2.5m Isaac Newton Telescope (INT)

General Description of the Telescope

The INT is an equatorially mounted (polar-disc/fork type) telescope with a 2.54 metre diameter f/2.94 primary mirror made of Zerodur (a ceramic/glass product of zero thermal expansion). There are two focal stations used from a possible three. These being a corrected f/3.29 Prime Focus and an f/15 Cassegrain focus. The f/50 Coude focus was never commissioned.

An equatorially mounted telescope only needs to track at sidereal rate in R.A. to maintain the object at the focal plane. The polar axis points to true north and is inclined at the latitude angle equal to that of the observatory (approx 28.75o N). n.b. In this document, the terms HA (hour angle) and RA (right ascension) I have used freely. They simply mean the movement of the telescope in the east - west direction; the meridian being the north - south centre line of the axis. Hour angle being the more correct term as this value changes with time.

The tube is of open construction with the optical components (primary and secondary mirrors) supported at the extremities of the upper and lower Serrurier trusses. These attach to the top end ring and the massive centre box section. The top end ring supports the secondary mirror assembly with its focus drive or the prime focus turntable.

When observing at the f/3.29 Prime focus, the secondary mirror assembly is removed and the Prime Focus Cone Unit (PFCU) with Wide Field Camera (WFC) fitted.

Fitted to the centre box section are the trunnions which form the declination axis. These are supported on gimbal mounted bearings on top of the fork tynes. This complete assembly is mounted onto the polar axis disc. The base section housing the polar disc bearing is mounted on a spherical load point at the north end of the north/south centre line. The polar axis disc is supported by hydrostatic pads.

At the bottom of the lower Serrurier trusses, the mirror cell is attached to the tube by four high tensile bolts. The mirror cell contains the 2.5m primary mirror along with the mirror support system and associated electronics. Below the mirror cell is the Cassegrain instrument rotator and cable wrap.

The Cassegrain instrumentation with electronics racks is fitted below the rotator. The normal configuration consisting of the Acquisition and Guider box (A&G box), the Intermediate Dispersion Spectrograph (IDS) and the Faint Object Spectrograph (FOS)  n.b. FOS now de-commissioned. These can be removed and a specialist instrument fitted if necessary. The A&G box is seldom removed.

The electronics for controlling the telescope is housed in racks (bays)  in an air conditioned room on the second floor at the west side of the building known as the CLIP centre (Control, Logic, Interlock and Power). The telescope control computer, a DEC ALPHA is located here as is the ING Time Service, Data Acquisition System computers (UDAS) and some of the older instrument controllers e.g. The PFCU MMS and integrating TV system (GRINNELL).

For a detailed description of the INT refer to:   The INT Operation and Maintenance Manual Vol. 1

The total weight supported by the hydrostatic bearings (the Polar axis disc), is made up of the individual weights of each telescope component. These being:
Secondary mirror assembly 1000kg
Prime Focus Cone Unit 1000kg
Top end ring 9000kg
Telescope Tube 2000kg
Telescope Tynes  28000kg 
Centre box section  25000kg
Mirror cell 9000kg
Mirror 4361kg
Rotator and cable wrap  6000kg
Acquisition Guide Box 1250kg
IDS instrument 2250kg
FOS instrument 250kg
Total weight  ~ 90  metric tonnes



Top end ring

The prime focus top end ring is suspended centrally in the tube on four pairs of struts known as the spider vanes and can contain either the secondary mirror assembly when observing at the Cassegrain focus or the prime focus imager (WFC + PFCU). Precision V blocks plates are fitted to the top end ring tube to ensure that the optical axis alignment is maintained when changing between the secondary mirror or prime focus imager.

Associated drawings:  TB1 201  TB1 286

Secondary mirror

The secondary mirror cell contains a central spigot which passes through the centre of the mirror enabling it to be held both axially and radially. A system of counterbalanced levers control the axial deflection at the outside edge. The cell is mounted on a barrel which can move axially between rollers.

A 6A stepper motor driving a lead screw through a slipping clutch and gear train adjusts the focus position. Limit switches control the overall range of movement. The motor is controlled by a stepper motor drive board in the CLIP centre. If a limit is reached when the telescope is in Computer Mode, Engineering Mode will be returned and it will be necessary to drive the mechanism in the opposite direction to clear the limit before computer control can be returned. The position of the secondary mirror (or prime focus camera) is measured using absolute and incremental piston transducers

Originally the secondary mirror had a double petal mirror cover system, but this proved troublesome and was removed many years ago, however, the motors are still fitted to the tube.

Associated drawings:  TB1 291

Telescope tube

The telescope tube is an open Serrurier truss design with a rectangular centre box section to which the declination bearing trunnions are attached. The lower trusses are closed by the main mirror cell, the upper trusses by a fixed end ring which supports the prime focus structure. The main mirror petals are suspended below the centre box section on A-frames which support the baseplate carrying the petals and drive assembly.

The Serrurier trusses have been designed to provide equal and parallel deflections of the top end ring and the mirror cell at all altitude angles. The trusses are tubular steel sections with spade ends which are attached to the centre box section with brackets mounted on the top end ring. A cable duct runs along one of the truss tubes carrying cables, fibre optics and compressed air to the prime focus connector panel.

Associated drawings:  TB1 201

Centre Box section

The centre box section (or CUBE) is a square fabrication of four box girder sections. Bearing trunnions are fitted to two opposite sides of the cube to form the Declination axis. The Serrurier trusses are secured within the centre box section. Extra supporting brackets have been clamped around the trusses and secured to the top of the centre box section to reduce an inherited flexure component when the telescope is moved away from the zenith.

Primary Mirror Cell

The primary mirror cell is a steel fabrication consisting of a flat disc with a box section cylindrical surrounding wall. The mirror cell is bolted to the lower trusses by four equally spaced bolts, the shear being taken by angled wedges at each of the four cardinal positions. Stiffening the assembly is achieved by radial webs. Below the cell is an extension piece which carries the Cassegrain instrument turntable and the cable wrap.

Associated drawings:  TB1 220

Primary Mirror covers

The primary mirror cover is built up of sixteen petals made from an aluminium honeycomb material for strength and lightness. The drive is through an endless chain driven from two 3 phase motors with worm-boxes. Each petal contains a slipping clutch so that on opening or closing, each petal is driven to a fixed stop position.

The mirror covers are controlled from two push buttons mounted on the services panel of the Engineering rack in the control room. Limit switches operating relays sense the OPEN and CLOSED condition. Indicator lamps within the push buttons show when the petals are fully open or closed. The power is then removed from the motors.

Mirror status is also sent to a CAMAC PR2402 input register (Address: B4 C3 N12 A1 - Bits 6 and 7) .

The contactors for driving the mirror cover motors and the mirror cover relay control circuit board are now located in the middle section of the Mains Distribution Cabinet in the CLIP centre.

n.b. This board is a new fabrication which ONLY includes the primary mirror cover relays. As the secondary mirror cover has been removed, the relays haven't been included. However, should there be a problem, the old relay board will work in this position.

Associated drawings:  TB1 260  ED020

Mirror support system

n.b. It should be noted that the original mirror support system using an airbag and bi-state controller were replaced (circa 1996) with a WHT type system. Thus much of the information in the Grubb Parsons manuals is now obsolete. For more information, refer to the manual: WHT/INT MIRROR SUPPORT SYSTEM

The mirror cell floor has a machined surface where 36 pneumatic pads; known as Belloframs, are fitted in three concentric rings for supporting the mirror. The belloframs are divided into 3 sectors of 120o with each sector being controlled by a JOUCOMATIC servo control valve. These valves control the air pressure (input supply = 100psi) to the twelve belloframs in each sector ensuring that the mirror alignment to the optical axis stays correct.

Three load cells (axial definers) measure the compressive/tensive forces acting on each sector and send signals back to the Mirror Support PID electronics crate for processing. This drives the JOUCOMATIC valves so forming a servo loop. A radial defining system consisting of counterweights and cranks is used to balance the mirror and to prevent it from twisting. The components of the mirror support system are:

Pneumatic Pads (Belloframs)

Each pad consists of a cast base with a gas (compressed air or N2) inlet connection. A diaphragm sleeve is clamped to the base with a V clamp ring. A rolling diaphragm is held between these two parts with its central horizontal surface bonded to the support pad piston. As the air pressure in the control system varies, the piston can rise or fall on the rolling diaphragm to apply the required supporting forces to the mirror. There are also rest pads fitted in the base of the cell which support the mirror when the pneumatic system is not pressurised.

Axial Definer system

Three of these units; consisting of an adjustable rod and load cell  (one for each sector) are fitted between the cell floor and brackets cemented to the mirror's edge. In relation to the north/south alignment of the telescope, these are mounted at the N, SE and SW positions.

These are used to maintain the correct alignment of the mirror to the optical axis of the telescope at any attitude and thus set the working height for the mirror.

Adjacent to each axial definer unit are dial gauges which are used to measure the height of the mirror within the cell. It is necessary to read these before and after mirror re-aluminisation to ensure that the mirror has returned to its correct operating position.

Associated drawings:  TB1 222  TB1 227

Radial Definer system

This weight relief system consists of 36 counterweights with cranks fitted around the mirror cell. The cranks pivot in brackets cemented to the circumference of the mirror. These provide the required tangential supporting force for the mirror

The radial position of the mirror within the cell is fixed by three radial (transverse) adjustable defining rods each fitted with a load cell. Like the axial definers, these are designed to ensure that only pure compressive loads are measured. These are mounted at the NE, NW and S positions.

The radial position of the mirror is monitored at four points using dial gauges. These can be withdrawn from the mirror's edge when the mirror is removed for re-aluminising. As with the axial definers, these must be read before and after mirror re-aluminisation to ensure correct alignment.

Associated drawings:  TB1 221  TB1 223

Mirror Support PID Control

The electronics for the mirror support system are housed in a 3U crate mounted externally on the mirror cell. This is a PID (Proportional - Integral - Derivative) system and consists of the modules: The loads on the AXIAL and RADIAL load cells are indicated on panel meters near the bottom of the engineering control rack. (Drawing: ED026). Overtravel switches in the cell operate if forces in excess of 100lbf are measured. An alarm indicator on the control desk illuminates of this happens.

n.b. When the new mirror support system was installed, the two boards in Card Crate 6:

(which controlled the old airbag system)  were modified to work with the PID system. Although some of the old circuitry is used, the above ED drawings as shown in the Grubb Parsons file are not correct. Refer to the manual: WHT/INT MIRROR SUPPORT SYSTEM kept in the electronics workshop which contains all the new drawings and detailed instructions for setting up the mirror support system.

Cassegrain Turntable

The turntable below the mirror cell provides mounting for the Acquisition and Guidance unit (A&G Box) and the Intermediate Dispersion Spectrograph (IDS) combined with the Faint Object Spectrograph (FOS).  The A&G box is seldom removed and a non-common user  instrument can be bolted to this. The purpose of the turntable is simply to enable the SLIT of the spectrograph to be rotated in relation to the object under observation.

The rear face of the mirror cell provides the mounting for both the Cassegrain turntable and it's cable wrap. The turntable is in the form of a ring supported on a 1219mm diameter wire race slewing ring and is driven by a single motor/gearbox unit with an anti-backlash drive. The pinion engages with a spur gear cut into the circumference of the table.

The motor is a DC torque motor type: G19M4 fitted with a tacho-generator to provide velocity feedback and a solenoid operated brake. An idler shaft with a pair of pinions mesh with the turntable spur gear and drive a 36000 count 5 digit BCD absolute encoder giving a possible 360o position indication. A limit switch limits the travel of the table to +/- 200o .  This constraint is due to the parameters of the cable wrap.

Associated drawings:  TB1 272

Cassegrain cable wrap

The cable wrap mechanism consists of two loops of KABELSCHLEPP plastic drag chain which surround the turntable. This is driven by a pin from the turntable. Each loop passes around a roller mounted on brass bushes in the open ends of an almost annular aluminium channel section known as the banana. Cables are fed in from the mirror cell connector plate via two slots into the ends of the Kabelschlepp. To prevent bunching, the cables run in compartments formed by spacers fitted between rollers on the links. On exit from the wrap, the cables pass through two slots to the Cassegrain connector plate.

The cable wrap cable population is a important factor for trouble free operation. This as been achieved by having the same number of cables (and of a similar diameter) in both chains. The cables are clamped at each end of the chains. This is important because if the cables stretch or pull through the clamps, hernias can form and damage the cables.

The electronic cubicles for the Cassegrain instruments are mounted on the underside of the cable wrap ring.

Associated drawings:  TB1 274

Polar Disc

The Polar disc with two tynes, extension cone and hub form the polar axis. It is supported axially by five hydrostatic pads and radially by three hydrostatic pads. The axial pads consist of three fixed height units and two bellows units. The radial pads comprise of two fixed height and one bellows unit.

Associated drawings:  TB1 120  TB1 130

Hydraulic system

The hydraulic plant consists of two assemblies:
  1. The power unit consisting of the oil tank, pump and the pressure switch located on the ground floor.
  2. The accumulators providing the required working pressures are located on the second floor below the north pier.
When the hydraulic plant is running, oil is pumped through small pipes via jet blocks to the 3 radial pads and 5 axial pads which control the flow of oil to support the polar disc at approximately 0.1mm above the pads. The oil drains into a trough and then returns to the main tank.

Two types of accumulators are used. One maintains a set oil pressure in the bellows pads and also serves as a buffer. This effectively takes out pulsations produced by the oil pump which could be transmitted through to the telescope.  The larger accumulator is connected to the main feed to maintain oil pressure to the polar disc for several minutes in the event of oil pressure failure. This allows the telescope to come safely to rest.

The oil pump is switched on from the engineering control rack, but can be disabled by changeover switches in the oil plant room on the ground floor.

On each pad there are a series of 4 microswitches. When the hydraulic plant  is running, these sense when the polar disc is raised off the pads. n.b. In practice, due to hysterisis in the microswitches, these are very difficult to adjust and its not usual to see several oil pad alarm lights illuminated on the control desk even when the working oil pressure has been reached. This is not a problem

The important alarm light is oil pressure. Failure of the oil pressure switch contacts not closing will cause a break in the alarm and interlock chain and will prevent the telescope being driven in HA both in computer and engineering mode. However, for engineering purposes, the telescope can be moved in DEC quick and slow motion without the oil pump running.

Associated drawings:  TB1 180  to  TB1 189


When observing at prime focus, motorised counterweights on the N and S sides of the cube are used to compensate for the extra weight of the WFC.  These are run down to their lowest position when the WFC is being used. The weight boxes are driven by lead-screws using a motor/gearbox unit. A handset is plugged into the CUBE control box and can be switched to drive the weights either seperately or together.

Fine balancing of the CASS turntable is achieved by loading weight plates into slots below the mirror cell or with weights added or removed from racks mounted on the cable wrap ring.

Cable runs

Cables from the Cassegrain turntable after passing through the cable wrap are terminated in the first instance on a connector panel mounted above the mirror cell. This allows the removal of the mirror cell from the tube e.g. When re-aluminising the mirror. The important ones being: Cables which terminate on the Cass connector panel, but which don't pass through the wrap are: and those to prime focus (the important ones) which pass through a hole in the cube and enter a cable tray on a truss tube: All the above cables and services pass into the cube through the hollow DEC axis bearing (drive side). Movement of the telescope in declination is accommodated by alteration of the depth of the cabling which occupies a compartment within the cube. Other cables that pass through the DEC bearing into the cube are: Cables that go up the DEC tyne, but do NOT pass through the bearing enter under the DEC clamp cover. These being: All these cables pass through the hollow polar axis where they are clamped and then enter five flexible tubes called the elephant trunks. The other end of the trunks are fixed under the floor behind the telescope polar disc area. Each trunk is long enough to allow the total permitted +/- movement of the HA axis without exerting strain on the cabling (which could cause friction when the telescope moves in HA).

Cables that don't pass through the elephant trunks serve the HA (RA) drive of the telescope:

All the cables run as a group under the raised observing floor (now covered in plywood paneling), passing into the CONTROL ROOM near the west end. They then drop through a hole under the raised floor into the CLIP centre Most are terminated in Bays 5 and 6 on numbered tagstrip blocks.

n.b. Cables that go to the control desk and engineering rack come back up from the CLIP centre. There is very little in the way of cabling coming from the telescope that terminates in the control room. Just some of the groups of fibre optic cables and a couple of alpha cables used for RS232 links. These can be found under the raised floor below the observers working area at the west end of the control room.

Associated drawings:  TB1 120  TB1 141  TB1 146  TB1 153  TB1 155  TB1 156  TB1 157  TB1 220

Declination and Hour Angle (RA) drives

Two systems are used for driving the telescope in DEC and HA :
  1. A Quick Motion drive for fast slewing the telescope
  2. A Slow Motion drive to enable the telescope to reach its final position and to enable the telescope to track (in RA) once the target is acquired.
A clamping arrangement allows the telescope to disconnect from the SM drives during fast slewing.

DEC and HA Quick Motion drives

QM drive is generated through a pinion driving a large spur gear fitted to the DEC and Polar axis. A Brown Boveri  M26 printed armature motor with integral gearbox is used. A tachogenerator within the motor provides velocity feedback.

When the telescope is tracking, the HA QM motor provides an anti-backlash torque to keep the worm and wheel in constant mesh by applying a force in the opposite direction to the rotation of the wormwheel.

On the DEC axis, the QM motor is powered on to maintain a constant load (anti-compliance torque) to the DEC SM tangent arm.

Associated drawings:  TB1 143  TB1 153

DEC Slow Motion drive

The SM drive is generated through a tangent arm (known as the banjo). An Inland (T-7203-C) torque motor with integral tacho-generator to provide velocity feedback drives a recirculating ball screw and anti-backlash nut connected to the tangent arm via a lead-screw, but only over a limited travel. This being  +/- 2.5o from the centre of travel.

There are limit switches and slotted opto switches (operated by a metal strip mounted on the anti-backlash nut) that determine the position of the tangent arm. These being:

The pre-limits (which operate about 0.5o before the final limits) are fleeting alarms which will sound the audible alarm, but will not trip the drive. They will clear if the telescope is driven away from them.

If a final limit is reached in computer mode, engineering mode will be forced. It will then be necessary to turn the LIMIT keyswitch to the O/R (override) position and drive the telescope out of the limit manually. Once the limits are cleared, the DEC SM drive will automatically re-centre itself.

n.b. It should be noted that once the demanded position is reached, the DEC QM and SM drives to all intents and purposes are held stationary to keep the DEC axis rigid. This is done by keeping the motors powered up. The only time the DEC SM drive operates is when the telescope is tracking and autoguider corrections are needed.

A hollow shaft Baldwin incremental encoder is mounted on the lead-screw shaft. This sends positional data back to the TCS via CAMAC. However, since the DEC axis was fitted with a Heidenhain tape encoder, the Baldwin is no longer read. See:  Heidenhain encoder

Associated drawings:  TB1 141-02  TB1 154  TB1 155  TB1 155-02

HA Slow  Motion Drive

The HA slow motion drive also uses an Inland (T-7203-C) torque motor with integral tacho-generator which drives the worm shaft. A hollow shaft BALDWIN incremental encoder is connected to the worm shaft. This sends positional data back to the TCS via CAMAC.

Once the demanded position is obtained and the SM clamps are applied, the HA SM motor turns the wormwheel at sidereal tracking rate to keep the object under study on the focal plane. Autoguider corrections are sent to the HA servo by either speeding up or slowing down the SM motor.

HA limits are fitted each side approximately 100o from the mid position. The pre-limit switches operate 7o before the final limits are reached.  The telescope should NEVER be driven pass the pre-limits.

Associated drawings:  TB1 141  TB1 141-02

DEC and HA Quick and Slow Motion Clamps

The DEC quick and slow motion clamps are contained within a housing along with the main drive spur gear for the DEC axis on the west side of the telescope. The QM clamps are located under the large cover on the outside of the DEC axis. The SM clamps are under split covers on the opposite side of the spur gear  i.e. Facing towards the telescope tube.

n.b.  To get access to the DEC clamp motors or microswitches, the large cover needs to be removed. Mechanical support staff should be called as the cover is heavy and requires the use of the dome crane.

The HA quick and slow motion clamps are located in the polar axis disc and requires removing cover plates to get access to them.

Two groups of shoes fitted with friction pads, each driven by an endless chain and 3 phase motors make up the QM and SM clamp assemblies. The clamps are spring loaded and work off a cam driven by the chain. The chain is always driven in the same direction. Microswitches operated by a seperate cam determine whether the clamps are engaged or are free.

The QM clamps are for engineering use only and are always engaged. These are controlled by the CLAMP keyswitch on the engineering rack. This puts the telescope in AXIS FREE mode so that it can be moved by hand for maintenance. The axis free indicator will illuminate if the QM clamps are released.  The telescope MUST BE IN BALANCE before the QM clamps are operated.

The SM clamps will be dis-engaged when the telescope is fast slewing and engaged when the telescope is at the demanded position and is tracking. Sending the telescope to zenith park will also engage the SM clamps when the telescope has stopped.

Before the SM clamps are engaged or dis-engaged, a series of conditions needs to be meet. Logic circuits with relays fitted in the CLAMP LOGIC tray in Bay 2 in the CLIP centre control the clamp action.  See the Clamp Logic  for more information.

The 3 phase contactors for both the DEC and HA QM and SM clamp motors can be found in the middle section of the Mains Distribution Cabinet.

Associated drawings:  TB1 144  TB1 156  TB1 146-02  TB1 156-02

DEC and HA SM Clamp Diagnostics

As an aid to diagnosing clamp problems, close to the contactors are two DIN rail mounted boxes each fitted with a pair of LED's. These show the state of the SM clamps and switch alternately as the clamps change between ENAGED and DISENGAGED states. The critera being:  That ONLY one of the LED pair will be illuminated. If either:
  1. Both LED's are ON or
  2. Both LED's are OFF
There is a problem! The most likely cause being the cam operated microswitches which break every few years. There are spares in the INT electronics store.

The INT servo drive system


In computer mode, as in the case of the WHT and JKT, the INT is controlled by a DEC ALPHA computer running VAX VMS. The software controlling this is called the Telescope Control System (TCS). CAMAC  (Computer Automated Measurement And Control) is used to interface the control systems such as the positional data from the DEC and HA encoders and to drive the servo amplifiers through a DAC.

Other systems read or written to the TCS via CAMAC include the time service, dome drive and position, focus and turntable control. Autoguider corrections are processed using a VME system connected directly to the DEC ALPHA computer. A engineering (manual)  mode is also available for driving the telescope, dome and other systems.

Engineering mode

Manual controls for the telescope are located on the engineering control rack. The telescope axes can be selected for Quick Motion or Slow Motion operation by  push buttons. The QM or SM drive rate is derived from potentiometers adjacent to the push buttons. In most cases, these will be at their maximum settings since the inputs to the pre-amplifiers ramp up to the voltage set by the rate potentiometers. This ensures a slow velocity build up to the telescope motors. HA and DEC +/- direction buttons then direct the demanded rate (voltage) to the servo electronics.

Quick motion (HA and DEC) servo

In manual mode, on selection of QM, the SM clamps will disengage telescope movement from the slow motion drive. The DC voltage set by the QM rate potentiometer is compared with the voltage from the QM motor's tacho-generator. The comparision is made in the QM pre-amplifier circuit boards. These boards being:

RA:  EC3   DEC:  EC5  (in Card Crate 4).

The error (or difference) voltage generated from the comparison is then taken to the QM power amplifiers. The tacho voltage backs off the input demand until the required velocity is achieved.  The tacho signal is also used to disable the SM clamps from being engaged until the telescope has come to rest.

Slow motion (HA and DEC) servo

Pushing the SM button will engage the SM clamps which will lock the telescope axis to the SM drive. At the same time this will disable the QM drive.

The demand signal after buffering is passed through an active notch filter which is tuned to the computer loop sampling frequency of 20Hz. This prevents mechanical resonance of the telescope. It is then amplified and compared to the tacho signal in the SM  pre-amplifier boards. These being:

(RA:  EC1   DEC:  EC4  (in Card Crate 4)

The error signal generated from the comparison is taken to the SM power amplifiers. The tacho voltage backs off the input demand until the required velocity is achieved. As the operating conditions of speed and load on the worm/wheel encompasses an area in which slip-stick may be anticipated, an additional dither signal is introduced into the servo.

There is only one difference between the HA and DEC slow motion pre-amplifiers. This is the facility to centre the DEC slow motion drive.. This is achieved by micro switches which are connected to the clamp logic electronics. According to whether the lead screw is left or right of centre, a  + or - voltage is applied to the DEC SM pre-amplifier which will drive the lead screw until the DEC CENTRED switched is activated.

Computer Mode

Providing the alarm and interlock logic is in the correct state, turning the keyswitch from engineering to computer mode (and pushing the green reset button), control will be passed to the TCS if its running.

The CAMAC system is clocked at 20Hz (50mS). During this period, the TCS will have read the time service (UTC), have converted UTC to Local Sidereal Time (LST), will have read the HA and DEC encoder values and computed the velocity demand value for the DAC. Other tasks such as moving the dome to compensate for telescope position are also performed during this period.

In computer mode, two servo loops are generated:

Position loop

The INT uses a combination of encoders; these being: which the TCS reads to determine the required velocity demand signal. The data from the tape encoders are averaged to reduce errors due to ovality of the tape.

When co-ordinates have been entered for RA and DEC, the TCS calculates how far the telescope needs to move and at what velocity. The SM clamps on both axes are released and the telescope will drive to reduce the error between the DEMANDED and the ACTUAL  position co-ordinates. The greater the difference is between them,  the faster the telescope will move.

This is achieved by the TCS reading (at 20Hz) the Ferranti encoder (HA) and Heidenhain encoder (DEC) during fast slewing until the error between actual and demanded has reached a low level (minutes of arc). When this is achieved, the TCS will switch from QM to SM drive. The SM clamps are engaged and the Baldwin encoder (now HA axis only) brought into the calculation to increase the positional accuracy to within seconds of arc.

n.b. It should be noted that the HA and DEC demand rates are chosen so that when the telescope axes are driven in QM, they arrive close to the demanded position at the same time.  The change over from QM to SM drive in HA and DEC always happens simultaneously.

Computer generated levels are sent to a CAMAC output driver module (Address: B4 C3 N15 A0 ) to engage or dis-engage the SM clamps. These signals are sent to the CLAMP LOGIC circuits which in conjunction with the state of relays in the ALARM and INTERLOCK crate decide whether the SM clamps can be operated safety.

The TCS will drive the telescope to constantly reduce the error between the demanded and the actual position co-ordinates. When tracking, the position error on the TCS Info Display will be very low and usually show 00:00:00.

Velocity loop

A digital word (the DEMAND value) generated by the TCS is sent to the servo pre-amplifiers via a CAMAC digital to analogue converter module (DAC). This samples at 12bits.  (Address: B4 C3 N14)  The channels being:
  1. A0  HA slow motion drive rate
  2. A1  HA quick motion drive rate
  3. A2  DEC slow motion drive rate
  4. A3  DEC quick motion drive rate
The DAC is set for Bi-polar mode operation; the output levels being: +/- 5V. A value of 2048 being equal to 0V  i.e. No demand being sent to the servo amplifiers. Once tracking is achieved, the HA slow motion axis will drive at sidereal rate (a velocity demand of  ~ 1990 ) being sent to the DAC A0 channel.

The tacho signal is used to maintain the velocity and to switch in the QM motors which act as anti-backlash devices when the demanded position has been reached.  These motors work by applying a light force in the opposite direction to which the telescope is tracking. This occurs when the SM clamps have been re-engaged and the demanded position reached. The TCS Info Display will then indicate that the telescope is tracking.

Position Encoding

There are 3 types of encoders used on the telescope. All are INCREMENTAL types. These being:
  1. Ferranti  grating tape encoder (HA axis)
  2. Heidenhain  tape encoder (DEC axis)
  3. Baldwin optical encoder (slow motion HA and DEC axes)

Ferranti grating tape encoder

The Ferranti grating tape encoder is used as the COARSE encoder on the HA axis only. It comprises of a stainless steel tape wrapped around the edge of the polar axis disc. The tape has a line spacing of  2.5 lines/mm. The reading head optics convert this to 5 lines/mm and work using the Moire Fringe Effect.

Moire Fringes are formed when a section of the grating  known as the index is superimposed on a scale grating of identical structure, but with two sets of lines at a slight angle. A collimated beam produced from a IR emitting GaAs chip at the focus of the lens projects the index line structure onto the scale. Fine adjustments of the angle between the index and the scale gratings enables the phase of the signals to be set accurately to 90o . Any distortion in the grating or curvature of the tape will cause the phase to deviate from the quadrature condition.

There are three reading heads mounted 120o around the tape to compensate for disc (tape) ovality. Each reading head produces two sinusoidal signals with a 90o phase shift between them. Depending on which one is leading the other by 90o, this will be the lead pulse, the other the lag pulse.

The signals from each head are sent to the Ferranti processing modules in the CLIP centre. Here they are converted into two TTL pulse trains, but still retaining their phase shift. The lead pulse will cause the modules to count up, the lag pulse to count down.

The pulses from the three processing modules are then sent to individual CAMAC RGO32Bit up/down counters (Address: B4 C2 N 2-5-8) where the serial data is converted to a parallel binary up/down value which the TCS can read.

Heidenhain  tape encoder

This replaced the older Ferranti encoder system on the DEC axis (circa 1995). It was done to improve telescope pointing and can resolve angular position to a much higher accuracy than the Ferranti tape. The tape rulings are extremely fine and cannot be resolved by eye.

Note: Fitting this system was a success, so good in fact that the DEC Baldwin encoder is no longer read by the TCS. The reason being that encoding directly from the DEC axis removes the error that was always present due to the Baldwin being fitted to the slow motion mechanism. It was planned to replace the HA Ferranti tape also, but this was never done due to cost!

The Heidenhain system uses two heads to reduce error due to ovality in the tape. The output from the heads are TTL quadrature signals that go directly to two CAMAC RGO32BIT up/down counters (Address: B4 C2 N 11-14)

Baldwin optical encoders

A Baldwin optical incremental encoder is mounted on each slow motion motor drive shaft. These are used to improve the positional accuracy which the Ferranti tape encoder cannot achieve. When the TCS reads the encoders, the value obtained from the Baldwin encoder is effectively combined with the averaged tape encoder value to improve the resolution.

IMPORTANT:  The HA axis will NEVER get into tracking mode as the position error generated will be too high without a working Baldwin encoder, but this is NOT the case for the DEC axis which uses the Heidenhain tape encoder and does not require the Baldwin.

The Baldwin optical incremental encoder has two quadrature outputs of 12000 counts per revolution. There is also a 1 pulse per rev signal generated.  The output signals go first to line driver boxes mounted close to the encoder and then on to CAMAC RGO32Bit up/down counters (Address: B4 C3 N 3-6).

These modules multiply the number of pulses by a factor of four making the total number per revolution to 48000 counts. One Baldwin bit is equivalent to 0.01875 arc seconds in RA and 0.01 arc seconds in DEC.

Note:  The Baldwin encoders have long been replaced with versions that use an LED for a light source. So the ominous task of replacing encoder lamps is no longer neccessary!

Encoder problems

If one of the Ferranti or Heidenhain reading heads (or a RGO32bit counter) develops a fault,  the telescope pointing will be completely wrong and it position unknown. In a severe case (high order bits missing) this could cause the telescope to shoot off until a limit or horizon switch is activated. If the position error is too high, the SM clamps will NOT engage .

Oil and grime on the tapes can also cause problems. Before suspecting a faulty Ferranti or Heidenhain encoder reading head, the tape should be cleaned.

Cleaning the Ferranti tape

The Ferranti grating tape comprises a steel tape wrapped around the edge of the polar disc. It is recommended that this tape is cleaned every 2 years, due to dust and grease accumulating and therfore the read out heads loosing pulses. There are 3 read out heads seperated a 120 degrees.  To get access to the tape , the read out head has to be removed, undoing the nut on the spring loaded bolt. Take th bolt away. The head can now be removed, by pushing the unit agains the 2 spring loaded bolts on the front and tilting the unit of the track. The positioning won't be lost, as this is defined by the metal bar.

You now have access to the tape for cleaning and inspection. You need a second person in the control room to move the telescope while cleaning the tape with propanol. The telescope has enough movement both ways to clean the tape, first removing the head mounted at 300 degrees, and then repeat the procedure after removing the head at 60 degrees. This will avoid removing the head mounted at 180 degrees, to which the access is much more difficult.

In the case of HA, (or if tracking cannot be achieved) a faulty Baldwin encoder or its RGO32Bit CAMAC module could be the problem. n.b. As the Baldwin encoders now use an LED light source, it is more likely to be a faulty RGO32Bit counter than the encoder itself.

Zeroset Module

This consists of a proximity sensor which detects the position of a metal target mounted at the zenith (DEC) and  meridian (HA) positions on the telescope. The output signal from the sensor is fed to a zeroset module that gives a output pulse which can only occur when a narrow datum window is active. This window can be shifted electronically +/- relative to the target and also the zeroset pulse width adjusted. Using this system, a pulse is generated which serves as a very accurate position datum.

To zeroset (clear) all the encoder RGO32bit counters at startup; in the case of the Baldwins, this pulse is logically ANDED in the line driver boxes with the signals:  lead, lag and 1pulse per rev. The datum pulse is also used to clear the registers of the Ferranti and Heidenhain RGO32bit counters this being applied to the zero ref input of the module.

Refer to (Drawing: ED079) which shows the general arrangement. The zeroset module setup is critical to set up.  Refer to the technical folder on the zeroset module if it needs to be re-calibrated.

Associated drawings:  BB1 006

Alarm and Interlocks

This consists of an interlock chain of relay contacts in series which must all be closed to pass +24V through to the CLAMP LOGIC tray. The relays associated with this are actived by the: These relays are mounted on wire-wrapped boards in Card Crate 5 in the CLIP centre (Drawing: ED033 3 sheets). Operation of  these alarms will return the telescope to engineering  mode if computer mode was selected. The status of these alarms are shown by indicator lamps on the control desk and if activated, an audible alarm will sound. Although the audible alarm can be muted (ACCEPT button), they must be cleared.

Alarm status is also read by a CAMAC input register (address: B4 C3 N13 A0) and is shown on page 3 of the TCS Info Display window.

In the case of the telescope moving into a limit switch, the spring biased LIMIT keyswitch needs to be turned to O/R (override) and held whilst the telescope is driven out from the limit. Once the alarm is cleared, the keyswitch can be released and normal operation resumed.

The HORIZON LIMITS are a special case and stop the telescope (at any attitude) if the tube is getting close to the horizon (about 5o). When moving the telescope to ACCESS PARK, the horizon alarm will operate. As the telescope needs to move downwards further to gain access to the WFC when working at prime focus, operating the LIMIT keyswitch and re-selecting QM DEC will allow the telescope to move down to its final limit.  When doing top end instrument changes, a clamp on the dome balcony operated by a hand wheel locks the tube ring when the telescope is in an out-of-balance condition.

The alarms below are fleeting alarms which simply sound an audible warning and illuminate an indicator. These do not trip the servo power amplifiers. Pushing the ACCEPT button mutes the alarm.

Tie bar

When an instrument is changed at the Cassegrain focus or the mirror cell removed, a TIE prop is swung out and inserted to lock the DEC axis at the zenith position.

Note:  Although there is a tie bar alarm relay, the contacts in the interlock chain have been shorted across for reasons unknown (at least to me!).  In engineering mode, it is possible to move the telescope in DEC without the oil pump running and the tie prop inserted.

Do NOT attempt to move the telescope when tied. Serious mechanical damage may occur and the power amplifiers can burn out due to excessive currents being drawn.

Associated drawings:  TB1 541  BB1 124 to BB1 130

Clamp Logic tray

This is located at the top of Bay 2 in the CLIP centre. It consists of 3 logic boards and a series of plug in relays. The +24V relay supply for the tray comes in on TS85 from the ALARM and INTERLOCK crate. If an alarm is active, this voltage will be NOT be present.

Signals from the engineering rack push buttons and relays, HA and DEC SM clamp micro-switches, DEC centralising micro-switches and CAMAC are opto-isolated on Board 1 (Drawing: ED039) before being presented to the clamp logic circuit.

The clamp logic circuit on Board 2  (Drawing: ED040) consists of a series of TTL gates which read the state of the switches,  push buttons and relays. Decisions are made according to truth tables whether to release or engage the SM clamps. Other signals are examined to check if the switch-over between the QM and SM pre-amplifiers can take place. Once SM drive is selected, a torque loop is engaged with a signal being sent to the QM pre-amplifiers. The DEC centralising logic is also contained on this board as is the logic to be able to select computer mode.

Board 3 (Drawing: ED038) contains the power transistor relay drivers and relays which operate the 3 phase SM clamp motor contactors. n.b. It should be noted that the contactors on the tray are obsolete. The clamp contactors are now located in the middle section of the Mains Distribution Cabinet and marked as HA and DEC SM clamps. The EMERGENCY STOP, power amplifier RESET and associated relays are also found on this board or the plug in relays within the tray. A break in the alarm chain or the Emergency stop pushed will cause the 50V supply to be removed from the power amplifiers.

When in computer mode, clamp control is passed to a CAMAC OD2407 output driver (Address: B4 C3 N15 A0). Bits used are:

  1. A0  HA SM clamps engaged
  2. A1  HA SM clamps disengaged
  3. A2  DEC SM clamps engaged
  4. A3  DEC SM clamps disengaged
Other relays within the Clamp Logic tray control the lamps in the switches which show QM or SM clamp status and whether engineering or computer mode is selected. A set of relay contacts (on RL1, 2 and 15) send emergency stop, reset button and computer/engineering status to CAMAC  (Address : B4 C3 N16 A0).

A recent addition was a small board with a relay to enable the TCS to switch from computer to engineering mode. The relay is operated from a CAMAC output driver (Address : B4 C3 N15 A1). A new drawing: ED062A shows the switching arrangement and link through connections.

Refer to:  INT Technical Manual 1 for more information on the Clamp logic tray and truth tables.

Associated drawings:  TB1 550  BB1 302  BB1 303  BB1 304

Servo Pre-amplifiers

The pre-amplifier boards are located in Card Crate 4 (Bay 3) in the CLIP centre. These being:

Quick Motion:

Slow motion: The input DEMAND voltages from the engineering control rack rate potentiometers or CAMAC are routed through relays on these boards depending on whether engineering or computer mode has been selected. This being determined from the clamp logic circuit.

On the SM pre-amplifiers, the demand signal after buffering is passed through an active notch filter which is tuned to the computer loop sampling frequency of 20Hz. This prevents mechanical resonance of the telescope. It is then amplified and compared to the tacho signal.  The error signal generated from the comparison is taken to the SM power amplifiers. The tacho voltage will back off the input demand until the required velocity is achieved.

As the operating conditions of speed and load on the worm/wheel encompasses an area in which slip-stick may be anticipated, an additional dither signal is introduced into the servo. This signal is generated used an oscillator constructed from CMOS gates (IC5)

The HA and DEC QM pre-amplifiers contain a speed interlock circuit (IC4) driven from the tacho signal. The output drives a relay with contacts in series with the contactor coil and its switching transistor. This prevents the SM clamps being applied if the QM speed is too great.

Associated drawings:  BB1 120  BB1 121  BB1 123

Power Amplifiers

There are six power amplifiers distributed in Bays 1 to 3 of the CLIP centre. As the amplifiers are interchangeable, their functions are marked on the side of the cabinet. They are: The six is a working (hopefully!) spare kept in Bay 3.

The input power to the amplifiers is derived from full wave rectified; three phase transformers with anti-surge circuits mounted in the bottom of two of the bays. These PSU's connect to the amplifier trays through a 20A fuse and smoothing circuit. Each amplifier is capable of delivering 30A at 40V. However, the current is limited to a maximum of 16 amps.

The INLAND power amplifiers are mounted on a removable tray with the connections being made via a Lucar crimp tag terminal strip. n.b. All amplifiers (including the ones used in the JKT) are compatible with each other and can be interchanged between the INT and JKT telescopes and the positions they occupy.

See: INT (JKT)  INLAND power amplifiers  for more details and how to change a faulty unit.

The INLAND amplifier is modular in that it consists of a 'plug-in' encapsulated hybrid servo amplifier mounted on a frame consisting of two large heatsinks. A mains driven fan draws air through the frame. The hybrid amplifier drives 12 power transistors (2N3773) connected in an H-bridge configuration; 3 per leg, 6 mounted on each heatsink. n.b. It should be noted that the 3 transistors in each leg of the bridge need to be matched for gain so that the current passing through them is evenly distributed.

A bi-metal strip thermostat which opens at 70oC is mounted on each heatsink. If it operates (usually a sign that the fan has failed), an alarm output will shutdown the telescope drive. This alarm will also illuminate the P.A. Overload indicator on the control desk.

The output to the motor is protected with a 25A fuse. In series with the fuse is a shunt with cables connected across it going to the motor current meters on the engineering rack. Each power amplifier has 3 indicator lamps on the front panel. These are:

Cassegrain Turntable

The turntable is driven via a reduction gearbox by a DC torque motor with integral tachogenerator. Motor power is delivered by an Inland servo power amplifier.  Limit switches which are connected using by-pass diodes enable the turntable to be driven out of a limit. (drawing: ED003)

The Cass turntable pre-amplifier board EC6 (Drawing: ED048) is located in Card Crate 4 along with the RA and DEC servo pre-amplifiers. It enables the turntable to be driven in a +/- direction and at SLOW or FAST speeds, either manually with push buttons in engineering mode or via a CAMAC output driver (Address: B4 C3 N11 A0) when under computer control.

Turntable position is measured using a 36,000 count BCD encoder (Drawing: ED050). An 18 bit line driver box mounted near the encoder carries the encoder data to the CLIP centre (Drawing: ED049). A line receiver/driver board (Drawing: ED018) with the aid of a decoder converts the BCD data to degrees and minutes of arc before passing the signals to a CAMAC input register (Address: B4 C3 N10 A1) that the * TCS can read. The outputs from this board are also taken to an  LED display on the engineering rack (Drawing: ED016).

The conversion is done using a PROM DECODER (Drawing: ED015 sheets 1/2) which contains a look up table in firmware. Both the Cass TT line receiver and prom decoder boards are located in Card Crate 6 (Bay 3).

Historical note:   Originally a Beckmann display for the Cass TT was available on the control desk, but when the TCS moved over to DEC ALPHA computer, this display no longer works.  So ignore drawing: ED017. At one time, a Cass TT handset with LED position readout could be plugged into a mirror cell connector, but the cable got ripped out of its connector many years and we never managed to work out how it was connected? i.e. The Grubb Parsons drawings for this were wrong!

* For reasons unknown, the TCS after converting the BCD values to binary has to invert the top 3 bytes (whole degree part) to make sense out of the data.

A block diagram showing the Cass TT layout can be found in (Drawing: ED002)

Associated drawings:  TB1 457  TB1 490  BB1 004  BB1 005  BB1 131  BB1 301

Focus control

The secondary mirror assembly and the prime focus cone unit (PFCU) use identical stepper motors for the focus drive. They are driven from a double height Eurocard stepper motor drive board in the left hand side of Card Crate 4/5. Drawing: ED088 shows the general arrangement. The stepper motor controller is a standard RGO MMS board (as used in the INT A&G box and IDS controllers).

The circuit diagram is kept with the MMS circuits in Document Folder 15 in the INT electronics workshop.

The control board drives the motor at two speeds: FAST or SLOW.  There are three input lines for direction, speed and power. The latter is used to de-energise the motor coils to save power when motion ceases. Focus control can be either via the push buttons; SLOW and FAST, + or - DIRECTION on the engineering rack or via a CAMAC output driver module, (Address: B4 C3 N11 A1) when computer mode is selected.

UPPER and LOWER limit switches protect the mechanism and if active will force engineering mode if computer mode was enabled. The mechanism needs driving out from a limit switch manually before computer control can be re-enabled. Limit switch status and focal station selected: PRIME or CASS (this is determined by internal wire links which are linked differently within the CASS and PRIME focus motor cable connectors) can also be read by CAMAC,  (Address: B4 C3 N12 A1).

Associated drawings:  BB1 312  BB1 313  BB1 315 ED076  ED077  ED088

Focus Position Indication

Two systems are used. Both are identical for the Cass secondary mirror and the Wide Field Camera (mounted on the PFCU unit).

Absolute transducer

This uses a Linear Variable Capacitor Displacement Transducer (LVDT) with an transducer signal processor and LED display (E725)  manufactured by RDP Electronics Ltd. The piston on the transducer is spring loaded against the moving surface.

The E725 transducer signal processor and a junction box are mounted in a large box near the top of the prime focus connector panel. A cable with a 9-way D-conn. carry the signals between the junction box and the transducer.

Manufacturers documentation on the LVDT and the E725 display unit can by clicking on the links.

The output from the E725 unit is RS485 2400bds 8bit NoPar 1-stopBit protocol. RS485 is used, because the run to the CLIP centre is excessive. Data is carried down over a 2-wire link.

In the CLIP centre, the data is sent to a CAMAC 3340 comms module (Address: B4 C3 N18). An RS485 line receiver re-converts the data back to RS232 and this unit is plugged directly into the D25 connector on the 3340 module. A small mains plug type PSU powers the line driver.

The calibration procedure for the LVDT can be found here

A suitable Absolute Focus Display has still to be identified to mount into the Engineering rack in the control room.
For this reason, a VT220 display monitor standing on top of the control desk is used to show the raw data from the absolute focus encoder. This VT220 display is connected in parallel with input of the above mentioned CAMAC 3340 comms module.

Incremental  transducer

To improve focus position accuracy, a SONY LVTD is mounted alongside the ASL transducer. It is has a spring loaded piston which MUST make a good contact with the driving surface. The piston has been known to stick, but cleaning this with a solvent to remove grime cures the problem.

A single cable connects the transducer to the SONY processing module with LCD display. This is mounted in the side of a larger box (located at the bottom of the prime focus connector panel) to gain access if needed to the programming buttons. The processing module has an RS232 serial I/O configured for: 2400BDS 8N1 protocol. Because the cable run to the CLIP centre is excessive, an RS422 line driver carries the data down a 4 wire link.

Within the large box, a small PSU is fitted which powers both the SONY processing module and the RS422 line driver. As the display gives off a green glow in the dome, a black taped mask is put over this, but can easily be removed if needed to gain access to the programming buttons.

In the CLIP centre, the data is sent to a CAMAC 3340 comms module (Address: B4 C3 N19). An RS422 line receiver plugs directly into the D25 connector on the 3340 module. A small mains plug type PSU powers the line driver.

More information on this unit can be found at:  SONY focus transducer in the Ops-team web pages.

Focus tracking

To allow the focus to track due to the expansion and contraction of the Serrurier truss with temperature, five platinium resistance thermometers (Type: PT100) are fitted to the telescope tube, top end ring, cube and against and above the mirror.

The mirror air sensor is fitted below one of the petals. The mirror sensor is mounted on a sprung arm and rests against the top edge of the mirror. A side cover needs to be removed from the mirror cell to gain access to this sensor. All the sensors (with the exception of the mirror) are clamped under a nylon mount to the surface to be monitored. Heat sink compound is spread onto the sensor and surface to improve thermal contact. The temperature sensors are mounted in small diecast boxes.

Located under a side panel within the cube is a box containing the temperature transmitters. These provide an output current drive which is proportional to the resistance of the sensor (100 ohms at 0oC). The outputs from these transmitters go to Board EC8  (Drawing: ED072) fitted with precision 1K resistors in Card Crate 4 in the CLIP centre.

The voltages developed across these resistors are taken to a CAMAC ADC 1232 module (Address: B4 C2 N20) enabling the TCS to process the temperature data and adjust the focus if so required.

The ADC channels are:

  1. A0  Top ring
  2. A0  Cube
  3. A2  Mirror air
  4. A3  Mirror edge
  5. A4  Tube
It should be noted that only the TUBE sensor is read from which focus corrections are made.

Associated drawings:  TB1 714  BB1 310

Last edited:  Nov 2012  rjp