|
SOME PROPERTIES
of the
NORDIC OPTICAL TELESCOPE
Arne Ardeberg Lund Observatory Nordic Optical Telescope Scientific Association
Contents:
ABSTRACT
A short description is given of the Nordic Optical Telescope and some of its most important features. Heaviest emphasis is given to properties being of high priority. These properties are intended to define a low cost compact telescope in a small enclosure. With a moderately large aperture, highest weight has been given to design features providing optimum image quality. Further, the telescope has only a Cassegrain focus station, which should minimize instrumental polarization and favour image quality and speed. High interest in observations in the infrared wavelength range, also in day time, emphasizes the need for high pointing and tracking accuracy. The telescope has an altazimuth mounting, a fast, f/2.0, primary mirror of Ritchey-Chrétien type, made of Zerodur, with a diameter of 2.56 metres and an aspect ratio of 13.5. The effective f ratio of the telescope is 11.0 and the unvignetted field of view 25 arc minutes. The image quality specification is for the combined optical system 80 per cent geometrical energy within 0.3 arc seconds in passive mode. For the integrated optomechanical system, the corresponding specification is 80 per cent geometrical energy within 0.4 arc seconds for zenith distances smaller than 60 degrees. A focal reducer (to be installed) gives an effective telescope f ratio of 5.0 over a field of ten arc minutes. An integrated servo system under complete computer control manages rotation of the telescope with respect to azimuth and altitude axes, rotation of instrument adapter, pneumatic support system for the primary mirror, focusing movements and lateral displacements plus tilting (if necessary) of secondary mirror, autoguider corrections, rotation of building and data handling for implementation of refraction correction. In order to improve optical quality, wobbling has not been included for the secondary mirror. Instead, focalplane chopping has been preferred. For ancillary instrumentation mechanically attached to the Cassegrain focus, the maximum free distance behind the adapter is 1500 millimetres and thecorresponding maximum weight is 250 kg. More and heavier instruments can be mounted elsewhere and have optical fibre feeding from the instrument adapter. Over the available sky, pointing and tracking accuracies have been set at two arc seconds and one tenth of an arc second, RMS, respectively. The accuracy of both pointing and tracking should be unaffected by wind with speeds up to around 22 m/s. However, safety reasons dictate closing of dome already for lower wind speeds, depending on circumstances. Instructions to the observer are communicated via the control system. A field rotator is integrated in the instrument adapter. The zenith singularity has a size of 30 arc minutes times 20 arc minutes. For an object on the optical axis, dome vignetting affects a field of 1.3degrees extension around the zenith singularity. The lowest resonance frequencies for the azimuth and altitude axes are 7.4 and 14 Hz, respectively. The sky available, limited by safety software in altitude, extends to -56 degrees in declination. Decided measures have been taken to minimize effects of telescope turbulence. This includes greatest restrictivity with heat sources but also cooling of ambient air. Limitations with respect to telescope operation at high wind speed and in high relative humidity are indicated by the control system, advising the observer in these respects.
Key words: Telescope - Mounting - Mechanical Structure - Optics - Image Quality.
Introduction
In 1983 and 1984 agreement was reached in the four Nordic countries Denmark, Finland, Norway and Sweden concerningconstruction of a major observing facility for Nordic astronomers, the Nordic Optical Telescope (NOT). It was intendedto cover observing needs in the Northern Hemisphere for astronomers from the four participating countries and be of high quality for both the optical and the infrared wavelength ranges. The telescope was to be erected in a place offeringoptimum conditions. Already before funding was decided, Nordic astronomers had agreed that the obvious choice for telescope site was at La Palma, Canary Islands.
The definition of major design features for the NOT was not an entirely easy matter. The main reason for this was that it should, as well as possible, satisfy the needs of research groups, working on programmes and with observational methods of a very diversified nature.
The wish to reach faint limiting magnitudes and high spectral resolution called for a large collecting surface. This was not controversial among astronomers but had to be matched against existing budgets. It was unanimously decided to push for high observing power through a design, featuring compactness.
The imaging capability of the telescope was given special attention. It was quickly decided to choose image quality as first scientific priority. Obviously, image quality is of prime importance not only for imaging in itself but also improves very effectively the resulting observing power of the telescope for most purposes. Evidently, giving first priority to image quality implied high emphasis on site, optics and mechanical structure as well as on design of enclosure.
Scientific interests in high accuracy polarimetric work meant that effects of instrumental polarization had to be kept as small as possible. This, in turn, calledfor an optical solution with a minimum of elements and a maximum of optical symmetry (Andersen, 1981, 1982).
Already early, it was agreed that the telescope should be of high quality notonly for observations in the optical wavelength region but also for work in the infrared spectral range. This implied that it had to be infrared-clean and have high pointing and tracking accuracy.
In many ways, image quality of a telescope is, from a scientific point of view, much more interesting than the size of the collecting surface. It is often pointed out that decrease of prevailing image size by a certain factor gives a result for observing power equivalent to that of an increase of the size of the collecting surface by the same factor. In practice, this is only part of the truth.
For observations of stellar objects free from effects of crowding there is equivalence between decrease of image size and increase of collecting surface regarding limiting magnitude and/or observing speed. However, as soon aseffects of crowding have to be taken into account, decrease of image size is much more advantageous than increase of collecting surface. In many cases, an increase of the collecting surface does not improve conditions at all, whereas decrease of image size may turn out crucial. Examples are crowded stellar clusters, aggregates and multiple stars and also galaxies for which detailed structure is the target. Stating this in a slightly different way, it may be said that for many observational problems, a somewhat smaller telescope can, with increased observing time, obtain the same results as a large telescope, provided that observing conditions are stable. On the other side,it is easy to find many examples of observing problems for which increase of collecting surfaces does not imply any improvement.
In conclusion, it can be stated, that high image quality is always a great asset and often a necessary prerequisite for our possibility to solve observing problems. Moreover, concerning increase of observing power, decrease of image size is an imcomparably more economic solution than increase of collecting surface. As a result, for the NOT, image quality was entitled to the preference.
For all types of optical observations, the choice of telescope site is a matter of utmost consequences. For far too many potentially powerful telescopes, output flow and quality are, in practice, limited by doubtful quality of sites rather than by limiting instrumental factors. Obviously, the more sophisticated the telescope, the more critical becomes the choice of site. In the case of the NOT, the site problem was given ample attention. Taking into account all major parameters defining site quality for observations in the optical and infrared wavelength regions and, at thesame time, assigning high weight to image quality, it was decided to locate the telescope on the premises of the Observatorio del Roque de los Muchachos. The observatory grounds cover a generous area (Sanchez, 1985), coinciding with the highest part of the Caldera de Taburiente (a crater of very large dimensions and with an origin still open to dispute), which, in turn, completely dominates the island of La Palma. Within the observatory area, differential site testing (Ardeberg and Andersen, 1990) was made of places selected on account of local topographic features. As a result of this testing and negotiations with other groups belonging to the Comité Cientifico Internacional (CCI), it was decided to erect the NOT at Cruz del Fraile, a mountain ridge defining approximately a North-South line.At its South end linked to the summit of the Caldera de Taburiente, Roque de los Muchachos, the North end of Cruz del Fraile is at a distance of around 350 metres from the summit, nearly exactly in the prevailing upwind direction.
Located at 2382 metres above sea level, the NOT site is the highest of the observatory. Whereas the other telescopes define a loose cluster close to Fuente Nueva, the NOT is several hundred metres away, accompanied only by its own service building and infrastructure.
More details on the site of the NOT can be found in a separate description.
The complete installation for the NOT comprises the telescope in its building, a service building and infrastructural elements (Ardeberg, 1985, 1987). Items belonging to NOT infrastructure are a lifting device for the primary mirror, a non-break-power installation, a transformer and a heat exchanger. Finally, an access road is included with space permitting proper access to the telescope building for vehicles, including heavy trucks. In order to secure optimum image quality, the telescope is installed close to the North extreme of the mountain ridge Cruz del Fraile. With this position, prevailing winds should result in a maximum of undisturbed laminar flow. At the same time, with the wind from theleast favourable direction, from the Roque de los Muchachos summit, maximum turbulence relaxation should be due, as a result both of distance from the summit and of local topography. Both effects were verified during site testing.
The service building has been erected about 85 metres South of the telescope building, close to prevailing downwind direction. It is located somewhat lower than the telescope building, in a natural cavity in the Cruz del Fraile ridge.
Normally, and always during observation, the mirror lift is kept in collapsed position close to the service building. Also the non-break-power device, thetransformer and the heat exchanger are placed close to the service building. The access road links telescope and service buildings to the main observatory road close to the Roque de los Muchachos summit.For more details regarding telescope building and infrastructure, reference is made to special descriptions of these items.
The telescope has an altazimuth mounting. Clearly, part of the reason for this choice is of economical nature, as this type of mounting, with its compactness, reduces both weight and cost of the telescope proper, at the same time as it allows for a rather small and, thereby, low cost enclosure. However, altazimuth mounting also has some further, rather important consequences regarding image quality, pointing and tracking. Finally, as compared to the more conventional equatorial type of mounting, altazimuth mounting implies some restrictions concerning field rotation and zenith singularity.Regarding the consequences of telescope compactness, reference is made to sections describing mechanical structure, optics, mirror cells and wind sensitivity. Further reference is made to sections dealing with enclosure, image quality, pointing, tracking, field rotation and zenith singularity.
The NOT has a welded structure of steel with a total height in zenith position of around 9.4 metres. The corresponding height above the upper fork base, or above the observing floor, is approximately 7.2 metres. The total weight of the telescope structure is about 44 tons. The moving part of the telescope has a total weight close to 35 tons.
The largest and heaviest single structure of the telescope is the fork. It is, in total and including the cover for the altitude gear, around 5.6 metres high with 4.1 metres above the upper fork base. The total width is 6.1 metres in the direction of the altitude axis and 4.6 metres in the direction perpendicular to the altitude axis. The centre of the altitude axis is 2.9 metres above the upper fork base. The total weight of the fork, excluding bearings, gears and drives, is somewhat above 17 tons.
The second most prominent telescope structure is the tube. Excluding theinstrument adapter, but including the primary mirror cell and the top unit, it has a total length of around 5.5 metres. The distance between the surfaces of the primary and secondary mirrors is 4.2 metres. The centre section has a width of 3.4 metres and the top ring a diameter of 3.0 metres. The total weight of the telescope tube, excluding optical elements and instrument adapter, is somewhat below seven tons.
Mirror blanks are made of Zerodur by Schott Glaswerke in Mainz, Federal Republic of Germany. Zerodur is a glass ceramics material of highest homogeneity and longterm dimensional stability. It has an exceedingly low thermal expansion coefficient, being around 1/10000000 1/K, or even lower. For practical purposes, this is fully negligible as long as reasonable precautions are taken concerning temperature conservation of and around mirrors.
For the mirror blanks, strict requirements were defined concerning birefringence, striae and presence of bubbles and inclusions. These requirements were met with ample safety margins. For the primary mirror blank, the stress birefringence resulting from permanent stresses was found to be lower than or equal to 10 nanometres per centimetre and the stress birefringence caused by streaks of striae was below detection level. The numbers and sizes of bubbles and inclusions were far belowrequirements inside the critical zone and well below requirements outside it. The average linear coefficient of thermal expansion was found to be below detection limit, or less than 1/100000000 1/K, for temperatures between 0 and 50 degrees C. For the secondary mirror blank, stress birefringence resulting from permanent stress was smaller than 4 nanometres per centimetre and stress birefringence caused by streaks of striae was, as for the primary mirror blank, below detection level. Bubbles and inclusions were not significant and the average linear coefficient of thermal expansion was found to be -6/100000000 1/K.
Optical figuring was made in the Optics Laboratory at Tuorla, outside Turku in Finland, an institute of the Turku University. The laboratory is blasted into the rock, about ten metres below ground. This ensures thermal and mechanical stability. A vertical shaft above the laboratory, ending in its upper part in a 1 m telescope, provides a tower of 25metres height and excellent possibilities for test measurements of the mirror under figuring, maintaining its working position. Thus, the programme mirror can remain on its supports throughout the figuring procedure. This increases accuracy and decreases figuring time. The shaft has been thermally insulated. The thermal stability of the laboratory is better than 0.5 degree C, also over very longperiods, and the integrated turbulence over the optical test path is of the order of 0.1 arc seconds. For further details regarding the Optics Laboratory at Tuorla, reference is made to Korhonen et al. (1985). Manufacturing of optics for the Nordic Optical Telescope is described in Korhonen et al. (1985) and in Haarala et al. (1988). The primary mirror has a diameter of 2.56 metres and an aspect ratio of 13.5. This means that it is a mirror of the new generation, of low weight and necessitating a sophisticated support system in order not to deform significantly. However, it is still rigid enough not to need active optics control to maintain its correct surface figure. The weight of the primary mirror is 1925 kg. The surface of the primary mirror is of hyperbolic Ritchey-Chrétien type, and its f ratio is 2.0. The choice of a surface of Ritchey-Chrétien type was made to ensure a large field of highest optical quality, free from spherical aberration and coma. The central hole in the primary mirror has a diameter of 340 millimetres. The primary mirror is faster and thinner than those of all other telescopes of similar size so far constructed (Korhonen et al., 1985; Korhonen, 1987).
The secondary mirror has a diameter of 0.51 metres and an aspect ratio of 6.3. The axial radius of curvature is -2287 millimetres and the surface is hyperbolic. The weight of the secondary mirror is 35kg. These parameters define a somewhat conventional type of secondary mirror. The solution is perfect for the NOT, especially taking into account the decision to replace, for observations in the infrared wavelength region, the function of a wobbling secondary mirror with that of a focal plane chopper. Further comments regarding wobbling secondary mirror versus focal plane chopping can be found below under "Top End". The equivalent focal length of the telescope is 28160 millimetres. The combined optical system gives, at the Cassegrain focus station, an effective f ratio of 11.0. This corresponds to an image scale of 0.13 millimetres per arc second. The unvignetted field of view is 25 arc minutes in diameter. The total field of view is 30 arc minutes in diameter with a maximum relative vignetting effect of 2.5%.
Specifications on the optics are severe. The combined optical system should, in passive mode, give 80 per cent geometrical energy within 0.3 arc seconds. In practice, the result was that the primary mirror alone had to deliver 80 per cent geometrical energy within lessthan 0.25 arc seconds in passive mode. This, in turn, meant that it had to be figured over the complete surface to an accuracy better than about 30 nanometres RMS.
The Ritchey-Chrétien optical system gives freedom from spherical aberration and coma. The aberrations remaining off the optical axis are field curvature and marginal astigmatism. The radius of the field curvature is -963 millimetres. For this reason, without correction, optimal image quality will be obtained over a limited field only. In practice, image spread due to field curvature will be below a tenth of an arc second out to a (radial) distance from the optical axis of somewhat more than 2.5 arc minutes. This corresponds to a field of more than 40 millimetres in diameter, which is larger than the surface covered by all linear detectors presently available. Out to a (radial) distance from the optical axis close to 5 arc minutes, image spread due to field curvature is still not more than three tenths of an arc second. As a result, effects of field curvature may, for most purposes, be regarded as insignificant forfields up to 10 arc minutes or around 80 millimetres in diameter. For larger fields, image spread increases rapidly as a function of distance from the optical axis and field correction becomes a necessity. For the zone limited by 25 and 30 arc minutes diameter, respectively, to be used for autoguiding, proper focusing gives an image quality corresponding to 80 % geometrical energy inside from 0.5 to 0.8 arc seconds. This is fully acceptable also for guiding of highest precision or with an rms error smaller than one tenth of an arc second.
For use in imaging of extended objects and other types of observation for which a telescope f ratio lower than the standard value 11.0 may be favourable, a focal reducer will be made available. It converts the effective telescope f ratio to 5.0 and covers a field of ten arc minutes.
With a 2.56 m primary mirror with an aspect ratio of 13.5, good image quality requires a primary mirror cell of considerable sophistication. In practice, the mirror cell has to ensure that deviations of the mirror surface always, also in difficult positions, are smaller than optical tolerances. This means that all deviations larger than around 30 nanometres have to be avoided (Andersen, 1986; Andersen and Jessen, 1985).
The solution chosen is a circular mirror cell with a box section and a pneumatic support system. The primary mirror is floated axially on air bellows. These bellows are symmetrically distributed along three concentric rings. Three supporting sections are included. Each section has 15 metal bellows and one fixed point with a load cell. The pressure of the supporting air bellows is adjusted to guarantee that they accurately take the weight of the mirror. The air pressure to be applied to the totally 45 bellows isdetermined from the forces measured for the fixed supports (Ardeberg and Andersen, 1988, 1990).
For each of the three supporting sections, the force on the corresponding load cell determines the air pressure to be applied to the 15 supporting bellows. This air pressure is adjusted to provide a resulting force on the load cell close to zero. For this, a closed loop control system is used. The supporting bellows are provided with compressed air from a compressor in the ground floor of the telescope building, connected to the bellows via a system of pipes and hoses. All pressure adjustment is done completely automatically and under full electronic control (Andersen, 1986).
Further, the primary mirror cell has to support the mirror in the transverse direction. The transverse supports have to take load only in a plane perpendicular to that of the altitude axis. The mirror cell has 20 transverse mirror supports,working in parallel. They are of push-pull type with lever arms and counterweights. The lever arms transmit forces to the edge of the mirror.
As compared to a more conventional type of mirror support system, based on use of counterweights, the choice of a pneumatic mirror support system offers a number of important advantages. The weight is considerably lower. In itself and also through its implications on the weight of other structural details, this translates into a lower cost. In addition, residual errors in the floating supports do not result in a cumulative force on the fixed supports.
The support system described has a basic bandwidth of between two and three Hz. At the same time, the resonance frequency of the fixed point suspension of the primary mirror is around eleven Hz. For the corresponding oscillation mode, an electromechanical active damping device has been installed.
Apart from the advantages described, all duly used to improve performance anddecrease cost of the telescope, there is an advantage still pending. A pneumatic support system of the type chosen, acting on a primary mirror with an aspect ratio as high as 13.5, is, potentially, a good basis for installation of an active optical system. In this context, it should be clearly stated, that the design of the telescope, including the cell for the primary mirror, has, in itself, been made with purely passive optics in mind. Calculations performed demonstrate that the NOT, in a passive mode, should reach the specifications adopted regarding image quality. At the same time, we have hope that we will reach an image quality regime on the extreme side for earthbased telescopes. In itself rather exciting, this calls for a certain preparation for a possible further development.
At the same time as the mirror cell has to prevent undue deformation of the primary mirror, it has to carry the weight of the instrument adapter and the ancillary instrumentation attached to the Cassegrain focus station. With a total maximum distance of 1500 millimetresbehind the adapter allowed for such instrumentation, an adapter weight of 900 kg and a maximum allowed weight of the ancillary instrumentation in Cassegrain focus of 250 kg, forces on the mirror cell can be considerable. This implies that high rigidity is mandatory. Reference is also made to details under "Instrumental Adapter".The cover unit for the primary mirror consists of six equal sections, roughly triangular in shape. These sections move three and three together, closing in an overlapping fashion, sealing both section joints and the sky baffle joint.
For a telescope intended for use in the optical as well as in the infrared wavelength ranges, the design of the top end and the secondary mirror has to be discussed with special care. This is especially emphasized by the fact that some of the requirements due for observations in the optical and infrared wavelength regions, respectively, tend to be somewhat incompatible.
In order to keep the light path as clean as possible for work in the infrared wavelength region, the top unit should be constructed with a diameter smaller than that of the secondary mirror. This is not in conflict with requirements for observations at optical wavelengths, as long as there is good coordination between mechanical and optical designs. For the NOT, such coordination has been given high attention.
Normally, the major conflict of interests concerning observations at optical and infrared wavelengths, respectively, iscentred on the decision whether or not to install a device in the top unit allowing wobbling of the secondary mirror. A wobbling secondary mirror is, for observations at infrared wavelengths, a straightforward solution in order to provide an improved discrimination between radiation from the target object proper, and that from the surrounding sky. However, to be of significant help, chopping must have both a frequency sufficiently high to compensate for variations of the background radiation, and a throw large enough to allow also for observations of extended objects and/or objects in contaminated sky regions. For the same reasons, chopping should be possible with respect to two axes.
Ideally, the chopping frequency should be higher than 30 Hz. Further, maximum throw should be at least several arc minutes. Concentrating on observations at infrared wavelengths only, these requirements can be met, although theyare, inevitably, in mutual conflict. With a top unit counterweight outbalancing the secondary mirror and its cell and being wobbled in counterphase to this unit, static and dynamic balance can be maintained, keeping the tube structure free from significant effects of vibration. Even for observations at optical and near infrared wavelengths, a wobbling secondary mirror may, in itself, be of use. However, in general terms, a wobbling secondary is a device causing much concern for such observations. First of all, it may be rather detrimental to image quality, violating the principle of fixed optomechanical elements. Also with sophisticated design, it is hard to avoid completely a negative influence on image quality. The situation becomes especially conflictive when image quality is given high priority, as is the case for the NOT. Further, introduction of a wobbling secondary will almost inevitably cause problems regarding pointing and tracking. Again, the higher the requirements, the more serious become the detrimental effects. It should be noted that decrease of pointing and tracking quality is, inmost respects, even more serious for work at infrared wavelengths than for corresponding work in the optical wavelength region.
Whereas a wobbling secondary mirror is rather useful for observations at infrared wavelengths, especially in the thermal infrared part of the spectrum, it is not the only alternative for improvement of discrimination between radiation from the target object and the surrounding sky, respectively. Also devices for focalplane chopping can today be made rather effective, especially if wavelengths above several micrometres are given lower weight. In many cases, a focal plane chopping device seems a perfect alternative to the problem normally resulting from solutions involving installation of a wobbling secondary mirror. Further, and most important, larger two dimensional detectors for imaging at infrared wavelengths are rapidly becoming commonly available. This fact implies that the practical consequences of the choice between different solutions for sky chopping are heavily decreased. Nevertheless, for the NOT, the alternative sky chopping solutions involving a wobbling secondary mirror and a focalplane chopping device, respectively, were thoroughly studied and compared. The outcome was, that for wavelengths shorter than around ten micrometres, the two alternative solutions did not seem to give very different results for sky suppression, whereas for longer wavelengths the wobbling secondary mirror tends to give increasingly superior performance. With highest scientific emphasis on wavelengths shorter than ten micrometres, and taking into account the limitations posed by atmospheric watervapour for longer wavelengths, it was decided that the most reasonable overall solution was to adopt focalplane chopping, leaving the secondary mirror fixed. For further details regarding alternative sky chopping solutions for the NOT, reference is made to Olofsson (1984).
To improve image quality, the top end may be moved not only in focusing mode but also laterally. In addition, it can be tilted, although this should, normally, not be necessary. These movements are all discussed below, under "Mechanical Structure".
The NOT has two sky baffles, one for the primary mirror and one for the secondary mirror. The sky baffle for the primary mirror has a total length of 2053 millimetres, a base diameter of 445 millimetres and a top diameter of 385 millimetres. The sky baffle for thesecondary mirror has a length of 630 millimetres, a base diameter of 620 millimetres and a top diameter of 715 millimetres.
Together with the quality of mirror blanks and optical figuring, the degree of sophistication of the mechanical structure define the limiting performance level of the telescope, especially with regard to image quality. In consequence, considerable attention has been paid to design of these structures. In particular, the complete mechanical structure has been the subject of detailed finite element analysis.
With the fast, f/2.0, primary mirror, optomechanical deviations are critical and have to be avoided to any significant extent. The adoption of a Cassegrain focus station further stresses the importance of an excellent mechanical structure. This is due both to the weight of ancillary instrumentation behind the primary mirror cell and to the need for a comparatively large distance between the rear plane of this mirror cell and the fork base, leading to relatively long fork arms.
Requirements on the fork can be fully taken into account in a relatively easyway. From the finite element analysis, the size of the fork can be chosen on the "safe" side. In practice, this means that the fork be made rigid enough to keep its resonance frequency above critical level. This has been fully accomplished for the NOT. Details are given below under "Resonance Frequencies".
For the telescope tube, the problem of optomechanical performance is somewhat more difficult. The basic requirement is that, in all operating positions of the telescope, the relative positions and orientations of the primary and secondary mirrors be kept constant. In total, the telescope tube has to carry not only its own weight but also that of mirror, mirror cell, instrument adaptor and ancillary instrumentation. This calls for a high quality primary mirror cell, dealt with above under "Cell for Primary Mirror". However, it also requires a stiff tube and an adequate counterweight system.
From results of finite element analysis, the structure of the telescope tube has been chosen so as to guarantee necessary rigidity. (Further details regarding telescope rigidity may be found under "Resonance Frequencies"). Four remotely adjustable counterweights are attached to the center section. The counterweights are moved in a simultaneous fashion. This will adequately take care of the problem of varying weight and size of ancillary instrumentation. As the instrument adapter rotates to compensate for field rotation, the telescope counterweights can not, in a proper way, take care of asymmetric weight distribution caused by ancillary instrumentation. This type of effects has to be compensated by counterweights on the adapter. The weight of each telescope counterweight is 200 kg and its total working range is 800 millimetres. It is added, that some unbalance can be accepted without negative effects. Assembly drawings can be found in Nordic Telescope Scientific Association Techn. Rep. (1986).
Even with maximum efforts concerning primary mirror cell, tube structure,counterweight system and top end, there will, with changing telescope position, be certain inevitable relative displacements between the primary and secondary mirrors. To compensate for these effects, the secondary mirror can be moved in various ways.
First of all, the secondary mirror can be moved along the optical axis, for focusing. The total focusing range is 8 millimetres. Further, the secondary mirror can be moved laterally with respect to the optical axis. The lateral movement is along a direction perpendicular to the altitude axis. As a result, the centre of the secondary mirror can, laterally, be moved freely along a line in the top-end plane with a stroke of 0.6 millimetres. Finally, if needed, the secondary mirror can be tilted with respect to the optical axis. The maximum tilt is one degree. Theoretically, no tilting of the secondary mirror should be necessary.
For ultimate fine tuning of image quality, an autoguider is available. At the same time as the use of this facility is recommendable, it should be stronglyemphasized that autoguiding is no substitute for an adequate mechanical structure. For optimum image quality over longer exposure times, the appropriate solution is an adequately working mechanical system, including an integrated servo system taking care of critical movements, plus an autoguider. Reference is also made to details given in "Instrument Adapter".
An integrated servo system (Jannerup, 1986) under computer control manages all movements and handles all data relevant for proper alignment and image quality. This means that the electronic system automatically controls
- rotation of telescope with respect to azimuth axis - rotation of telescope tube with respect to altitude axis - rotation of instrument adapter - support system for primary mirror - focusing movement of secondary mirror - lateral displacements of secondary mirror - tilting of secondary mirror (ifnecessary) - implementation of autoguider corrections - rotation of building - handling of data on temperature, pressure and relative humidity of ambient air plus continuous calculation and implementation of refraction data The bandwidth of the main servo loops is around 5 Hz. This is sufficiently high to guarantee a tracking accuracy, as seen by the detector, of better than a tenth of an arc second RMS.
The total image quality specification for the telescope is that in a passive mode and over a field of ten arc minutes, the integrated optomechanical system should give resulting energy concentration corresponding to 80% geometrical energy within 0.4 arc seconds for all positions with zenith distances smaller than 60 degrees. For larger zenith distances, the corresponding image quality specification states 80% geometrical energy within 0.7 arc seconds. These are instrumental image qualities well matching the best atmospheric image qualities observed. For maintenance work, it is necessary to lock the telescope in certain positions. For this reason, a safety lock pin is available on one of the fork arms. Withthis lock pin, the telescope tube can be locked in zenith position and at four degrees above the horizon.
With high priorities for optimum image quality and minimum effects of instrumental polarization, there was high preference for a single focus station implying a minimum of optical elements and a maximum of optical symmetry. As a result, it was decided to make the NOT a single focus telescope of Cassegrain type, at the same time not precluding later addition of a Nasmyth focus station. Some of the instrumental consequences of the choice of a Cassegrain focus station are discussed above under "Cell for Primary Mirror" and "Mechanical Structure". It is added, that the simpleoptical solution for the Cassegrain focus tends to be favourable also for pointing and tracking accuracy. Finally, together with a telescope with altazimuth mounting and a telescope building rotating with the telescope, choice of Cassegrain focus gives favourable conditions for cabling, especially for experimental setups. More details regarding cabling aspects are given in a description of the NOT building (Ardeberg, 1990).
For improved flexibility, especially regarding rapid and comfortable exchange of ancillary instrumentation, provision is available for pseudo-Nasmyth connection of instruments. This solution is based on a fibre optical feed through the Cassegrain instrument adapter. It allows rapid exchange of several pieces of ancillary instrumentation in a standby fashion. Regarding connection and exchange of ancillary instrumentation, more comments can be found elsewhere. The distance between the vertex of the primary mirror and the focus is 962 millimetres. For the Cassegrain focus, the free distance behind the instrumental adapter is 1500 millimetres. The corresponding maximum weight of a piece of ancillary instrumentation to be attached in Cassegrain focus is 250 kg. In both cases, the values can be regarded as generous. Also instrumentation considerably larger and/or heavier can be used, connnected via optical fibres. They can then be placed in or on the fork structure, on the observing floor or in more remote places. Further details concerning pseudo-Nasmyth connection of ancillary instrumentation are given under "Instrument Adapter" and in instrument descriptions.
In most types of engines working with bearings, smoothness of running depends, to a large extent, on the quality of the bearings. This is especially true for telescopes. The reason is that for telescopes, requirements on smooth operation are hard, at the same time as the critical movements are very slow. Inpractice, this means that telescope bearings have to operate with very low friction. It is equally important that significant effects of play are avoided.
Friction values low enough to be acceptable for telescope applications are provided by ball bearings of good quality,as long as diameters are small enough. With bearing diameters exceeding one metre, it is increasingly difficult to guarantee adequate friction values. In this case, hydrostatic bearings constitute an attractive alternative. If hydrostatic bearings are chosen, measures must be taken to minimize effects of heat dissipation from the hydraulic power supply. Adequate temperature control is essential both for the thermal stability of the telescope environment and for the viscosity of the oil.
The NOT has azimuth and altitude bearings. For the azimuth axis, horizontal loads are taken up by a ball bearing. This has a diameter of only 750 millimetres and is not exposed to more than marginal forces. As a result, friction is not a significant problem for this bearing. Also, with a small internal preload, effects of play are, in practice, fully eliminated. Both the friction and the heat dissipation of the azimuth ball bearing are fully insignificant.
For vertical loads, the azimuth axis has a bearing with a diameter of 4000 millimetres. With a diameter of this size, adoption of a ball bearing would imply a considerable risk for significant and unpredictable friction. In consequence, as a safe solution, a hydrostatic bearing has been chosen.
The hydrostatic azimuth bearing has three circular pads fixed to the yoke structure. The bottom part of these pads is a cylinder with an outer diameter of 250 millimetres and an inner one of 230 millimetres. To the five millimetres high space inside the supporting cylinder, oil is pumped under a pressure of 30 bar. This raises the pads to the moving position, with the supporting cylinders floating on an oil film with a thickness of around 30 micrometres.
The pads of the hydrostatic bearing rest on a surface consisting of a steel ring, ground to high precision. This steel ring defines the horisontal support plane for the yoke, and it has an oil sump for the hydrostatic bearing. The steel ring is supported by a steel structure with a machined surface. This steel structure rests in the concrete telescope foundation.In normal running mode, the total heat dissipation in the hydrostatic bearing is somewhat below one kilowatt. Water cooling of the hydraulic power supply is used to eliminate the most immediate potential heat dissipation problem. The oil temperature is kept constant at +5 degrees C. Thermally, the oil will not affect the telescope structure and only marginally the telescope base. It is recalled, that the complete hydrostatic bearing system as well as the base and upper part of the telescope foundation are subjected to forceful cooling by air flushing in the cooling jacket below the observing floor.
The altitude axis bearings have an outer diameter of 750 millimetres. This is small enough for ball bearings to constitute a perfect solution. Radial groove ball bearings have been chosen. The two altitude bearings are aligned to high precision in order to avoid excessive friction. Both bearings are preloaded with the yoke structure acting as a preload spring.
With the bearing and preload systems described, total friction in azimuth is ofthe order of 10 kilopondmetres. In altitude, the total friction is around 20 kilopondmetres. Both friction values can be regarded as rather favourable. It is noted, that, with drive motors disengaged, the telescope can be easily moved manually in azimuth as well as in altitude. Including ancillary instrumentation, the total moving mass of the telescope is close to 35 tons.
The NOT is driven in azimuth and altitude via gears. For each axis, there is a large gear wheel with a diameter of 2400 millimetres and two small pinions on two motors. The pinions have diameters of 100 millimetres. The gears have been machined to very high precision with special tooth compensation to assure smooth running. The tooth-to-tooth accuracy of the two largest gears is guaranteed to be better than four micrometres RMS. Accurate pointing of a telescope is an important feature for many reasons. First of all, for all types of observations, good pointing is a prerequisite for safe work, especially regarding fainter objects. Further, even if relatively bright objects are observed, high pointing accuracy means faster work. Especially for brighter objects, pointing and setting of objects can, if the pointing is of low quality, often occupy more time than the observations proper. Finally, for observations in the infrared wavelength region, accurate pointing is mandatory, as objects are often hard or impossible to locate. For observing in day time, these difficulties are greatly emphasized.
The basic prerequisite for accurate pointing is that the mechanical structure reacts in a manner predictable within limits correspondingly smaller than the pointing accuracy aimed for. Above, under "Mechanical Structure", details have been given regarding rigidity of fork and tube. Further, it has been shown howthe integrated servo system, under computer control, handles corrections necessary to compensate for inevitable relative displacements between the primary and secondary mirrors.
Naturally, high pointing accuracy can be obtained only with encoders of good quality and sufficient angular resolution. This is dealt with below under "Encoders". In addition, correction for effects of refraction has to be made quite exactly. For this purpose, temperature, pressure and relative humidity of ambient air has to be monitored continuously with sufficient accuracy. More details regarding corrections for effects of refraction are given below, under "Refraction".
For the NOT, the accuracy of blind pointing has been specified as two arc seconds RMS over the complete sky available. For differential pointing over limited fields, the corresponding accuracy should be of the order of a few tenths of an arc second.
As stated above, image quality is a matter of high priority for the NOT. This immediately calls for very accurate tracking. It must be added, that accurate tracking is, in itself, of high importance for most types of observations. As with pointing, tracking of high quality is a necessity for observations in the infrared wavelength region, and especially so for observations carried out in day time.
The tracking accuracy of a telescope depends on the quality of the mechanical structure, mirror cells, mirror support systems and servo systems. These details have been discussed above under "Mechanical Structure" and "Cell forPrimary Mirror". Further, tracking accuracy is affected by uncontrolled friction and defects of gears. This is taken up below under "Bearings" and "Gears". Finally, wind, and especially wind with high frequency variations, can influence the tracking quality in a rather significant way, if adequate precautions are not taken. More details regarding this are given below under "Sensitivity to Wind". The specification regarding tracking accuracy for the NOT was defined as one tenth of an arc second RMS. This refers to the complete sky available and to effects with frequencies of 0.01 Hz and higher.
With altazimuth mounting, field rotation is inevitable. For many types of observations, this is of no consequence.However, for certain types of programmes, field rotation must be fully avoided. One example is longslitspectroscopic observations over the surface of a galaxy, aiming at spectral and/or radialvelocity analysis as a function of distance from the centre of the galaxy. Another, rather simple, example is spectroscopy of a star with one or more nearby component(s).
Several types of optical rotators have been constructed. Most of them easily take care of effects of telescope field rotation. However, at the same time rather negative effects are added, such as light loss and decrease of image quality.Especially the light losses can be considerable.
In order to avoid completely these effects, the NOT has been equipped with a field rotator as an intermediate instrument item between the cell for the primary mirror and the instrument adapter. With this rotator, residual effects of field rotation should be fully negligible over the entire field accessible, 30 arc minutes. For further details regarding the field rotator, reference is made to "Rotator" below.
For a telescope with altazimuth mounting, a zenith singularity is unavoidable due to the fact that there is a practical limit to the azimuthal velocity of the telescope and of the telescope building.
For the NOT, the azimuthal velocity has an upper limit, corresponding to the slewing speed, of three degrees per second.As a result, the minimum zenith singularity possible has a size of around 30 arc minutes times 20 arc minutes, or about 0.2 square degrees.
It should be noted, that close to the zenith, the dome will introduce vignetting, due to limitations in the maximum angle to which the upper hatch can be opened.The surface around the zenith position, for which vignetting is introduced, has an extension of 1.3 degrees for an object on the optical axis. From a practical point ofview, the zenith singularity as well as the sky surface around the zenith affected by vignetting are of negligible consequences.
For good tracking and image quality, a stiff mechanical structure with high resonance frequencies is an absolute necessity. This need is highly emphasized in the presence of significant wind buffeting. Further, great care must be taken to avoid transmission through the foundations and/or ground of effects of lower resonance frequencies of the building and dome structure.
The azimuth axis has its lowest resonance frequency at 7.4 Hz. For the altitude axis, the lowest resonance frequency is 14 Hz. These are values which can be classified as very safe, taking into account basic structural rigidity as well as effects of wind buffeting. Coupling between building and telescope has been minimized. Asa special precaution, foundations of telescope and building have been completely separated. The foundation of the building rests in a surface rock layer with a thickness of around 1.5 metres. The foundation of the telescope goes down to another rock layer with its upper surface about three metres below ground. The material between the two rock layers consists of rather soft material. Further, the rail for the rotating telescope building has been placed on rubber sheets.
Correct choice and matching of encoders are essential for accurate positioning of the telescope. The NOT has a total of five encoders. Two encoders are installed for position readings of the azimuth axis and two for corresponding readings of the altitude axis. Finally, one encoder is installed in the rotator for readings of rotator position.
For readings of the azimuth and altitude positions, the choice is a combination of one static and one incremental encoder. The two identical solutions include one absolute Stegmann encoder and one semiabsolute Indoctosyn encoder. The Stegmann encoder has an absolute accuracy of 80 arc seconds. The Indoctosyn encoder repeats itself and is absolute within one cycle of one degree. Combination of the two encoders results in an absolute positional accuracy somewhat better than two arc seconds. Over limited fields, differential accuracy should be considerably better.
The rotator has been equipped with a Stegmann absolute encoder only. The resulting accuracy of 20 arc seconds is fully sufficient for this purpose. With the largest telescope field possible, 30 arc minutes, the corresponding maximum error over a two hour exposure corresponds to only a few hundredths of an arc second.
In the top end unit, two linear transducers are installed. These transducers transmit data for positions of the secondary mirror in focus direction and in lateral direction. In both cases, the accuracies delivered by the linear transducers are beyond what is needed for practical purposes.
For telescopes with conventional pointing and tracking accuracies and for which fine object tracking is done either manually or applying direct corrections from an autoguider, refraction effects are of minor importance. In such cases, effects of refraction are noticeable only for longer exposures of objects at large air masses and for which exposure and guiding, respectively, are made with light of significantly different wavelengths.
With high blind pointing and tracking accuracy, effects of refraction are much more noticeable. For the NOT, already blind pointing should be accurate to around two seconds of arc (Laustsen and Klim, 1985). This means that one might expect to achieve accuracies for differential pointing in limited fields which are of the order of a few tenths of an arc second. This, in turn, implies that the telescope, as spinoff data, can deliver astrometric positions of reasonable quality. Further, and for many purposes of high importance, we should be in aposition to achieve tracking accuracies better than a tenth of an arc second RMS.
With expectations for differential pointing and tracking down to and better than a few tenths and one tenth of an arc second, it is obvious that effects of refraction have to be taken quite seriously. In practice, this calls for inclusion of accurate refraction corrections as an integrated part of the servo system for the telescope. In the case of the NOT, such corrections are based on local meteorological data obtained from a meteorological station placed at a distance of 50 metres from the telescope building. These data are received and processed by the telescope control system, appropriate corrections being executed through the servo system.
The sky available to the observer is, in principle, limited by the geographical latitude of the site, the altitude of the telescope, surrounding landscape and restrictions imposed through hardware and software arrangements. In the case of the NOT, practical limits to the sky available are defined by the geographical latitude and software limits.
With a geographical site latitude of +28°45'20.5", and a software limit at 4.5 degrees above the horizon, the sky available for observations includes the complete Northern Hemisphere plus the Southern Hemisphere north of declination -56. For practical purposes, the corresponding limitation in declination should be regarded as around -50.
It must be pointed out, that for normal observations it is prohibited to attempt to override the software limitation in height above the horizon. For absolute safety, a mechanical endstop is set at four degrees above the horizon. Also for altitude positions beyond the zenith, the telescope has a software limitation set at 3.5 degrees beyond zenith. The corresponding mechanical endstop is at four degrees beyond zenith.
The mechanical endstops in altitude consist of shock absorbers acting between the moving part of the telescope and the stationary counterpart. A compression of the oil-filled shock absorbers results, via a current limiting switch, in cutting of the power to the altitude motor.
With telescope positioning under computer control, warnings will be issued via the control system, as soon as the altitude limits are approached. Positioning orders including passing of the lower altitude limit will not be executed. Instead, an explanation will be delivered via the control system.
All turbulence with the exception of intrinsic atmospheric site turbulence may be described as man-made turbulence. This can, in turn, be divided into observatory turbulence, enclosure turbulence, telescope turbulence and instrumental turbulence. All of these four components of man-made turbulence are very important (Ardeberg and Andersen, 1990). However, in this context we are only concerned with telescope turbulence. Telescope turbulence can be seen as composed of turbulence resulting from telescope structure and from the primary mirror, respectively. Both components can give considerable contributions, especially important in the case of telescopes with tight image quality error budgets.
Turbulence from the telescope structure has two main sources. One is the telescope structure proper, the other consists of devices attached to the telescope structure. The telescope structure in itself produces turbulence dueto heat exchange with ambient air and sky.
Most conventional telescopes have unfortunately large amounts of electronics devices and motors attached to the telescope structure. Further, the heat dissipation from these installations is often quite considerable, a severe disadvantage emphasized by the closeness to the optical path.
Nocturnal heat flow from the primary mirror is a very prominent obstacle to high image quality. Normally, the heat capacity of primary mirrors is very large, with thermal time constants being of the order of a number of hours, including several days in some cases. The critical position of the primary mirror highly emphasizes its role as a source of severe thermal inbalance.
A hot primary mirror threatens image quality in two ways. First, a nonuniform temperature distribution within the mirrortends to deform the mirror surface. This is not, in itself, to be referred to as telescope turbulence. Second, a very important contribution to telescope turbulence is provided by the hot air layer above the mirror surface. The convective activity of this air layer can easily dominate the resulting imaging performance for telescopes with tight total error budgets in image quality.
For the NOT, effects of telescope turbulence have been taken very seriously. Decided attempts have been made to minimize to the extent possible all sources of telescope turbulence discussed.
A possible temperature gradient along the telescope structure has been given special attention. Obvious sources of such a gradient are the hydrostatic bearing and general heating of the telescope base and yoke.
Introduction of a thermal gradient along the telescope structure via the hydrostatic bearing has been minimized in two ways. First, the hydrostatic power supply iscooled. Second, the hydrostatic bearing as well as the lower part of the yoke and the base of the telescope are enclosed in the heat trap between the observing floor and the ground floor of the telescope building.
General heating of the telescope base and yoke has been minimized through a series of precautions. First, all spaces in the ground floor of the telescope building are kept at low temperature, carefully maintained by air conditioning, excess heat being ducted away. Second, these spaces are heavily insulated. Third, a "cooling jacket" constitutes an effective protection against all remaining vertical heat leakage.
In order to minimize the risk of temperature gradients along the telescope tube, the complete air volume around the telescope is subjected to active temperature control. For this purpose, both floor cooling and cooling of the air in the higher part of the dome have been established. For the heat exchange between the telescope structure and nightsky, further studies on site seem necessary. Further, the centre section of the telescope has its inner surface insulated.
The amount of electronics devices and motors attached to the telescope structure has been restricted to an absolute practical minimum. Cabling and remotely placed electronic devices have been preferred whenever possible. At the same time, heat dissipation of remaining units has been kept as low as possible.
On the telescope structure itself, a total of 15 motors are present. These are two azimuth motors, two altitude motors, one focusing motor for the secondary mirror, one motor for lateral displacement of the secondary mirror, two motors for possible tilting of the secondary mirror, six motors for the flaps of the mirror cover and, finally, one motor for the telescope counterweights. Normally, only the four main drive motors and the two rotator motors are running during observations. The two motors for azimuthal movement of the telescope have heat dissipations which are very low. Further, both motorsare cooled. Moreover, they are located in the cooling jacket environment. As a result, thermal influence on the telescope structure should be negligible, even if it has to be noted, that high wind forces can deteriorate the picture somewhat.
Also the two motors installed for altitude movement of the telescope have very low heat dissipations. They run in a continuous mode during normal observing. Like the azimuth motors, the altitude motors are cooled. This should reduce the thermal influence of these motors to negligible amounts.
The motor installed in the top end for focusing movements of the secondary mirror has a very low heat dissipation. Further, during normal observing activity, it is activated only in an intermittent mode. It seems fully justified to regard thermal influence of this motor as insignificant.
The necessity for lateral movements of the secondary mirror is very limited. The motor for this purpose, installed in the top end, has very small heat dissipation.The thermal effects of this motor can be regarded as insignificant.
Angular adjustment of the secondary mirror is achieved through axial corrections in two of the three support points. For these corrections, two motors are installed in the top end. They have very low heat dissipations. During normal observing, these motors are not active. Thus, they have no thermal influence.
The six motors installed in the centre section for operation of the flaps of the mirror cover are inactive during observing work. Thus, they have no thermal influence. It is noted, that it belongs to normal prudence to open the mirror cover well in advance of observing initiation.
On the centre section, there is finally a motor attached for movements of the counterweights. This motor is activated only in connection with exchanges of ancillary instrumentation attached directly to the Cassegrain focus station. For this reason, it has no thermal influence.
In addition to the azimuth and altitude motors, a total of four encoders are installed. All have heat dissipations which are negligible.
The last item to be mentioned in connection with heat dissipation along the telescope structure is the rotator unit. The rotator has two motors and one absolute encoder. The two motors have very low heat dissipations and the absolute encoder a negligible heat dissipation. For some observing programmes, the rotator can be maintained inactive, in which case it has no thermal influence.
It is added, that the top end unit contains two linear transducers. These are used to measure displacements of the secondary mirror along and perpendicular to the optical axis. The heat dissipation in these transducers is insignificant.
Finally, there are certain possible thermal effects from the instrument adapter and from ancillary instrumentation. The thermal influence of the adapter is discussed under "Instrument Adapter",whereas those of ancillary instrumentation are detailed elsewhere.
For the NOT, mirror blanks are of Zerodur. The very low thermal expansion coefficient of this material is an excellent guarantee that distorsions of the surface of the primary mirror will be below detection level even for worst possible practicle temperature differences between the primary mirror and ambient air.
Creation of a heated, convective air layer above the primary mirror is a problem much more difficult to solve. Flushing of the mirror surface with cool air is far from simple and is not evident as remedy. For a primary mirror as large as that of the NOT, air flushing of the surface may well add more convective activity than it eliminates.
Clearly, the most efficient cure against mirror turbulence, at least in principle, is to maintain the mirror temperature equal to that of ambient air. This involves two difficulties, both hard to solve. These difficulties concern thermal relaxation and temperature prediction. The heat capacity of the primary mirror is quite low because of the high aspect ratio. The resulting time constant is of the order of a couple of hours. Still, thermal relaxation has a time scale far from insignificant compared to diurnal and nocturnal cycles.
Problems resulting from the time scale of thermal relaxation of the primary mirror are severely emphasized by difficulties encountered in forecasting nocturnal ambient temperatures. Unfortunately, these difficulties imply that also with optimum preparations, measures against telescope turbulence originating in the air layer above the surface of the primary mirror will, unavoidably, contain a component not fully within control. The best possible remedy is to keep the temperature of the telescope ambience as constant and close to forecasted night temperature as possible with a preference for slightly lower temperatures. This is the policy followed for the NOT. It is added, that a study of local temperature conditions is foreseen. This will improve both forecasting of night temperature and measures to maintain temperature control.
Wind affects observing conditions in many ways. In general, complete absence of wind is favourable for telescope pointing and tracking but not necessarily the optimum condition concerning the microthermal climate of the site and for telescope induced degradation of image quality. Wind of modest speed should be of no consequence for the operation of an adequately constructed telescope, at the same time as it might well improve the microthermal climate of the site andfurther turbulence due to the telescope. High wind speeds tend to influence both telescope operations and microthermal site climate in a negative way.
Regarding telescope behaviour, the primary negative effect of high wind speed is buffeting. Clearly, wind with a low bandwidth influences telescope operating parameters less than wind with a spectrum characterized by a prominent high frequency part.The telescope parameters most immediately influenced by wind (buffeting) are pointing, tracking and sensitivy to dust. To some extent, also telescope turbulence will be affected. The tracking and telescope turbulence directly influence image quality.
Concerning pointing and tracking, a minimum sensitivity to wind buffeting can be achieved with a mechanical telescope structure with maximum high resonance frequencies. In this respect, more details can be found under "Resonance frequencies" and "Mechanical Structure" above.
The dependence on wind characteristics of sensitivity to dust are evident. For more details, reference is made to "Sensitivity to Dust" below.
Both the degree and effects of telescope turbulence may be significantly influenced by wind. This type of wind sensitivity depends on various parameters. In the case of the NOT, given its rather low level of telescope turbulence, such effects should be close to negligible. For the NOT, protection against effects of wind can be achieved through adequate combination of the positioning of gates and hatches. At the same time, great care should be taken for wind speeds higher than 10 metres per second, especially in case of gusty wind.
In itself, the telescope should maintain its pointing and tracking quality even for wind speeds up to 22 metres per second. However, due to the risk for presence of dust in the air at higher wind speeds, the dome should be completely closed (gates and hatches) for wind speeds above 20 metres per second. For the same reason, all gates and the lower hatch should be completely closed for wind speeds higher than 12 metres per second. Further, for wind speeds above 15 metres per second, the telescope should not face the upwind direction nor any direction within 45 degrees of the upwind direction.
Regarding the safety limits detailed above, instructions will be given via the control system. This includes warnings when limits are approached. Further, the control system takes over in an automatic safetymode, if instructions are not duly followed. In this way, possible temptations to continue observing beyond safety limits should not affect the telescope adversely. It is important to remember that the dome must be completely closed (gatesand hatches), as soon as the telescope is subjected to wind vibration. This applies also to wind speeds lower than the limits stated above. Continued exposure to wind vibration will seriously affect gears and bearings. Details regarding hatches and gates are given elsewhere.
No part of the telescope nor of the ancillary instrumentation should be exposed to water, neither in the form of precipitation nor as condensation. For this reason, the dome should be completely closed (gates and hatches) as soon as there is risk for condensation and/or precipitation.
First of all, when the sky is overcast, the upper hatch should be maintained closed. Further, the dome should be completely closed (gates and hatches) as soon as the local relative humidity is 95 per cent or higher. The same applies when condensation and/or precipitation is/are observed, whatever the reading for the local relative humidity. It is highly recommendable to pay special attention to circumstances as soon as the local relative humidity is above 80 per cent and/or when it is in a state of rapid increase.
If the local relative humidity has reached values of or above 95 per cent and/orprecipitation and/or condensation has/have been noted, the dome should remain completely closed (gates and hatches) until the local relative humidity has fallen to values below 90 per cent and signs of precipitation and condensation do not appear.
Higher levels of humidity are especially dangerous when coinciding with high amounts of dust in the air. This is discussed below under "Sensitivity to Dust". Concerning hatches and observing wall gates, details can be found elsewhere.
Already small amounts of dust in the air are detrimental for telescopes. Accumulated effects on sensitive parts can be very serious and heavily decrease operational quality. Examples of such sensitive parts are gears, bearings and optical surfaces.
In principle, the most direct way to control the presence of dust in the air is through use of dust meters. Such meters are available commercially. Some of them are rather reliable. However, most of the dust meters are designed to operate in conditions rather different from those at sites for optical telescopes. Many of them are integrating devices. If long integration times are necessary for consistent data, the applicability to site conditions is doubtful.
In the absence of reliable dust meter data, the presence of dust in the air has to be taken with greatest concern. It is especially alarming when coinciding with significant wind speeds and/or higherlevels of humidity. As soon as the wind speed exceeds 10 metres per second and dust is present in the air, the gates and lower hatch of the dome should be closed. Further, in these circumstances, the telescope should not face the upwind direction nor any direction within 45 degrees of the upwind direction. If dust is present in the air, the dome must be completely closed (gates and hatches) when the wind speed exceeds 12 metres per second.
Presence of dust in the air may provoke condensation at levels of local relative humidity much lower than those normal for air free from dust. This type of condensation may occur very rapidly. Therefore, coincidence of higher levels of humidity and presence of dust in the air calls for special attention. Details regarding hatches and gates are given elsewhere.
With a telescope with altazimuth mounting, field rotation has to be taken into account. For many types of observing programmes, field rotation is hardly noticeable, and for even more types of programmes it is without consequences, except for autoguiding. At the same time, for some types of observing programmes, field rotation is rather important. This has been discussed above under "Field Rotation". The field rotator provided for the NOT rotates the field to an accuracy of 20 arc seconds. For all practical purposes, this is more than sufficient, as discussed under "Encoders and Transducers". At the same time, the rotator can be used to position ancillary instrumentation to selected angular positions at corresponding accuracy. This may, for example, be used in case of long-slit spectroscopy of galaxies and nebulosities.
With a blind pointing accuracy of two arc seconds for the NOT, use of a finder telescope is superfluous for normal observing. However, for testing and debugging in initial phases and after reassemblies in connection with aluminizations of telescope optics, access to a finder telescope is often helpful. For this reason, the NOT has been provided with a finder. The finder telescope is a refractor with a free aperture of 153 millimetres and a focal length of 2130 millimetres. It has a rigid tube and is firmly attached to the centre section of the NOT. A modest CCD camera is used as detector. The image resolution is close to one arc second and the limiting magnitude with normal scanning frequency is around V=7.
The instrumental adapter for the NOT is intended to provide interface between the telescope and ancillary instrumentation. At the same time, it includes some observing functions common to most types of observation. Among the functions possible, one may mention field viewing, periscope viewing, autoguiding, image quality monitoring, offset guiding, autocollimation for alignment verificationof telescope optics, complete checking of prevailing image quality, framegrabbing and limited photometry. In addition, the instrumental adapter holds a permanently mounted standby CCD camera, possible to switch in within a matter of seconds. The main purpose of this latter arrangement is to permit the observer to take advantage also of excellent image quality prevailing for shorter periods only.Further, the adapter carries a Hartmann-Korhonen device for online determination of optical quality.
With a considerable number of electronic devices, the instrumental adapter has, intrinsically, a heat dissipation far from negligible. For this reason, a special cooling unit has been designed. With thisunit active, resulting heat dissipation from the adapter is negligible.
For further information regarding the instrumental adapter, reference is made to a detailed description by Florentin Nielsen (1989). Assembly drawings can be found in Nordic Optical Telescope Scientific Association Techn. Rep. (1986).
In addition to improvements of alignment, tracking and thermal control, some projects are in course for further upgrading of our telescope. Among these projects we mention, first, installation of a system of temperature probes designed to monitor temperatures of ambient air, enclosure, air volume around the telescope, on and below the observing floor and of various parts of the telescope, including both optical and mechanical elements. Resulting data will be used for further improvements of routines for thermal control. Second, withan aspect ratio of 1:13.5, the primary mirror is well suited for active mirror support corrections. We have taken up design of corresponding modifications and supplements. Third, advanced laboratory simulations, supported by activities for the design of the Large Earthbased Solar Telescope, entrusted to our telescope group, strongly indicate that air flushing of the surface of the primary mirror can significantly improve turbulence conditions. We intend to follow this up, and a possible design is presently under discussion.
Andersen, T.B. 1985: Optical Specifications and Performance of the Nordic Optical Telescope, Technical Report from the Nordic Optical Telescope Scientific Association
Andersen, T.E. 1981: 2.5 m Telescope Design Study, Birkerød, Denmark
Andersen, T.E. 1982: 2.5 m Telescope, Addendum to Design Study, Birkerød, Denmark
Andersen, T.E. 1986: Mirror Cell Pressure Regulator, Technical Report from the Nordic Optical Telescope Scientific Association
Andersen, T.E., Jessen, N.C. 1985: Deformation Calculations of the Primary and Secondary Mirrors of the Nordic 2.5 m Optical Telescope, Technical Report from the Nordic Optical Telescope Scientific Association
Ardeberg, A. 1983: The Case for a 2.5 m Telescope, Proc. Nordic Astronomy Meeting in Oslo, Aug. 17, 1983, Inst. Theor. Astrophys. Blindern, Oslo, Report No. 60, p. 7
Ardeberg, A. 1984: Ancillary Optical Instrumentation - Provisions and Options, Proc. Nordic Astronomy Meeting, Sept. 3-5, 1984, Obs. and Astrophys. Lab. Univ. Helsinki, Report 6/84, p. 121
Ardeberg, A. 1985: Nordic Optical Telescope, Vistas in Astronomy 28, 561
Ardeberg, A. 1987: On the Nordic Optical Telescope, Observational Astrophysics, Methods and Techniques in Optical Astronomy, Proc. Nordic Research Course, Brorfelde, June 1-12, 1987, ed. R. Florentin Nielsen, p. 105
Ardeberg, A. 1990: The Enclosure of the Nordic Optical Telescope, Nordic Optical Telescope Scientific Association.
Ardeberg, A., Andersen, T. 1988: VLT Design Implications of the Nordic Optical Telescope, Proc. ESO Conf. on Very Large Telescopes and their Instrumentation, Garching, 21-24 March 1988, ed. M.-H. Ulrich, p. 183
Ardeberg, A., Andersen, T. 1990: Low Turbulence - High Performance, SPIE 1990 Symp. on Astronomical Telescopes & Instrumentation for the 21st Century, Advanced Technology Optical Telescopes, to be published
Florentin Nielsen, R. 1989: Cassegrain Adaptor for the Nordic 2.5 m Telescope, Technical Note from the Nordic Optical Telescope Scientific Association
Haarala, S., Korhonen, T., Lappalainen, T., Sillanpää, A. 1988: Optical Tests of the Primary Mirror for the Nordic 2.5 m Telescope, Proc. ESO Conf. on Very Large Telescopes and their Instrumentation, Garching, 21-24 March 1988, ed. M.-H. Ulrich, p. 589
Jannerup, O. 1986: Design of Digital Main Servos for the Nordic 2.5 m Optical Telescope, Technical Report from the Nordic Optical Telescope Scientific Association
Korhonen, T. 1987: Optics for the Nordic Optical Telescope, Observational Astrophysics, Methods and Techniques in Optical Astronomy, Proc. Nordic Research Course, Brorfelde, June 1-12, 1987, ed. R. Florentin Nielsen, p. 198
Korhonen, T. Haarala, S., Piironen, J., Sillanpää, A. 1985: Manufacturing Optics for the Nordic 2.5 m Telescope, Departm. Phys. Sci. Univ. Turku, Report R 84
Laustsen, S., Klim, K. 1985: Telescope Pointing and Tracking, Technical Note from the Nordic Optical Telescope Scientific Association
Nordic Optical Telescope Scientific Association 1986: 2.5 m Telescope Assembly Drawings, Technical Report NOTSA
Nordic Optical Telescope Scientific Association 1986: 2.5 m Telescope Adaptor Assembly Drawings, Technical Report NOTSA
Olofsson, G. 1984: Do we need a Wobbling Secondary for NOT?, Proc. Nordic Astronomy Meeting, Sept. 3-5, 1984, Obs. and Astrophys. Lab. Univ. Helsinki Report 6/84, p. 171
Svärdh, I. 1989: User Manual for the Nordic 2.5 m Telescope Control System, Technical Report from the Nordic Optical Telescope Scientific Association
|
||
|