Telescopes – I.

Optics & Mounts

Dave Kilkenny

1
Contents
1 Introduction 3
1.1 What is a telescope ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Atmospheric transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Aberrations 4
2.1 Chromatic aberration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Spherical aberration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Coma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Oblique astigmatism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5 Field curvature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.6 Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

3 Telescope basics 11
3.1 Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3 Light-gathering power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.4 Resolving power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.5 Seeing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

4 Telescope configurations 17
4.1 Refracting telescope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Prime focus and Newtonian reflectors . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.3 Gregorian reflector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.4 Cassegrain reflector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.5 Schmidt camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.6 Maksutov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.7 Coudé focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.8 Nasmyth focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

5 Telescope mounts 27
5.1 Equatorial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2 Alt-az . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

6 The Largest Telescopes 32

2
1 Introduction
This lecture assumes knowledge of basic optics – reflection, refraction, dispersion, diffraction,
interference and so on.

1.1 What is a telescope ?

A telescope – as far as astronomy is concerned is just a “light bucket”. In almost all applications,
we are not interested in magnification, but simply in collecting enough photons to measure.
Usually, this means collecting enough photons to achieve a desired “signal/noise” ratio, always
important in astronomy.
With larger telescopes, we increase the collecting area, so we increase the number of photons
collected. This means that we can observe fainter objects, but we can also improve resolution.
This can be spatial resolution, but it is often more important to get better spectral resolution
(eg. by using echelle spectrographs) or better time resolution for “high-speed” photometry
or spectroscopy. This means that one can take very short exposure (direct or spectroscopic)
measures whilst retaining good signal/noise.

1.2 Atmospheric transmission

Astronomy developed as a “visual” region science – a natural result of the sensitivity range of
our eyes. Radar development led to the discovery of extra-terrestrial radio sources (the Sun and
the galactic centre) and the development of radio astronomy. The schematic of the transmission
of the Earth’s atmosphere show why infrared astronomy tended to be developed at high altitude
observatories and the study of ultraviolet, X-ray and γ-ray astronomy was really restricted to
high-altitude balloon observations and the advent of satellite astronomy (IUE, ROSAT, EUVE,
IRAS, HST, ....).

Figure 1: Schematic of the atmospheric transmission as a function of wavelength.

3
1.3 Units
The basic unit of length in physics is the metre – in the case of visible light, the units usually
used are nanometres (10−9 m), so that visible light is about 400 – 700 nm.
However, in astronomy, many other units are used – partly as a matter of historical inertia, but
mainly for reasons of convenience. We thus have, for example:

• Optical – Ångstrom (Å). Traditional optical range unit (10−10 m)

• Infrared – Micron (µm). Near infrared ∼ 1 – 5 µm.
• Radio – mm “microwave”.
• Radio – cm. eg. 21cm line of neutral Hydrogen.
• Radio – Frequency/Hertz (Hz). eg. 21cm = 1420 MHz.
• X-ray, γ-ray – Energy (eV). eg. 1 KeV = 2.4 × 1017 Hz = 12.4 × 10−10 m

2 Aberrations
The history of telescope development and indeed much of modern telescope design is substantially
affected by optical aberrations and trying to minimise them.

2.1 Chromatic aberration

Chromatic aberration occurs in lenses (not mirrors) because blue light is refracted more than
red, so the focus for blue light is slightly closer to the lens than for red light.

Figure 2:

Chromatic aberration can be corrected using (for example) two different kinds of glass – with
different refractive indices – to form an achromatic doublet.

4
However, even if we work with reflecting surfaces (or monochromatic light) we encounter other
aberrations.
In elementary geometrical optics, it is customary to consider (for example) the formation of images
by a lens by considering light rays very close to the optical axis (paraxial rays) and making
the approximations sin θ = θ; cos θ = 1. This results in first order or Gaussian theory, where
optical elements are essentially perfect, and gives, for example, the well-known lens equation and
so on.
One way of approaching reality is to look at a McClaurin expansion of the sine of an angle:
θ3 θ5 θ7
sin θ = θ − + − − ........
3! 5! 7!
It is easy to see from the table below that inclusion of only the second term on the right-hand
side of the equation will result in a quite close fit to the “exact” result. Use of the θ3 term results
in third order theory and enables determination of the Seidel aberrations.

Table 1: Values of sin θ and the first three expansion terms.

θ sin θ θ θ3 /3! θ5 /5!
10◦ 0.17365 0.17453 0.00089 0.00001
20◦ 0.34202 0.34907 0.00709 0.00004
30◦ 0.50000 0.52360 0.02392 0.00033
40◦ 0.64279 0.69813 0.05671 0.00138

Figure 3: The five Seidel aberrations.

5
2.2 Spherical aberration
Spherical surfaces are easy to figure, but light striking the outer part of a spherical lens – or
mirror – focuses closer to the lens/mirror than light from the inner part. For a lens, spherical
aberration can be removed by designing an achromatic doublet so that the spherical aberrations
cancel. For a mirror, it is common practice to use paraboloidal rather than spherical surfaces.

Figure 4: Spherical aberration effect in a spherical mirror.

Figure 5: Spherical aberration.

Figure 6: Simulation of the effect of spherical aberration on the image, passing through focus.

6
The Southern African Large Telescope (SALT) has 91 separate mirrors of approximately
1m diameter each. These are all figured with spherical surfaces because:
• spherical is easier and therefore somewhat cheaper – important when there are 91 of then
(97, including spares).
• The mirrors have to be integrated into a single surface – adjusted to form a single optical
element – this is considerably easier with a spherical surface than any other conic section.
• Individual mirrors can be used anywhere on the surface of the sphere. Any other surface
would require that each mirror would have a unique location and could not be used elsewhere
on the surface. (the Keck telescopes actually do this – but it’s hard).
• With a spherical surface, “spare” mirrors can be used anywhere on the surface. Mirrors can
be replaced by freshly aluminised mirrors on a planned and regular basis. This would not
be possible with an aspheric surface.

Figure 7: Simulation of SALT. Notice the “tracker beam” and alt-az mount.

As we have already seen, a major drawback of a spherical lens or mirror is spherical aberration.
The primary mirror of SALT will therefore have this aberration which will be corrected by a
complex and well-designed lens system – the spherical aberration corrector (SAC)

7
2.3 Coma
Images formed from off–axis rays, are distorted (quite badly for paraboloids !). If we think of
images from successively larger radius circular strips of the mirror, these combine to form an
elongated image (“comet-like” = coma or comatic images). The further from the optical axis, the
worse the coma gets.
Field correcting lenses can be used to reduce the effect of coma.

Figure 8: Schematic illustrating coma.

Figure 9: Simulation of the effect of coma on the image, passing through focus.

8
2.4 Oblique astigmatism
The term “oblique” astigmatism is used to distinguish astigmatism in a lens or mirror, other
than that which is produced intentionally by a cylindrical surface or, for example, in the human
eye (most correcting spectacles will include a correction for astigmatism). Oblique astigmatism
(like coma) is an off-axis aberration and all uncorrected lenses and mirrors will suffer from it.

Figure 10:

Figure 11: Simulation of the effect of oblique astigmatism on the image, passing through focus.

9
2.5 Field curvature
In the absence of spherical aberration, coma and astigmatism, the focal surface of a spherical
mirror or lens will lie on a paraboloidal surface called the Petzval surface. If astigmatism is
present, the curvature will be more severe than the Petzval surface.

Figure 12: Field curvature.

In the days of glass photographic plates, it was possible to force a plate to curve into the shape
of the curved focal surface. With modern solid state detectors, a telescope might need expensive
“field-flattening” optics to operate over a “wide” field.

2.6 Distortion
Even if it is possible to get rid of “point” source aberrations – the on-axis spherical aberration
and the off-axis coma and astigmatism – and the field curvature, an optical system can still suffer
from distortion. This is effectively a slight variation in magnification across the field and results
in the well-known “barrel” and “pincushion” distortions – so-called because they look nothing
like either a barrel or a pincushion.

Figure 13: Field distortion.

10
3 Telescope basics
3.1 Speed
The brightness of an image is proportional to the light collecting area, so:

B α a2
where a is the diameter of the telescope aperture
But for an extended object (such as a galaxy) the more the image is spread out by the optics
(the “scale” of the telescope), the less will be the brightness in energy/unit area on the detector.
The spread is proportional to the focal length of the telescope, so for extended sources:
B α (a/f )2

Amongst the parameters defining a telescope are thus the focal length (f) and aperture (a), or
the:
focal length
focal ratio = f/a =
aperture
The focal ratio is sometimes called the speed or f-ratio of the telescope. A small f-ratio means
faster speed because of a brighter image (for an extended source). It thus means shorter exposure
times to measure a certain brightness of object but it also means that the scale of the telescope
is smaller.

3.2 Scale
The scale of a telescope is the way in which angular size in the sky translates to linear size at
the focus of the telescope – which would normally be on the detector. Scale is usually expressed
in arcseconds/mm.
Simple geometry gives:
scale, s = 206265/f

where s is in arcsec/mm, and the focal length of the telescope, f, is in mm. (and there are ∼
206265 arcseconds in a radian).

Example: at the prime focus of a 4m telescope with an f/3 primary,

focal length, f = f ratio × aperture = 12m = 12000mm
so,
scale = 206265/12000 = 17.2 arcsec/mm

11
3.3 Light-gathering power
The Light-gathering power of a telescope is simply its ability to collect light and is therefore
proportional to the collecting surface area, or a2 , if a is the aperture (usually by “aperture” we
mean the diameter of the primary mirror).
For point sources (such as stars), ideally the image is concentrated into the same area irrespective
of focal length (or f-ratio) so “speed” depends only on aperture.
If we compare the SAAO 1m telescope to the human eye (with a maximum aperture ∼ 8mm),
the relative light gathering power will be (1000/8)2, or a factor of nearly 16000.

3.4 Resolving power

Light passing through a circular aperture suffers Fresnel diffraction, so that even an infinitely
small point source will show a somewhat diffuse image with an interference pattern – the diffrac-
tion pattern.

Figure 14: Diffraction pattern produced by a circular aperture. The central spot is the Airy disk.

In astronomy, the radius of the first minimum in the diffraction pattern is referred to as the Airy
disk. The angular radius is given by:
λ
θ = 1.22
a
Actually sin θ, but we can write sin θ = θ because the angles are very small. And the linear
radius is:
λ
θ = 1.22 f
a
where:
f = focal length
a = aperture (diameter of the objective)
λ = wavelength

12
Example: For white light (λ ∼ 5600Å = 5.6 × 10−5 cm, given a lens of diameter 4cm and a focal
lenght of 30cm, the Airy disc has an angular radius:
5.6 × 10−5
1.22 × = 1.71 × 10−5 radians = 3.5 arcsec
4
and a linear size of:
30 × 1.71 × 10−5 = 5.1 × 10−4 cm.
For a point source (star), the central disk is thus ∼ 0.01mm in diameter.

Two sources are said to be resolved (Rayleigh criterion) when the peak of the central maximum
of one falls on the first dark ring of the other.

Figure 15:

The minimum resolution of a telescope in the optical (say 5600Å) is then:

λ 1.22 × 206265 × 5.6 × 10−5 14.1
θmin = 1.22 (radians) = = (arcsec)
a a (cm) a(cm)
This is the diffraction-limited resolution – which is generally unattainable due to atmospheric
effects (The Hubble Space Telescope is an obvious exception to this).

Examples: SAAO 1m telescope θ ∼ 0.14 arcsec

Human eye θ ∼ 50 – 60 arcsec
Jodrell Bank 75m θ ∼ 9 arcsec (at λ ∼ 20cm)

13
Figure 16: Demonstration of Airy disk/difraction-limited resolution with various sizes of objective/primary mirror.
From left – 13 cm, 50 cm, 2.4m and 5m.

A couple of other image effects are worth noting here:

• Images of bright stars will often show diffraction “spikes” (see figure) from the “spider”
which support the secondary mirror.
• Bright images on photographic plates often show halos due to halation – internal reflection
of scattered light within the (photographic) glass plate.

Figure 17: Optical effects from bright images – diffraction and halation.

3.5 Seeing
In the previously given examples, we have seen that a small aperture (4 cm) will result in
diffraction-limited resolution of several arcseconds, whereas for even a modest-sized 1m tele-
scope, this figure is near to 0.1 arcsecond. Since the effects of the atmosphere tend to degrade
images by amounts of the order of an arcsecond, diffraction-limited resoltion is rarely attained by
earth-based telescopes - at least in the optical. Radio telescopes are usually diffraction-limited
because they operate at so much longer wavelenghts.
Light entering the atmosphere travels through increasingly denser, higher pressure air which, in
the lower few kilometers of the atmosphere, also increases in temperature. Since refractive index

14
is a function of density and temperature for a given wavelength, the refractive index increases
nearer the ground. This effect is systematic and therefore essentially predictable. However,
the atmosphere also contains random and unpredictable turbulent and thermal variations which
introduce lateral variations in refractive index on a timescale up to 100 Hz.
These variations distort the plane wavefront which arrives above the atmosphere and can produce
the same effect as transient lenses and prisms – refraction, diffraction and dispersion. The most
common of these distortions cause “ripples” in the refractive index which are comparable to the
telescope aperture and cause brightness and colour scintillation of star images – what is seen as
“twinkling” of the star by the naked eye.

Figure 18:

The distortion of a plane wave by the effects described above – the scintillation of the image of a
point source such as a star – shows, on a very short time-scale (∼ 100 Hz) a pattern of “speckles”
in which the speckles are of the order of the Airy disk in size. A longer exposure adds all these
speckles into a much larger image than the diffraction-limited image and this is usually called the
seeing disk.

Figure 19: Repeated short (2 millisecond) exposures of Aldebaran. The speck at the head of the arrow is roughly
the size of the diffraction-limited image.

15
Figure 20: Schematic showing the effects of diffraction and seeing on stellar images.

16
4 Telescope configurations
We use telescopes because they give us:
• Light collecting power (light “bucket”).
• The use of modern detectors (including even the photographic plate) allow us to store or
“integrate” photons for many minutes.
• The use of larger apertures gives us better resolving power.
Recall that for many purposes, magnification is unimportant – stars are effectively point sources,
so magnifying them just magnifies the seeing disk. Telescopes give us the ability to resolve –
spatial resolution, spectral resolution and time resolution.

4.1 Refracting telescope

The earliest telescopes (eg. of Galileo, 1609; Kepler 1610) used lenses and are called refracting
telescopes or refractors. The text-book arrangement for a simple refractor is shown in the
figure.

Figure 21:

This arrangement is called afocal – both object and image are infinitely distant. The lenses are
confocal – separated by the sum of their focal lengths. The magnifying power of such a set-up
is:
f1 tan θ y1
magnification = = =
f2 tan α y2
Aberrations in such a simple arrangement would be unpleasant. Light losses would occur at the
four surfaces and in the two lenses.
Refractors generally have long focal lengths ( = good scale) and were good for astrometry. How-
ever, there are several drawbacks:

17
 • It is difficult to make large optically accurate lenses.
• It is difficult to support large lenses adequately at the edges, and such lenses are quite thick
in the middle (and so, very heavy).
• Chromatic aberration is a problem in a single lens.
• Glass absorbs light – especially in the ultraviolet.
Refractors also tend to be rather long for their aperture, and therefore need relatively large domes.
The 40-inch (1m) refractor of the Yerkes Observatory was the largest ever built.

Figure 22: The Yerkes 40-inch (1m) refractor. Big, innit ?

On the up side, refractors are generally mounted in a sealed tube which keeps the interior surface
clean. They also tend to be rugged and stable.
Reflecting telescopes or reflectors use mirrors as the optical elements because they:
• are easier to figure accurately (to better than λ);
• they only need to be figured on one surface, not two;
• can be supported more evenly and satisfactorily (including “active” optics;
• can be given high reflectivity, so light losses are small;
• can be made from lower quality glass (in the sense that defects are not important) but must
generally be made of low expansion glass (eg. “Pyrex”, “Zerodur”, “Astrositall”).

18
Because modern telescopes are large, the size excludes lens or dioptric systems and requires
mirror or catoptric systems, or combinations of mirrors and lenses catadioptric systems.
Some systems are described below.

4.2 Prime focus and Newtonian reflectors

A prime focus system has essentially a single element – the primary mirror. A Newtonian
system has a simple “flat” to move the focus out to one side of the telescope. The paraboloidal
primary avoids spherical aberration. The first reflecting telescope was designed by Newton in
1670.
These systems can be quite fast - typically around f/5 or so - useful for imaging.

Figure 23: Prime and Newtonian foci.

Prime focus observing used to require an observers “cage” and so was suitable for only the largest
telescopes. With increasing automatic and remote control, this is no longer true.
Newtonian focus was popular (and still is) for amateur instruments, but was less convenient for
large instruments because of the asymmetry of the arrangement. Again with increasing remote
access (for example, fibre optics) this is no longer true.

19
 Figure 24: The Anglo-Australian 3.9m Telescope (AAT).

Figure 25: The prime focus “cage” of the Anglo-Australian Telescope.

20
4.3 Gregorian reflector
The Gregorian reflector is a variant on the Newtonian reflector devised by James Gregory.
It has never really been popular because the long focal length – due to the secondary mirror
being outside prime focus – requires a long telescope tube (with potential flexure problems) and
therefore a relatively large dome.

Figure 26: The Gregorian reflector.

4.4 Cassegrain reflector

The Cassegrain variant, with the secondary mirror inside prime focus, has a much shorter length
and has been a popular configuration for decades. Typically f-ratios are f/12 to f/20.

Figure 27: The Cassegrain reflector.

(Note that in modern telescopes, some kind of electronic detector will generally be in the focal
plane, rather than an eyepiece !)
Variants on the Cassegrain system include:
• The Dall–Kirkham, which has a concave ellipsoidal primary and a convex spherical sec-
ondary. The figuring is easier with only one aspheric mirror, tending to give a better on-axis

21
 performance, but off-axis coma is about three times worse than a straight Cassegrain con-
figuration.
• The Ritchey-Chrétien, which has a hyperboloidal primary and a secondary which devi-
ates from a conic section such that the coma and spherical aberration are corrected at the
Cassegrain focus (such systems are called “aplanatic”). Ritchey-Chrétien configurations
are very popular where (wide-field) imaging is required – obviously because of the good
aberration characteristics. They are also compact systems.

4.5 Schmidt camera

Bernhard Schmidt’s (1930) telescope, often called a Schmidt camera has a spherical primary
(no coma) with a corrector plate at the centre of curvature of the primary to correct the spherical
aberration.

Figure 28: The Schmidt camera.

The Schmidt camera combines the advantage of relatively aberration-free images with a wide field-
of-view, in astronomical terms. The SAAO 1m telescope has a field-of-view of a few arcminutes,
whereas a Schmidt with comparable aperture (such as the ESO, Palomar or AAO 1.2m Schmidts)
will typically have around a 5◦ × 5◦ usable field.
The focal plane of a Schmidt is curved, so detectors need to be curved to fit this plane. In the
past, photographic plates were the typical Schmidt detector, and these would be mounted on a
former to bend them to the shape of the focal plane.
Schmidts are usually fast – f/2 to f/2.5. “Super Schmidts” can be as fast as f/0.8 with fields of
7◦ or 8◦ square.

22
Figure 29: Schematics for Schmidt and Maksutov reflectors (left) and a photo of the Anglo-Australian Schmidt
telescope (right).

4.6 Maksutov
Maksutov telescopes are very compact, wide-field systems with a fast primary and all surfaces
spheroidal. A well-designed Maksutov is virtually free from coma, spherical aberration and astig-
matism, but the catadioptric nature sets a limit on the size.

Figure 30: The Maksutov telescope.

4.7 Coudé focus

Coudé is French for “elbow”. Two flats bend the light beam at 90◦ to the optical axis usually
down the declination axis and then through another 90◦ down the polar axis. The advantage of
this is that heavy or bulky equipment can be installed in a controlled environment for stability
(mechanical, thermal or electrical). The long focal length usually means that such systems are
slow (typically ∼ f/35, but have a good scale.
The Coudé focus is being used less now that good optical fibres are readily available.

23
 Figure 31: Schematic of the Coudé focus.

4.8 Nasmyth focus

The Nasmyth focus uses a single flat to send light down one axis of rotation. Use of this focus
is common on alt-az mountings. With a flat which can rotate through 180◦ , two Nasmyth foci
can be used, one on each side of the telescope. A minor disadvantage is that the field rotates,
but this can be corrected by rotating the detector at the same rate.

Figure 32: Schematic of the Keck 10m telescope showing the Nasmyth focus.

24
 Figure 33:

Figure 34: Summary of reflecting telesecope configurations.

25
Of course, different configurations can exist in the same telescope structure. A telescope might be
used at prime focus, at Cassegrain focus (with the insertion of a secondary mirror), or at Coudé
or Nasmyth foci (with the insertion of suitable tertiary mirrors). This has the advantage that
different instruments can be left permanently on the telescope with the various mirrors being
used to select the desired instrument.

Figure 35: Schematic of possible Keck configurations.

26
5 Telescope mounts
There are two main types of telescope mount – equatorial and altitude-azimuth (“alt-az”)
– with many variations.

5.1 Equatorial
This type of mount has been popular, almost universal, since telescopes achieved any sort of
useful size. As we have seen, the equatorial system of co-ordinates (α, δ) is based on the rotation
of the Earth. A telescope with one axis aligned with the polar axis only needs to be driven around
this axis at the sidereal rate to track astronomical objects.

Figure 36: Equatorial mounts.

With sidereal time, we know where the zero-point of the system is, and can easily locate objects
at a given (α, δ) position. Some examples of equatorial mounts are given in the figures.

• The fork gives excellent sky access but is somewhat limited by flexure effects for high load-
bearing. More suitable for smaller telescopes than larger. This type of mount can be made
very fast (e.g. “YSTAR” – The Yonsei Survey Telescope for Astronomical Research.)
• The yoke is better for load-bearing but obstructs access to the polar region.
• The horseshoe is very strong and gives polar access.
• The polar disk is very rigid for a given size.

27
 Figure 37: Equatorial mounts: (a) Fork, (b) Yoke (also “English” mount), (c) Horseshoe (d) Polar disk.

Figure 38: Fork mount telescope. See also the picture of the Anglo-Australian Schmidt telescope in section 4.5

28
Figure 39: Mt Wilson Observatory 60-inch (1.5m) telescope. (Polar disc mount).

Figure 40: McDonald Observatory 82-inch (2m) telescope.

29
 Figure 41: Mt Wilson Observatory 100-inch (2.5m) telescope (Yoke mount).

Figure 42: Mt Palomar Observatory 200-inch (5m) telescope. Horseshoe mount – see also the picture of the
Anglo-Australian 3.9m telescope in section 4.2.

30
5.2 Alt-az
In the last couple of decades, telescope apertures have significantly exceeded the size of the long-
standing record-holder, the 200-inch (5m) Palomar “Hale” telescope (operational in the late 1940s)
and the Russian 6m telescope (operational in the early 1970s, but with many problems). This
generally means much bigger and heavier structures and that the altitude-azimuth (or “alt-az”)
mount is much more suitable.
As we have seen, use of an alt-az system means that the instantaneous altitude and azimuth of a
star must be calculated from the RA, Dec and local sidereal time – and this must be done very
rapidly and very frequently if the telescope is to track a celestial object smoothly and accurately,
because – unlike RA in the equatorial system – the rates of change of altitude and azimuth will
not generally be constant. The increasing speed and power of computers in the last ∼ three
decades has made this practicable.

Figure 43: Schematic of the Japanese “Subaru” alt-az telescope on Mauna Kea. See also the schematic of the
Keck telescope in section 4.8.

In engineering terms, the alt-az mount is pretty much essential for the current generation of
large (∼ 10m) telescopes. It is superior to equatorial mounts because, amongst other mechanical
advantages, the vertical loading is constant and symmetrical, azimuth rotation causes no loading

31
change or deflection, and the telescope tube flexure is in one plane, dependent only on zenith
angle.
There are a couple of (minor) drawbacks to alt-az telescopes:
• The necessity for continuous rapid calculation of altitude and azimuth. This has been solved
by the power of modern computers – so not really a problem at all.
• We have seen that the relationship between horizontal and equatorial co-ordinates can be
expressed by two equations such as:
sin a = sin δ sin φ + cos δ cos φ cos H

and
sin H cos δ
sin A = −
cos a

Note that, in the second equation, if the altitude (a) approaches 90◦ , the azimuth (A)
approaches infinity. What this means, of course, is that for an object which passes through
the zenith, as it does so, its azimuth jumps instantaneously from 90◦ East to 90◦ West (or
from 90◦ to 270◦ ). In practice, it means that for objects near the zenith, it is very difficult to
transform quickly enough and accurately enough from equatorial to horizontal co-ordinates
and a small patch of sky near the zenith is effectively inaccessible, even though the telescope
can easily point there. Simple observing programme planning should avoid problems from
this source.

6 The Largest Telescopes

Figure 44: Schematic of the larger telescope mirrors on Mauna Kea, Hawaii.

32
 Table 2: The largest operational optical telescopes

Telescope Aperture (m) Location Altitude (m) Comments

Keck I 10.0 Mauna Kea, Hawaii 4123 36 mirror segments
Keck II 10.0 Mauna Kea, Hawaii 4123 36 mirror segments
Hobby-Eberly ∼9.2 Mt Fowlkes, Texas 2072 91 mirror segments
Subaru 8.3 Mauna Kea, Hawaii 4100
Antu 8.2 Cerro Paranal, Chile 2635 VLT
Kueyen 8.2 Cerro Paranal, Chile 2635 VLT
Melipal 8.2 Cerro Paranal, Chile 2635 VLT
Yepun 8.2 Cerro Paranal, Chile 2635 VLT
Gemini (North) 8.1 Mauna Kea, Hawaii 4100 Gillett telescope
Gemini (South) 8.1 Cerro Pachon, Chile 2737
Multi-Mirror 6.5 Mt Hopkins, Arizona 2600 MMT
Magellan I 6.5 Las Campanas, Chile 2282 Walter Baade telescope
Magellan II 6.5 Las Campanas, Chile 2282 Landon Clay telescope
Bolshoi 6.0 Nizhny Arkhyz., Russia 2070 Large Alt-az telescope
Hale 5.0 Mt Palomar, California 1900 “200-inch”

Table 3: Large optical telescopes under construction.

Telescope Aperture (m) Location Comments
Large Binocular telescope 11.8 Mt Graham, Arizona 2 x 8.4m
Gran Telescopio Canarias 10.4 La Palma, Spain Keck based
SALT 9.2 Sutherland, South Africa HET based
LZT 6.0 B.C., Canada Liquid mirror

Table 4: Extremely large telescope studies.

Telescope Aperture (m) Comments
OWL 100 Overwhelmingly Large Telescope
Euro50 50
MaxAT 30 – 50
CELT 30 Californian Extremely Large Telescope
XLT 30
GSMT 30 Giant Segmented Mirror Telescope

33
 Figure 45: Schematic of the Large Binocular Telescope.

Figure 46: One of the mirrors for the Large Binocular Telescope.

34