I never fail to be astounded at the number of amateurs who do not keep their telescopes collimated. I wonder whether they are simply scared of doing more harm than good, or even damaging the instrument. Admittedly, the traditional methods of star collimation are best done with two people, but, for the webcam user, this is just not necessary. It must be said that not all telescopes, even expensive ones, have friendly mirror adjustment systems. I know of one world-class planetary imager who bought what he thought was the ultimate planetary telescope (at a cost of $13,000 for the optical tube assembly) only to find that the mirror collimation system was so complex that the telescope was virtually unusable. This was a telescope with excellent optics, made useless by the dealer's bodged primary and secondary mirror cells. The telescope was so bad in this regard that when the tube was horizontal the primary mirror was prone to being catapulted out of its cell by the spring-loaded mirror support! However good a telescope manufacturer is at making optics, make sure they or their dealers can make reliable adjustable mirror cells too.

Precise collimation is a necessary chore because even the finest telescope only delivers pin-point diffraction limited images over a very narrow field of view. This will be irrelevant for low-power views of galaxies and comets, but absolutely critical for high-power views. In this regard, the finest planetary telescopes are the long-focus Newtonians. I am talking here about instruments with (typically) apertures of 25 cm or so and focal ratios of 7 to 10 (i.e., focal lengths between 1.75 and 2.5 meters). A term you will often hear about in this context is the "sweet spot." The sweet spot of a planetary telescope is the diameter of the region within which perfect pinpoint star images appear. As you move further away from the optical axis of the telescope, stars become distorted by aberrations like coma and astigmatism. As you cross the sweet spot boundary, these aberrations just start to perceptibly degrade the star images. Needless to say, your planet needs to be within the telescope's sweet spot, ideally, bang in the middle. The physical size of this sweet spot is frighteningly small. With a simple Newtonian, coma starts to degrade the image first and the sweet spot diameter is proportional to the cube of the telescope's f-ratio. (Yes, you did read that right—the cube!) Just to give a few examples, an f/4 Newtonian (of any aperture) will only have a sweet spot that is a tiny 1.4 mm in diameter! An f/6 Newtonian will have a 4.7-mm sweet spot. Increase to f/8 and you get a very nice 11.2-mm sweet spot. If you are a real long-focus Newtonian fanatic, f/10 will net you a whopping 21.9-mm sweet spot. Converting these diameters to angles, for a 250 mm aperture, gives you sweet spot angles on the sky of 4.8, 10.8, 19.3, and 26.5 arc-minutes. As can be seen, the 250mm f/4 Newtonian has a sweet spot barely larger than the largest lunar craters. Conversely, the 250-mm f/10 Newtonian has a sweet spot almost as large as the full Moon!

The Newtonian is only one type of telescope. Compound telescopes (like Schmidt-Cassegrains, Maksutovs, and Maksutov-Newtonians) have a variety of sweet spot diameters, but they are all measured in arc-minutes and they all need collimating precisely for good planetary performance. The ever-popular Schmidt-Cassegrain Telescope (SCT) design can only be collimated by adjusting the tilt of the secondary mirror to reflect the optical axis of the primary straight down the drawtube middle; adjusting the primary mirror in such telescopes is not practically possible.

The final stage in the collimation of any planetary telescope has to be done on either a real star or an artificial star. This can be frustrating at first because, as you adjust the optics, the star will move, so it needs to be recentered. However, after a bit of experience, the collimating chore can become routine. With fine adjustments the test star will only move an arc-minute or two. Of course, if manufacturers made telescopes whose optics did not move around, you would only have to collimate a telescope once. Sadly, this happy state of affairs rarely exists unless you build yourself a custom long-focus Newtonian. Schmidt-Cassegrain telescopes have "conical" primary mirrors (i.e., they are thinner at the edges than in the middle). Such mirrors' relatively light weight can be supported by the telescope's central baffle tube alone. However, this raises the problem of SCT mirror flop. As you move the telescope around the sky, or flip the telescope about the declination axis of your German Equatorial Mounting (GEM), the mirror position will change. It may only change by a few arc-minutes, but that is enough to wreck the collima-tion. Fortunately, the three collimation screws on an SCT's secondary are easy to adjust, even if you do have to remove the dew cap to get at them. I know of one planetary imager who always leaves his SCT parked in the position he images in (i.e., on the meridian, at the planet in question's declination, so he can retain col-limation).

The modern webcam imager has a distinct collimating advantage over his visual counterpart, especially when using a telescope that is pretty close to perfect colli-mation to start with. You can simply attach a webcam to a high power Barlow lens or eyepiece and observe the test star on your computer screen, as you tweak the collimation for that final stage. You can have the slow-motion hand controller next to your PC to keep recentering the star too. This makes life a lot easier than darting from eyepiece to adjustment screw, back-and-forth, back-and-forth, bending and stretching while sweating away in thermal clothing! With a webcam and a laptop, everything can be done in relative comfort.

Let us now examine, in detail, the precise process of collimating a telescope from scratch and the tools that help to make this process relatively painless.

Basic Collimation of a Newtonian Reflector

Figure 3.2 shows the three phases in the basic daytime collimation of a Newtonian telescope: uncollimated, secondary mirror adjusted, and primary mirror adjusted. The Newtonian is an ideal telescope to collimate precisely because both primary and secondary mirrors can be adjusted; everything can be made "just right." The figure shows the view, through the drawtube, when the observer's eye is centered where the eyepiece normally sits. A plastic 35-mm film canister with a hole in the middle makes an excellent sighting tube for positioning the eye. Just follow the steps in the figure and the basic collimation is complete.

The best way to visualize the optical configuration of a Newtonian telescope, from a collimation viewpoint, is that there are two optical axes: the optical axis of the primary mirror and the optical axis of the eyepiece. The axis of the primary mirror is at right angles to the primary at the optical center of the primary (this center is usually assumed to be at the dead center of the circular glass mirror). If you have a thoughtful mirror manufacturer they will have marked this center with

Reflection of secondary and spider in primary ..¿ill

Inside of telescope tube, behind the secondary m


Primary mirror

Secondary holder, correctly centered behind the drawtube

Primary mirror

Primary now centered exactly in the middle of the secondary

Primary now tilted correctly towards secondary a b c

Figure 3.2. The three key stages in rough collimating a Newtonian reflector, prior to star collimation - this is far less critical in a slow Newtonian.

a tiny dot or ring. If not, it will help to mark the mirror yourself. As the dead center of the mirror lies under the "shadow" of the secondary mirror, this marker will not affect the performance. The photons from a star in the precise direction of the primary mirror's axis will be reflected and "focused" to a perfectly sharp image at the focus on this axis. (This is where your webcam needs to ultimately sit.)

The axis of the eyepiece is usually assumed to be at the center of the focuser drawtube. Of course, for the webcam user, it will really be the dead center of the webcam chip, after the obligatory Barlow or Powermate enlarging lens. But, when we are actually collimating the system before the star test, the eyepiece axis will, effectively, be the axis of the observer's sight tube, Cheshire eyepiece (read on), or laser collimator (also read on). The secondary mirror diverts the incoming light to the side of the Newtonian tube. The secondary will also divert the optical axis of the primary and the eyepiece (depending on how you look at the situation).

The main purpose of collimating is to align the primary and secondary mirrors to form one common axis. Normally, you do this by adjusting the position and tilt of the secondary mirror, and the tilt of the main mirror. In addition, the drawtube should, ideally, be parallel to the primary mirror's optical axis. With webcams, one has to assume that the chip inside the webcam is mounted at right angles to the webcam-drawtube adapter. Let us not get too paranoid at this stage!

One issue that often worries first-time Newtonian collimators is that when they look down their telescope tube they get the impression that the secondary mirror is not exactly centred in the tube; the flat secondary mirror appears to be offset toward the mirror end and away from the eyepiece end. In fact, this is perfectly normal, as the light cone that needs to be captured is slightly bigger at the bottom of the flat secondary mirror than at the top. This "offsetting" of the secondary mirror is routine in quality telescopes but the required offset is fairly small in long focal length Newtonians where the size of the secondary is so much smaller than the focal length. Some telescopes have offset secondary mirrors (fast, f/4 Newtonians and telescopes with tiny secondary mirrors really need them), some do not. The critical point here is that, when looking through a sighting tube at the reflection of the primary mirror in the secondary mirror, you should be able to see the whole of the primary. If you cannot, light is being lost. Ideally, the reflection of the primary should appear concentric in the secondary mirror. The offset formula is: minor axis/(4 x focal ratio). So for a 30-mm minor axis secondary on an f/7 Newtonian you need to slide the flat mirror 30/(4 x 7) = 1.07mm toward the primary and away from the eyepiece. In this case, you would need to be a perfectionist.

The good news for the planetary webcam imager, who is only interested in capturing objects in a field that is only 1 or 2 arc-minutes wide, is that once you have carried out the two-stage mirror alignment shown in Figure 3.2 you can move straight to collimating on a star (or even an artificial star). Collimation purists may well disagree with me here, but I am trying to steer a course between not collimating at all (seemingly, the approach of 90% of amateurs) and obsessive colli-mation. Yes, you can ensure that the drawtube/focuser is perpendicular to the tube (by removing the secondary holder and passing a pipe across the tube or by using a laser to project across the telescope tube). Yes, you can redesign the secondary support so that the secondary is offset precisely. But, to get those perfect planetary images you simply need to: 1) carefully tilt and rotate the secondary mirror until the primary mirror appears concentric within the secondary; 2) tilt the primary mirror until the secondary mirror is concentric in the primary; 3) conduct a star test. In stage 2, it is quite possible to achieve perfect collimation with a long-focus Newtonian by sheer good luck. This is because, if you have a spot marked on the mirror's optical center and can see the reflection of your eye and the sight-tube hole centered on that mirror spot, you may well be within a sweet spot that is 6 or 7 mm across. However, this cannot be relied upon. A star test must be performed, ideally using the equipment with which you will image.

Collimation Aids

Before we move onto the star test itself, I would like to say a few words about "gadgets" to help you collimate a telescope. The cheapest gadget, and one that can easily be home made, is a basic sighting tube as shown in Figure 3.3A. All this needs to be is a device just 31.5 mm in diameter (to fit the 31.7 mm diameter of a 1.25 inch eyepiece hole) with a small hole placed dead in the center, a few millimeters wide. This device ensures that the observer will hold their eye at the dead center of the drawtube when rough-collimating the secondary and primary mirrors. A useful addition is to place a white ring on the inside of the sighting device, around the observer's eye hole. This white ring will be obvious in daylight and its reflection will aid collimation.

Another gadget, available commercially, is a device known as a Cheshire eyepiece. This is little more than an eyepiece-sized sighting tube with a reflective washer-like insert, tilted at 45 degrees, and inserted into the sighting tube. It looks

Figure 3.3A. The Orion (USA) Collimating eyepiece is, essentially, an accurately machined tube with a central sighting hold mounted within a bright metal disc. Image: Jamie Cooper.

a bit like a miniature half-periscope. A gap in the side of the Cheshire allows light to be directed onto the reflective washer, creating a bright, doughnut-like reflection on top of all the other reflections seen during the collimation process. If the primary mirror center is marked with a spot or a ring, it enables accurate alignment of the primary spot with the doughnut's reflection, thus aligning the optical axes.

A more modern gadget is a laser collimator. Again, shaped in an eyepiece-sized package, this device shines the beam from a laser diode in a pencil-thin beam from the drawtube, onto the secondary and then onto the primary. If everything is lined up correctly, the laser beam will trace a return path exactly along the path of the beam, ending up with the laser beam pointing back at itself. With some laser colli-mators, the point where the return path strikes the laser is made easier to see by combining it with a Cheshire type design (i.e., a hole in the laser body side through which the laser's return journey can be verified). Such a device is shown in Figure 3.3B. Some caution is needed with laser collimators though, and not just from the point of view of eye safety. Unlike placing your swivelling eye at the hole of a sighting tube, the laser is simply blindly shooting its beam straight out of its casing. It is therefore wise to check that the telescope's drawtube really is perpendicular to the telescope tube and that the laser beam is emitted parallel to the laser beam collima-tor body. Some years ago, Sky & Telescope magazine tested a laser collimator whose beam was not emitted parallel to the collimator body. This would lead to an erroneous collimation. To check the machining accuracy of the laser collimator, mount the laser collimator in the telescope drawtube backward and point the laser at a wall (with the telescope drive off). Then rotate the collimator in the eyepiece. The spot on the wall should not move as the collimator is rotated. If it does, your collimator is a

Figure 3.3B. The BC&F Astro-Engineering Laser Collimator combines a laser collimator with a "Cheshire" type sighting hole, so you can see exactly where the laser beam is being directed. Image: Jamie Cooper.

lemon, unless it has some provision for adjusting the beam. This test is easier to perform if a spare drawtube is mounted rigidly in a workshop bench vice. After verifying (hopefully) that your laser collimator is OK, removing the secondary mirror and seeing where the laser beam ends up on the opposite side of the telescope tube will verify whether your drawtube and focuser is square-on to the tube. Personally speaking, I have never been that attracted to laser collimators. I have always reckoned that visually collimating by eye and then precisely collimating on a star is the best sequence of events. Laser collimators are often hyped as the ultimate collimat-ing tool, but if their limitations are not understood they can be of limited benefit (don't believe all of the manufacturer's hype). Unfortunately, collimating on a star can be thwarted by poor seeing, which is why I like the next gadget so much.

BC&F Astro-Engineering's Picostar, shown in Figure 3.3C is an artificial star generator that produces a point source illumination out of a 50-micron diameter fiberoptic ferrule. It allows a variety of illumination levels and, provided you have a long garden and a telescope that stays in collimation when the telescope is moved around, it is a godsend. For the device to work properly, it needs to be far enough away from the telescope that the 50-micron aperture appears smaller than the diffraction limit of the telescope. In practice, this means that a 200-mm instrument requires the Picostar to be placed at least 20 meters away and a 300-mm instrument needs the Picostar to be 30 meters away. In practice, many telescopes will need the device to be that far away anyway, as otherwise they could not achieve focus. With a Newtonian you may well need a drawtube extension tube to achieve focus. I have heard this type of adaptor referred to as a pervert tube as it enables a Newtonian to be focused on bedroom windows. It will not surprise anyone to learn that I own one.

In the pages above I have discussed, at length, the basic daytime collimation of a Newtonian reflector. This is because in home-made or commercial Newtonians, all of the component parts are often able to move and are all accessible for adjustment.

Figure 3.3C. The BC&F Astro-Engineering Picostar. An invaluable artificial star device with a 60-micron diameter light source. Image: Martin Mobberley.

The Newtonian is almost unique in this respect. With other telescopes, especially mass-produced ones, things are rather different. Refractors are largely collimated for life. The relatively light and long focus optics of commercial refractors are often fixed and the optics of mass-produced small Maksutovs are frequently fixed too (although in some cases they would be much better if they were adjustable). The ubiquitous Schmidt-Cassegrain, whether made by Meade or Celestron, has no provision for adjusting the primary. As we have already seen, the SCT primaries will often tilt by an arc-minute or two as the telescope is moved and thus nightly collimation becomes necessary. The optical axis of the primary mirror of an uncollimated commercial Schmidt-Cassegrain usually ends up pointing within a few millimetres of the centre of the instruments secondary, and a few minutes tweaking on that mirror's adjustment screws, will produce perfect collimation, at least, until the primary flops around a bit on it's central support. With a Schmidt-Cassegrain, there is only one daytime collimating step, adjusting the three screws on the secondary mirror (Figure 3.4) until the reflection of the primary in the secondary is concentric, as seen through a drawtube sighting tube. With regard to SCT mirrors "flopping" out of collimation, the larger the sCt, the

Schmidt Corrector Plate
Figure 3.4. The collimation screws on a Schmidt-Cassegrain can be found on the secondary holder mounted on the corrector plate. Image: Martin Mobberley.

more likely this is to happen, in my experience. I would like to say one thing in favor of SCT collimation though. Most Newtonian and Cassegrain systems have push-pull collimation systems. In other words, you fiddle about with two thumbwheels to get collimation: you slacken one thumbwheel then tighten the other at the primary mirror end. This can be a real nighttime hassle. Although you can only adjust the secondary with an SCT, the secondary mirror is light and the adjustment screws are spring-loaded. Therefore, you just tweak one of the three screws (or, ideally, small thumbwheel/knob screws) and the mirror moves. There is none of this push one way/pull the other hassle at the mirror end. Telescope makers please note! SCTs are a breeze to recollimate, even if they rarely stay collimated.

Nighttime Star Collimation

After daytime collimation, the next step for any telescope is nighttime star collima-tion. If you have a rigidly made Newtonian, a long garden, and an artificial star device, you need not bother about real stars or a clear night. An artificial star has a huge advantage over a real one—atmospheric seeing is not an issue. The textbook diffraction patterns will always be seen, even if tube currents and heat from the ground mean they are not quite perfect. For both Newtonian and Schmidt-Cassegrain collimation, the old-fashioned visual way of collimating on a star is a hundred times easier if an assistant is employed, so the telescope owner can look through the collimating eyepiece while the assistant adjusts the mirrors. However, as I have already mentioned, collimating with a webcam and a high-power Barlow is a much easier system. But the choice is yours. Either way, you will need to know what to see.

The first stage in star collimating a reflecting system is probably unnecessary if careful daytime calibration has already been carried out. It simply involves observing a first magnitude star at a decent image scale (say 250x visually), and well out of focus, and checking that the black hole (the shadow of the secondary mirror) is in the middle of the star's out-of-focus disk. Figure 3.5 shows the real image of an out-of-focus star, imaged with a well-collimated Newtonian, by expert imager Mike Brown. Note that the real image looks a bit different to the perfect simulation in Figure 3.6. That figure, like most of the collimation images here, was produced using a software simulation package called Aberrator (

Intra Focal Focus Extra Focal

Intra Focal Focus Extra Focal

2mm Inside 1mm Inside 1mm Outside 2mm Outside

All Tests taken at Prime Focus with TouCam Pro web cam, each image being the average of 200 frames

Figure 3.5. A real star test on a quality 200-mm Newtonian primary mirror made by Orion Optics (U.K.). The star is examined inside (intra) and outside (extra) focus. Image: Mike Brown.

2mm Inside 1mm Inside 1mm Outside 2mm Outside

All Tests taken at Prime Focus with TouCam Pro web cam, each image being the average of 200 frames

Figure 3.5. A real star test on a quality 200-mm Newtonian primary mirror made by Orion Optics (U.K.). The star is examined inside (intra) and outside (extra) focus. Image: Mike Brown.

Maksutov Collimation
Figure 3.6. A totally defocused star in a perfect telescope, simulated using Cor Berrevoet's Aberrator software.

If the black hole is not in the middle you are way off collimation and the mirror col-limation screws need to be adjusted until the shadow is centered.

The second stage in star collimation is far more demanding and can only be carried out when seeing conditions are reasonable (or with an artificial star). Typically, a second or third magnitude star is chosen (for a 200-300-mm aperture) and it must be well above the horizon so turbulence is minimized. A very high magnification is then used (600x or more for the visual observer and maybe 0.1 arc-seconds per pixel with a webcam) and the star is moved from well inside to well outside of focus while the diffraction rings are examined. There should be a bright dot in the middle and then a series of concentric dark and light rings out from the center. As the scope is moved through focus, this pattern should open and close smoothly and symmetrically (Figure 3.7 shows the slightly defocused view). If it fails this test, the mirror adjusting screws need tweaking. Of course, every time the screws are tweaked, the star will move and will need recentering in the field. Once the intra- and extrafocal patterns resemble a perfect textbook diffraction ring, the third step can be carried out.

The third and final step to perfect collimation can only be executed when seeing conditions are near-perfect—a rare event for most people. The set-up is the same as for step 2, except that the star is perfectly focused. We are now looking for the perfect Airy disk, a so-called "false" central disk, surrounded by diffraction rings of diminishing brightness (Figure 3.8). If the first ring is not uniform, or is incomplete (as in Figure 3.9), the collimation screws need tweaking by a tiny amount to achieve a complete and uniform first ring. This last test is so sensitive that even

Figure 3.7. A slightly defocused star in a perfect telescope, simulated using Cor Berrevoet's Aberrator software.
Figure 3.8. A perfectly focused star in a perfect telescope, with perfect collimation! A very high magnification is needed to see this view, along with a night of excellent seeing.

Figure 3.9. A perfectly focussed star in a perfect telescope, but fractionally out of collimation. Note the difference between this and Figure 3.8. A very high magnification is needed to see this view, along with a night of excellent seeing.

moving the scope around the sky will alter the situation with many telescopes! Hard-core planetary imagers will often tweak their telescope's collimation with this final step, before they commence imaging and even adjust the collimation throughout the night.

Well, we have now covered collimation, but it should be emphasized that for the star tests, the diffraction patterns seen will not resemble anything like the textbook patterns unless the night is perfect or you are using an artificial star. If you want to see textbook diffraction patterns on a less-than perfect night, try looking through a small aperture quality refractor or Maksutov at a very bright star. This will enable you to familiarize yourself with what should be seen. Users of giant Newtonians, e.g., 40 cm and larger, will have to accept that they may never see a perfect diffraction pattern or Airy disk unless they stop the instrument down. Not only are the transient, perfect, atmospheric columns (often called "cells") of stable air rarely more than 30-cm across, large telescopes take a very long time to cool down.

Of course, if even on a good night, when stars are not shimmering wildly, the Airy discs look distorted, and you have been through the telescope collimation process carefully, you may deduce that you have inferior optics. To be honest, this is very rare with optics from any of the major manufacturers. Competition is fierce in the modern telescope market and companies just cannot risk turning out poor optics. However, if you are using an old telescope, one from a backstreet mirror "cowboy," or one of dubious secondhand origin, the optics might be suspect. However, in my experience, most amateurs who complain of poor optics simply have failed to collimate them to the precision required to get diffraction-limited views of the planets. Also, a mirror cell that "pinches" the optics is often a source of nonperfect star images.

It can be fascinating to examine a perfect artificial star with a perfectly colli-mated Newtonian. The act of merely taking the rear mirror cell dust cap off, thereby altering the tube currents, can cause dramatic changes in the diffraction patterns of the star and makes you appreciate just how important the thermal properties of a telescope are.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


  • Ralph
    How important is collimation in planetary imaging?
    3 months ago

Post a comment