WO2009077799A1

WO2009077799A1 - Apparatus for eliminating angular velocity fluctuations of mass produced telescope mounts

Info

Publication number: WO2009077799A1
Application number: PCT/HU2008/000152
Authority: WO
Inventors: Attila Madai; Istvan Papp
Original assignee: Mda Telescoo Kft.; Meade Instruments Euro E Gmbh & Co. Kg
Priority date: 2007-12-19
Filing date: 2008-12-17
Publication date: 2009-06-25
Also published as: HU0700820D0; HU227934B1; HUP0700820A2

Abstract

The invention relates to an apparatus for eliminating angular velocity fluctuations of mass produced telescope mounts. The telescope mounts comprise a declination shaft (3) adapted for supporting a telescope tube (2), and a right ascension shaft (4) set perpendicular to the declination shaft (3), with driving motors (6, 11) having a drive control unit (5) being connected through drive trains (7, 12) to the declination shaft (3) and the right ascension shaft (4). The drive control unit (5) has an external control input (aut. port). The invention is essentially characterised by that an angular velocity encoder (1) is attached by means of a mechanical adapter installed subsequently (A) in a direct coaxial manner to the right ascension shaft (4). The output (Ki1) of the angular velocity encoder (1) is connected to the process signal input (Be1) of an electronic measurement and control unit (8, 13) that has a quartz oscillator (Osc), where the error signal output (Ki2) of the electronic measurement and control unit (8, 13), providing the correction signal, is fed back to the external control input (aut. port) of the drive control unit (5) of the telescope.

Description

Apparatus for eliminating angular velocity fluctuations of mass produced telescope mounts

The object of the invention is an apparatus for eliminating angular velocity fluctuations of mass produced telescope mounts, where the telescope mount comprises a declination shaft adapted for supporting a telescope tube, and a right ascension shaft (polar shaft) set perpendicular to the declination shaft, with driving motors having a drive control unit being connected through drive trains to the declination shaft and the right ascension shaft, and with the drive control unit having an external control input.

The inventive apparatus is capable of operating without applying an external reference signal (such as a guide star). The field of application of the invention includes commercially available mass produced equatorially mounted telescopes utilised for astronomical observations and measurements, and for analogue and digital astrophotography.

The design sought to be patented is an apparatus that is electrically connected to the drive control circuitry of a telescope mount and is mechanically connected to the right ascension shaft of the telescope mount in a direct manner (without any gearing or coupling). The apparatus is capable of stabilising the motion of the telescope during astronomical observation without the utilisation of an external reference signal (e.g. guide star position signal) such that the telescope accurately tracks the apparent movement of an observed celestial object to keep the object stationary in the telescope's field of view.

It is of common knowledge that in the course of astronomical observations performed using a telescope one of most important and most challenging tasks is pointing the telescope to and keeping it fixed on the targeted object for the whole duration of observation, 'meaning that the apparent motion of celestial objects, caused by the Earth's rotation, has to be compensated accurately (preferably with an accuracy of ± 1") during the usually few seconds-to-10-15 minutes observation duration utilising a motorised mechanism. However, atmospheric movements such as tropospheric and stratospheric winds, as well as other varying meteorological conditions (air temperature, humidity) cause the image of a star to "flicker" or "twinkle" to a certain extent due to the continuously varying atmospheric refraction; a phenomenon called "scintillation" in astronomy. Scintillation causes the wavefront of light coming from the star to be distorted, which results in that the star's image undergoes very fast random oscillations (typically with an amplitude of 3-5" and a frequency of 5- 20Hz) around its actual position. During extremely calm nights the oscillation amplitude may decrease jto 2-3" at sea level, and even further in observatories located at higher above-sea levels. To keep the thus produced image of a star (the so-called seeing disc) free from additional distortions, the tracking error of the telescope must be lower than 2" (that is, in the ± 1" range - hence the above specified accuracy limit).

A known solution for cancelling the apparent motion caused by the Earth's rotation consists in rotating the telescope about the right ascension axis (aligned with the rotation axis of the Earth) of a mechanism of two perpendicularly arranged moveable shafts applied for supporting and pointing the telescope. The right ascension shaft is rotated with an angular velocity corresponding to the Earth's rotation (but in the opposite direction) utilising a motorised gear drive. However, the limited accuracy of machining and assembly of the mechanical parts of the drive unit places restrictions on the real-world accuracy of the tracking mechanism, causing the angular veloc/ty of the telescope to fluctuate to some extent around the preset value. The targeted celestial object will thereby not remain exactly fixed in the field of view but will continuously move to a certain degree, which is unacceptable for accurate measurements.

It should be noted here that one arc second (1") is the 1/1.296.000th of the full 360° circular arc (360° x 60'x 60"); in other words it is the apparent angular diameter of a 25-mm diameter coin seen from a distance of 5 km. The implementation of a mechanical drive unit with such a high accuracy places extremely strict requirements on machining allowances of individual parts. In practice this means that very costly machining methods must be applied; and in most cases such accuracy is not feasible to reach for mass produced products. Today the best, premium-quality serially produced telescope mounts (e.g. the Paramount ME) reacjt a long term tracking accuracy of only ± 3" - ± 5" at a premium price corresponding to the quality of the mechanism, and the average level of accuracy for most commercially available telescope mounts is usually in the range of ± 10" - ± 20" but often deteriorates to ± 30" or even ± 40".

According to known solutions, external control is applied for improving the tracking ability of telescope mounts.

The most widespread of these solutions involves utilising a supplementary guide camera and/or guide scope (a so called "autoguider") that measures the position of a suitably chosen, sufficiently bright guide star located in the vicinity of the object to be observed to establish a correction signal for real-time motion control of the motors applied for rotating the telescope. (Thereby, the error signal to be fed back to the control system is established using a reference signal external to the telescope mount.) A solution utilising'' a guide star is disclosed in patent specification US

2007/0115545 A1. According to this solution, first a bright guide star is chosen, and then a drive control electronics and a guide scope equipped with a CCD camera (electronic eyepiece) are applied for determining motor velocity values required for automatically tracking the observed object using the measured and constantly updated position of the guide star and a star database stored in the memory of the control system. Since the telescope has a horizontal mount, in order to properly track the targeted object (the star to be observed) the telescope must be continuously rotated about both axes during exposition by continuously changing speed. The disadvantage of this solution is that angular velocity fluctuations of the telescope mount are not directly compensated. The error signal applied for motion control is calculated from average velocity values (by measuring the position of the guide star).

An advantage of systems utilising an autoguider is that all error types are corrected (irrespective of their source) in case the guide star is well-chosen (is close enough to the object to be observed and bright enough), but this kind of system cannot be applied in all real-world situations. Main disadvantages of known autoguider systems:

- it is often impossible or very difficult to find a sufficiently bright guide star in the vicinity of the object to be observed,

- in case a filter blocking a significant amount of incoming light (e.g. H-alpha) is used for the observation, a separate guide scope is necessary,

- the solution may not easily be automated because a suitable guide star has to be found and selected for each observation,

- if a fainter guide star is used, short-period tracking errors of the telescope mount cannot be fully eliminated due to the required relatively longer CCD integration time,

- in case of many observation tasks (e.g. nova and comet search, observation of variable stars) the time consuming initial setup causes time loss and deteriorates observation efficiency,

- the solution requires extra equipment (guide scope, guide CCD camera), increasing the combined mass that has to be moved by the telescope mount.

According to another, less widespread, known solution an angular position sensor (angular position encoder) is attached either to the shaft of the driving motor or to an intermediate element of the drive train. During calibration the encoder's signal is compared with the signal of an autoguider (or a reference encoder) to establish an error signal that is applied for training the telescope's own control system. During observations the angular position of the given shaft is measured by the encoder, and the control system applies the rpm corrections that were previously recorded. This method is known as "training of the control system". A solution based on training utilising an encoder is disclosed in the document WO 2005/101089. According to this solution the angular position encoder(s) or rotational speed encoder(s) are disposed on the shaft of the worm and/or the wormwheel of the telescope mount's drive train.

A modified autoguider-based control method may also be applied for the "training of the control system". In this case the "error signal" coming from the autoguider is recorded for the duration of one or more cycles of the rotary motion of the telescope mount's mechanism, and during observation it is "played back" to correct the rpm (angular velocity) of the driving motors. A common drawback of all solutions based on the "training of the control system" is that the correction is not based on real-time feedback and thereby the system is not capable of correcting the actually occurring errors. Disadvantages of training of the control system using encoders: - since stochastic errors have a significant share in the combined mechanical error of the mount, the recorded angular-velocity correction curve can never match the actually required correction values, thereby there will always be a residual error in the system;

- the encoder is capable of directly measuring mechanical errors arising upstream of its location in the drive train (that is, between the rotor of the motor and the shaft of the encoder), but the errors produced downstream of the encoder may only be detected indirectly, which introduces further uncertainty into the system,

- the coupling of the encoder's shaft to the drive train (through a friction-wheel drive, toothed- or flat-belt drive, or gear drive) will be the source of additional errors;

- in case the drive has multiple stages, the error curve produced as the superposition of periodic errors of the individual stages may include harmonics (beating of waves) that may be difficult or impossible to train in their entirety because their cycle time may be significantly longer than the training time usually applied (equalling typically one or two cycles of the worm).

The solution involving position measurement with encoders for the training of the control system may only be utilised for the control of state-financed professional giant telescopes used as primary instruments by astronomers, and even there only as an integral part of a greater control system. (For instance the UKIR telescope, or the Keck 10 m telescope, see also: Mark Trueblood & Russel

Merle Genet: Telescope Control; Willmann-Bell, Inc.; 1997 - section 6.6.2, p140. on the telescope of the Lick observatory).

In addition to the apparent motion of the objects to be observed (caused by the Earth's rotation) atmospheric refraction may also affect the measurements. Acting as a meniscus lens, the atmosphere alters the apparent position of stars in the sky. The effect is zero at the zenith and, as light rays coming from celestial objects have to pass through a refracting air layer of rapidly increasing thickness, increases almost exponentially towards the horizon. Atmospheric refraction shifts the apparent position of objects only in a direction perpendicular to the horizontal plane, increasing their elevation, but since the direction of the diurnal motion of celestial objects is not perpendicular to this direction (except at the moment of meridian transit), there is an apparent change in both equatorial coordinates (RA, right ascension, and DEC, declination).

The deviation caused by the effect of atmospheric refraction is virtually negligible in case of observations with relatively short duration (max. 2-3 minutes) carried out with shorter-focus instruments (with a focal distance under 1000-1500 mm) at elevations of 30-40° or above. For these kinds of observations it is sufficient to compensate the error caused by the angular velocity fluctuations of the telescope mount.

The objective of the invention was to overcome the disadvantages of known systems involving autoguider control or training based on a reference signal by providing an apparatus that may be easily adapted (without modifying the mechanism of the mount) to be utilised with commercially available mass produced telescope mounts of average or even below-average quality, and is capable of increasing the tracking accuracy of such telescope mounts to a level very close to the accuracy of highest-quality or even professional custom- fabricated mounts. Our invention is based on the recognition that for eliminating tracking errors the process signal should be measured directly at the last (most downstream) element of the drive train of the mount because the process signal measured in such a manner carries the combined errors of all the individual elements of the drive train, including periodic and stochastic components. We have also recognised that due to the stochastic error components the relevant information is carried by the momentary rpm (angular velocity) values, and by comparing these values with the values determined by the "principle of motion" of the object to be observed the error signal necessary for accurate motion control may be determined. The most important implication of our recognition is therefore that a realtime compensation determined utilising the measured actual angular velocity fluctuations should be ap^'plied to cancel the tracking errors of the mount and this goal is achievable with an appropriate electronics and encoder subsequently and easily installed onto the mount.

The inventive objective is accomplished by providing an apparatus specified in the introductory paragraph of the present specification, in which apparatus an angular velocity encoder is attached by means of a mechanical adapter in a direct coaxial manner to the right ascension shaft; the output of the angular velocity encoder being connected to the process signal input of an electronic measurement and control unit that has a quartz oscillator, and with the error signal output of the electronic measurement and control unit, providing the correction signal, being fed back to the external control input of the drive control unit of the telescope. The output of the measurement and control unit is connected directly or indirectly

(through a suitable level ςonverter unit) to the external control input, the so-called autoguider port, of the existing drive control unit of the telescope.

In a preferred embodiment of the apparatus according to the invention, henceforth called in accordance with astronomical terminology the sidereal variant (a variant with an angular velocity corresponding to the Earth's rotation), the measurement and control unit has an interpolator signal processing unit connected to the analogue output of the angular velocity encoder, and a microcontroller correlator unit connected to the output of the signal processing unit. A quartz oscillator is connected to the reference input of the microcontroller correlator unit, with the output of the correlator being connected to the external control input of the drive control unit of the telescope mount.

In a further preferred embodiment, the so-called "King-Rate" variant capable of providing refraction-corrected motion control, the measurement and control unit consists of a/ signal processing unit connected to the output of the angular velocity encoder, and a central computer unit disposed at the output of the signal processing unit, with a quartz oscillator being connected to the reference input of the central computer unit, and with the output of the central computer unit being fed back to the external control input (autoguider port) of the drive control unit of the telescope mount through a level converter unit. According to this embodiment the central computer unit has a bidirectional data transmission link to the drive control unit of the telescope through a standard serial data transfer channel. In case the apparatus is adapted to be utilised with a telescope mount where motion control is performed by an external control computer, the central computer unit of the measurement and control unit has a bidirectional data transmission link to the external computer. In a further preferred embodiment the central computer unit of the measurement and control unit also has an autoguider input.

The mechanical adapter applied for attaching the angular velocity encoder directly to the right ascension shaft of the telescope mount has a centring pin adapted for being connected to the threaded end of the right ascension shaft implemented (here) as a tubular shaft and for being fitted into the rotor of the angular velocity encoder, the rotor being also implemented as a hollow shaft, with the mechanical adapter also having a threaded retaining disc attaching the stator of the angular velocity encoder to the casing of the right ascension shaft, and with the adapter further having an external protective cover. The angular velocity encoder of the inventive apparatus is preferably a high- accuracy, high-resolution, analogue and/or digital-output magnetic or optical encoder having a configuration known per se.

The configuration and operation of the inventive apparatus will now be explained with reference to drawings illustrating specific embodiments, where Fig. 1.a is a semisectional view of the end portion of the right ascension shaft of a telescope mount, Figs. 1.b, Ic, 1.e are semisectional views of structural parts of a preferred embodiment of the inventive apparatus,

Fig. 1d is a semisectional view of a commercially available angular velocity encoder showing the mechanical configuration thereof,

Fig. 1f is an assembly drawing (shown in semisectional view) of elements depicted in previous drawings constituting the angular-position encoder attached to the end of the right ascension shaft utilising the inventive mechanical adapter, Fig. 2 is the schematic block diagram of a preferred embodiment - the so called sidereal variant - of the inventive apparatus,

Figs. 3.a and 3.b, respectively, show plots of angular velocity fluctuations measured on an actual telescope mount (EQ6) without the application of the inventive apparatus and with the apparatus of Fig. 2 connected; and

Fig. 4 is the schematic block diagram of a further preferred embodiment, the so-called "King-Rate" variant, of the apparatus according to the invention.

As it has already been mentioned, a high-accuracy, high-resolution angular velocity encoder 1 (momentary rpm sensor) is directly connected to the end of the right ascension shaft 4 pi the telescope mount in a strictly coaxial manner providing that the rotation of the encoder 1 is perfectly synchronised to the rotation of the right ascension shaft 4. A mechanical adapter ,,A", shown in component drawings Fig. 1.b, Fig 1.c, Fig 1.e, and assembly drawing Fig 1.f, is applied for connecting the encoder 1 to the telescope mount. The drawings illustrate embodiments adapted to be utilised with the commercially available EQ6-type (equatorial) telescope mount.

Fig. 1 shows a semisectional view of the end portion of the right ascension shaft 4 as it appears on an unchanged EQ6 mount. The rotary right ascension shaft 4 is supported by bearings in a stationary casing ,,H" (the bearing runner "cs" is also shown in the figure). The right ascension shaft 4 is implemented as a tubular shaft having a threaded end 4m. The threaded end 4m is intended for attaching a so called polar alignment scope applied for aligning the telescope with the rotational axis of the Earth before observations begin. Then the polar alignment scope can be removed as it is not required for either visual or photographic observations. The end of the right ascension shaft 4 is covered by a protection cap (not shown) retained by threading m2.

Fig. 1d shows the semisectional view of the commercially available angular velocity encoder 1 attached by the adapter ,,A" configured in accordance with our invention to the end of the right ascension shaft 4. The angular velocity encoder 1 consists of a rotor 1f implemented as a tubular shaft, and a stator 1a extending around the rotor. The encoder 1 also has a retaining tab ,,f and a retaining ring

,,gy" adapted for securing it to the telescope mount, (cable outlets for measurements signal wires are not shown in the drawing.) Fig. 1.c shows the centring pin A/1 of the mechanical adapter ,,A". The centring pin A/1 ensuresΛhat the right ascension shaft 4 and the angular velocity encoder 1 are connected directly, in a strictly coaxial manner. The centring pin A/1 has a threading ml at one end dimensioned such that it fits into the threaded end 4m of the right ascension shaft 4, while pin ,,t" disposed at the other end of the centring pin A/1 fits tightly into the tubular shaft shaped rotor 1f of the angular velocity encoder 1. (When manufacturing the mechanical adapter ,,A" it is of crucial importance to ensure that the threaded end ml and pin ,,t" of the centring pin A/1 are strictly coaxial, and it is also very important that the machined part has high dimensional stability.)

Fig. 1.b shows the semisectional view of the retaining disc A/2 of the mechanical adapter ,,A" of the inventive apparatus. A threading m2 (male) is disposed on the retaining disc A/2, which threading m2 is dimensioned to engage threading m2 (female) disposed on the casing ,,H" of the right ascension shaft 4 so as to be able to attach the⁷ retaining disc A/2 to the stationary casing ,,H". (As it has already been mentioned, threading m2 is originally intended for retaining the protection cap.) As it is seen in the assembly drawing Fig. 1.f, screws inserted during assembly into borings disposed on the front and side surfaces of the retaining disc A/2 are utilised for preventing the retaining disc A/2 from getting rotated after it is mounted on the casing ,,H", and also for retaining the retaining tab

,,f of the angular velocity encoder 1 (and therefore the stator 1a) and for oriented mounting of the protective cover A/3 shown in Fig. Ie.

Fig. 1.e shows a semisectional view of the protective cover A/3 of the mechanical adapter ,,A" according to the invention. The cover is applied for protecting the angular velocity encoder 1 and its retaining structural elements from dust, water and mechanical impact. Threading m3 (having the same dimension as the threading m2 of the right ascension shaft 4) is utilised for retaining the protection cap and comprises a cable outlet opening out for passing the signal output cables (outputs) of the angular velocity encoder 1 therethrough. Fig. If shows a semisectional view of the angular velocity encoder 1 mounted on the right ascension shaft 4 utilising the mechanical adapter ,,A" according to our invention. The above described elements are assembled in a way that should be straightforward after examining Figs 1.a-1.e. The assembly process is started by removing the above mentioned protection cap and polar alignment scope from the end of the right ascension shaft 4 that has threading m2. Next, the threaded end of the centring pin A/1 (having threading ml) is screwed into the threaded end 4m, and the retaining disc A/2 is attached to the end of casing ,,H" of the right ascension shaft 4. The angular velocity encoder 1 (more particularly, the rotor 1f thereof) is then pulled on the pin ,,t" of the centring pin A/1 , while the stator 1a of the encoder 1 is attached to the retaining disc A/2 by means of retaining tab ,,f" and screws. Finally the protective cover A/3 is placed over the already assembled elements and secured in an oriented manner utilising screws introduced into bores disposed on the front and side surfaces of the retaining disc A/2. (Before the protective cover is added it should be made sure that wires coming from the angular velocity encoder 1 are safely passed through the cable outlet opening out.)

It should be noted here that although the embodiment shown in Figs. 1.a-1.f illustrates a configuration designed to be used with a specific telescope mount type (EQ6), the angular velocity encoder 1 may be attached to other commercially available telescope mount types in an essentially same manner. Adaptation of the inventive solution to other telescope types is an undertaking easily accomplished by a person skilled in the art. The encoder (angular velocity or momentary rpm encoder) should be connected such that it always runs perfectly synchronised with the right ascension shaft and preferably that the connection may be made without altering the original telescope mount.

Fig. 2 shows a schematic block diagram of a preferred embodiment of the apparatus according to our invention. The apparatus shown in the drawing may be utilised for eliminating tracking errors of mechanical source arising in all commercially available mass produced telescope mounts that have an own telescope control system and/or autoguider control input. Utilising the inventive apparatus, these tracking errors (that is, errors caused by the mechanical inaccuracies present in the telescope mount) may be reduced to within a range of ± 1" (arc second). (Nowadays almost mass produced telescope mount types, even less expensive ones, fulfil this condition.)

As it is shown in Fig. 2, the telescope tube 2 is supported by the right ascension shaft 4 through a declination shaft 3 set perpendicularly thereto. The telescope tube 2 is moved by a driving motor 6 through a drive train 7. The driving motor 6 is operated by a drive control unit 5 connected to it. Driving motor 11 , also connected to the drive control unit 5, is applied for rotating the declination shaft 3 through a drive train 12. The drive trains 7, 12 typically apply a worm-wormwheel gear arrangement.

As it has been already explained referring to Figs. 1.a-1.f, an angular velocity encoder 1 is attached to the end of the right ascension shaft 4 directly (without any gears), in a strictly coaxial manner, utilising a mechanical adapter ,,A". According to a preferred embodiment the angular velocity encoder 1 is a high- resolution analogue-output magnetic encoder. The inventive control system utilises the two, sine and cosine-wave (90-degree phase shifted) signals appearing at the output KM of the encoder as process signals.

The output Ki1 of the angular velocity encoder 1 (momentary rpm encoder) is coupled to the process signal input Be1 of the interpolator signal processing unit 9 of a measurement and control unit 8 that is applied for multiplying the frequency of the signal of the angular velocity encoder 1. The output of the signal processing unit 9 is connected to one of the inputs of the microcontroller correlator unit 10 of the measurement and control unit 8. A quartz oscillator ,,Osc" is connected to the other input (the reference input ,,Be ref") of the microcontroller correlator unit 10. The output Ki2 of the microcontroller correlator unit 10 is coupled to the external control input aut.port of the telescope's own drive control unit 5, either directly or through an autoguider interface unit applied for providing required input signal levels. (The autoguider interface unit is not shown in Fig. 2.) The operation of the embodiment shown in Fig. 2 is described as follows: The applied high-quality magnetic (or optical) angular velocity encoder 1 emits 5000 pulses during- ⁷a single revolution of the right ascension shaft 4. This is multiplied by a factor of 256 in the interpolator signal processing unit 9 (to 1,280,000 pulses per revolution). Signals coming from the angular velocity encoder 1 also carry information on the direction of rotation. A single revolution is: 360° = 1 ,296,000" (arc-second); and thus the angular resolution is: 1 ,296,000 arc- seconds/1 ,280,000 pulses = 1.0125 arc-second/pulse. Celestial objects complete their apparent diurnal motion in 23 hours 54 minutes and 4 seconds (86,164 seconds), which gives the length of a sidereal day. In order to keep an object stationary in the telescope's field of view, the telescope should be rotated about its right ascension axis with a stabilised angular velocity corresponding to this value. The necessary "time base" (or base frequency) is thus 1 ,280,000 pulses/86, 164 sec = 14.855392 Hz. In case the pulse signal received from the right ascension shaft 4 has precisely this frequency, the rotational speed is accurate. One of the inputs of the microcontroller correlator unit 10 receives the frequency-multiplied signal of the angular velocity encoder 1 , while to the other, reference signal input ,,Be ref" the "time base" signal generated by the quartz oscillator ,,Osc" is fed. The two signals are compared fifteen times per second utilising the program stored in the microcontroller correlator unit 10. If the difference is outside the ± 1" range, a correction signal is sent (through output Ki2) to the external control input of the drive control unit 5, the correction signal being sustained until the error returns into the ± 1" range.

The embodiment of our invention shown in Fig. 2 is therefore capable of providing a constant-angular velocity drive matching the rotation of the Earth with an error in the range of + 1", utilising a simple apparatus that may be adapted for existing telescope mounts. In compliance with astronomy terminology, this embodiment is called the sidereal variant.

The software program of the microcontroller correlator unit 10 is designed such that the system is capable of deciding whether the angular velocity differs from the reference value because of drive errors or the difference is caused by a position change operation (a new object is being targeted by the so called GOTO function). In the latter case the difference is rapidly increasing, causing the system not to send a correction signal to the drive control unit 5. Fig. 3. a and 3.b shows the error diagram (angular velocity difference as a function of time) of the EQ6 mount according to our example. Fig. 3.a shows the error diagram of an original, uncompensated mount (as purchased in the store), whereas Fig. 3.b is the error diagram of a mount retrofitted with the inventive apparatus as described in relation to Fig. 2. The horizontal axis represents time T (seconds), while the vertical axis shows the angular velocity fluctuation in arc seconds. (The two diagrams are scaled differently.)

Looking at Fig. 3. a the following observations can be made: - errors caused by mechanical factors show periodicity; - cycle time is 8 minutes (480 s), the same as the cycle time of the worm of the final drive gear

- combined error is above 40";

- closer examination of individual cycles shows that differences caused by stochastic errors are superimposed on really periodic mechanical errors (no completely identical cycles can be found)

It follows from the above that in case errors were corrected applying the known "training" method, a relatively large error (in the range of 6-8") would still remain. Conversely, having a look at Fig. 3.b showing the error diagram of the inventive system it can be/maintained that

- periodicity has disappeared

- max. overall error only marginally exceeds 2".

Thus, measurement results indicate that the apparatus according to the invention operates in a way fully conforming to requirements set by the inventive objective. In case of such a small residual angular velocity fluctuation value (~2") the sidereal tracking error is unnoticeable, virtually disappearing inside the seeing disc.

However, as it has already been touched upon, under certain circumstances the observation errors caused by atmospheric refraction cannot be neglected. While the apparent movement of celestial objects resulting from the Earth's rotation may be completely compensated by rotating the right ascension shaft with a constant speed in a direction opposite to the rotation of the Earth, tracking the apparent movement caused by refraction requires the coordinated, varying-speed rotation of both the right ascension and declination shafts.

Consequently, - contrary to the sidereal variant - to compensate errors caused by refraction the right ascension shaft should be rotated with a varying speed depending on the position of the observed object in the sky, and also the declination shaft has to be rotated to a certain extent (the necessary correction values may be calculated exactly).

Thus, in addition fully compensating the errors arising from inaccuracies of the tracking mechanism of the telescope mount, the embodiment of the inventive apparatus shown in Fig. 4 is capable of compensating the "external tracking error" caused by atmospheric refraction, ensuring that the observed object is kept stationary in the field of view.

Structural elements identical to those shown in Fig. 2 are designated by the same reference numeralsin Fig. 4. As it is shown in Fig. 4, the output Ki1 of the angular velocity encoder 1 of the apparatus is connected to the process signal input Be1 of a measurement and control unit 13, more specifically to the process signal input Be1 of an interpolator signal processing unit 14 (similarly to the embodiment of Fig. 2).

In addition to performing closed-loop motion control for eliminating angular velocity fluctuations, in this embodiment the measurement and control unit 13 has to perform position control that is provided by a central computer unit 15 equipped with the necessary software programs.

The central computer unit 15 of the measurement and control unit 13 has one of its inputs connected to the output of the interpolator signal processing unit 14, while the other input "Be ref" (the reference input) thereof is connected to a quartz oscillator ,,Osc". In addition to that, the central computer unit 15 has a bidirectional data link (provided by a serial data transfer channel RS) to the drive control unit 5 of the telescope, more specifically to the data storage unit thereof, and according to a preferred embodiment it also has an autoguider input ,,Be aut.", to which an autoguider may be connected in specific cases. The output of the central computer unit 15 is connected through a level converter unit 16 to the external control input "aut.port" the drive control unit 5 of the telescope.

The apparatus of Fig. 4 is operated as follows:

- at the start of the observation the data necessary for refraction-corrected control, such as celestial equatorial coordinates of the target position, latitude of the observation location, local time data, and the gear ratio of the declination drive, are retrieved to the central computer unit 15 through the serial data transfer channel RS;

- in case the drive control unit 5 announces only the celestial equatorial coordinates of the object to be observed, missing data are retrieved from an external computer, also through the serial data transfer channel RS.

- using the retrieved data, the central computer unit 15 calculates in real time the horizontal coordinates of the targeted object and the necessary correction due to the effect of atmospheric refraction on the elevation of the object, and then the specific correction values for the RA and DEC coordinates are calculated by performing a mapping. (It should be noted here that calculations can be performed using continuous functions, see for instance Marek Miklόs, ed. Csillagaszat p87., Akademiai Kiadό 1989.)

- the calculated RA correction data are used by the central computer unit 15 to modify the (constant) angular velocity reference signal corresponding to the rotation of the Earth, the modified values being treated as a reference signal and compared with the output, signal of the angular velocity encoder 1 (the process signal), using the difference for establishing a correction signal for the drive control unit 5 (after the necessary level conversion),

- the calculated DEC correction data are applied for rotating the declination shaft in the necessary amount, also through the drive control unit 5.

In the embodiment shown in Fig. 4 the central computer unit 15 also has an autoguider input "Be aut" for optionally connecting an autoguider to the apparatus according to the invention. In this case refraction correction is not needed, because the autoguider signal carries all the information required for controlling the telescope mount, and therefore the signal received from the autoguider may be treated as an external reference for providing proper, refraction-corrected telescope control.

When an autoguider is connected, the apparatus is operated as follows:

- the connection of the autpguider is detected by the central computer unit 15,

- the central computer unit 15 receives the signals of the autoguider, which signals are subsequently treated as reference signals for controlling both the right ascension and declination shafts,

- between the reception of two successive autoguider signals the motion of the right ascension shaft is controlled using the (constant) reference signal provided by the quartz oscillator ,,Osc"., as it has already been discussed in relation to the sidereal variant shown in Fig. 2. Connection of an autoguider is not required but it simplifies the configuration of the apparatus while not being significantly costlier than the sidereal variant. By way of conclusion the most important advantages of the inventive apparatus are summarized below.

The apparatus may be added to any type of mass produced amateur and semi- professional equatorial telescope mounts as a standalone application both during manufacturing and as a retrofit. When applied as a retrofit, mechanical and electric connections may be made simply, without disassembling the original mount, in case of a wide range of commercially available telescope mounts. With the application of the inventive apparatus the accuracy of existing mediocre- quality mounts may be increased to a level exceeding even that of the best, most modern and most expensive telescope mounts.

Because the original drive mechanism and electronic motion control system of the telescope mount is left untouched during the retrofit, warranty is not voided. The required level of tracking accuracy may thereby be provided with a cost that is significantly lower than it would be if the same level was reached by increasing machining accuracy. *'

The inventive control system is capable of keeping the angular velocity of the right ascension shaft of almost any type of telescope mount (excepting only lowest quality mounts) within the range of two digits, that is, ± 1 arc second for a duration of 10 minutes. This level of tracking accuracy (within the above described limits) is sufficient for scientists performing the most demanding observation tasks.

With the application of the inventive apparatus the required ± 1 arc second accuracy may be achieved cooperating with the mount's own control system, without resorting to a guide star or any other external reference signal (even when the dome hatch is closed). The system corrects all mechanical drive errors in real time, and the control software is capable of detecting and using the telescope's targeting functions. No compatibility problems arise when the apparatus is connected to existing electronics. Also (under the above described circumstances) an autoguider camera may not be required, which makes astronomical observations simpler, more reliable, and faster. The King-Rate variant offers extra benefits, as it is capable of receiving and processing the autoguider control signal commonly used in today's telescope control systems. In case an autoguider signal is not present, the King-Rate variant provides accurate refraction-corrected motion control of the telescope mount to eliminate the effect of atmospheric refraction.

List of reference numerals

A adapter

A/1 centring pin

A/2 retaining disc

A/3 protective cover

1 angular velocity encoder

1a stator

1f rotor f retaining tab ml , m2, m3 threading out. cable outlet,dpening

2 telescope tube

3 declination shaft

4 right ascension shaft

CS bearing runner gy retaining ring t pin

4m threaded end

H casing

5 drive control unit

6, 11 driving motor

7, 12 drive train

8, 13 measurement and control unit

9, 14 signal processing unit

10 microcontroller correlator unit

Osc. quartz oscillator

15 central computer unit

16 level converter unit

Ki 1 output

Ki2 error signal output

Be1 process signal input Be ref reference signal input

Be aut autoguider input aut. port external control (autoguider) input

RS serial data transfer channel

Claims

1. Apparatus for eliminating angular velocity fluctuations of mass produced telescope mounts installed subsequently, where the telescope mount comprises a declination shaft (3) adapted for supporting a telescope tube (2), and a right ascension shaft (4) set perpendicular to the declination shaft (3), with driving motors (6, 11) having a drive control unit (5) being connected through drive trains (7, 12) to the declination shaft (3) and the right ascension shaft (4), and with the drive control unit (5) having an external control input (aut. port), characterised by that ,/ an angular velocity encoder (1) is attached by means of a mechanical adapter (A) in a direct coaxial manner to the right ascension shaft (4); where the output (Ki 1) of the angular velocity encoder (1) is connected to the process signal input (Be1) of an electronic measurement and control unit (8, 13) that has a quartz oscillator

(Osc), and where the error signal output (Ki2) of the electronic measurement and control unit (8, 13), providing the correction signal, is fed back to the external control input (aut. port) of the drive control unit (5) of the telescope.

2. The apparatus according to Claim 1, characterised by that the measurement and control unit (8) thereof consists of a interpolator signal processing unit (9) connected to the analogue output (KM) of the angular velocity encoder (1); a microcontroller correlator unit (10) connected to the output of the signal processing unit (9); and a quartz oscHlator (Osc.) connected to the reference signal input (Be ref) of the microcontroller correlator unit (10), where the output (Ki2) of the microcontroller correlator unit (10) is fed back to the external control input (aut. port) of the drive control unit (5) of the telescope.

3. The apparatus according to Claim 1 , characterised by that the output (KM) of the angular velocity encoder (1) is connected to the process signal input (Be1) of the measurement and control unit (13), with a central computer unit (15) being connected to the output of the signal processing unit (14); where a quartz oscillator (Osc.) is connected to the reference signal input (Be ref) of the central computer unit (15), with the central computer unit (15) having a bidirectional data transmission link to the drive control unit (5) of the telescope, and with the output of the central computer unit (15) being fed back to the external control input (aut. port) of the drive control' unit (5) of the telescope through a level converter unit (16).

4. The apparatus according to any one of Claims 1-3, characterised by that the mechanical adapter (A) applied for attaching the angular velocity encoder (1) directly to the right ascension shaft (4) of the telescope mount has a centring pin (A/1) adapted for being connected to the threaded end (4m) of the right ascension shaft (4) and for being fitted into the rotor (1f) of the angular velocity encoder (1), the rotor (1f) being implemented as a tubular shaft, where the mechanical adapter (A) also has a threaded retaining disc (A/2) attaching the stator (1a) of the angular velocity encoder (1) to the casing (H) of the right ascension shaft (4), and an external protective cover (A/3) releasably joined to the threaded retaining disc

(A/2).

5. The apparatus according to any one of Claims 1-4, characterised by that the angular velocity encoder (1) thereof is a high-accuracy, high-resolution, analogue and/or digital-output optical encoder having a configuration known per se.

6. The apparatus according to any one of Claims 1-5, characterised by that the angular velocity encoder (1) thereof is a high-accuracy, high-resolution, analogue and/or digital-output magnetic encoder having a configuration known per se.