we could already The individual steps of the alignment came an invaluable member of our team announce that the "INTRINSIC QUALI- procedure can be grouped in the follow- and worked with us the whole time. TY" of the telescope optics (80 % of the ing work packages: The image analysis confirmed im- light of a star concentrated in about 0.15 - Optical identification of the ALT-axis mediately the amount of decentring arcsec) had already been reached.