With a refractor, like with a SLR lens, you very rarely need to do this and I have seen very few refractor models that have the ability to perform a collimation. Usually moving the 2...3 lenses as a single group, basically (re)-centering this lens-combo to align with the optical axis / center of the camera or eye-piece..

I'm going to skip Newtonians & Dobsonians because I have no practical experience with those. Dobsonians suffer from field rotation during long exposures (unless you buy an expensive field rotator). Newtonians also often are sold with a simple Alt-Az mount and those mounts are less suitable for astrophotography. Both types of telescopes often are designed for visual observation and don't always have enough back-focus to support a CCD or SLR-camera.

I only briefly talk about Maksutov telescope : while it often is said they don't need maintenance or collimation, this assumes extremely high precision during manufacturing and no changes over the telescopes lifetime. I think it still may be necessary to ensure the corrector plate is parallel & centered. Not often, but like with a refractor, it might be necessary. Usually adjustment options are not provided.

With a SCT or RCT, you have to think what kind of tolerances there are and which of them you can correct with reasonable effort, while some are not critical
Let's assume the primary mirror as the reference
  • The secondary mirror can be off-axis in 2 direction
  • The secondary mirror can be tilted in 2 angles
  • The focuser can be off-axis in 2 direction
  • The focuser can be tilted in 2 angles
  • The OTA (relative to the primary mirror) can be off-axis in 2 direction
  • The OTA (relative to the primary mirror) can be tilted in 2 angles
  • The internal baffles may not sit perfectly centered and parallel to the optical axis
  • SCT : the tilt of the primary mirror changes due to tolerances on the focus mechanism and based on the angle of the telescope
  • RCT : The primary mirror is supposedly static but some models still suffer from mechanical flexture

Not all factors are practical or necessary to worry about. The baffles, for example, will cause vignetting and maybe reduce some protection against stray light. A hood is helpful for added stray light protection and vignetting can be dealt with in post-processing (e.g. gray-frames). However, collimation-techniques utilizing the shadow these baffles cast, have to consider this imperfection.

The tolerances & play in the SCT's focus mechanism causing tilt of the primary mirror also isn't easily dealt with. Furthermore, the SCT's primary mirror might move as the OTA is pointed towards the stars -- CELESTRON provides extra screws to minimize such issues but has no adjustments in case the primary isn't perfectly perpendicular to the optical axis.
Technically, aligning the focuser/camera isn't part of the "telescope's collimation" but to get the best image result, it has to be done, preferably during the same steps, even if it means adding an extra "collimation adapter ring"

What can be adjusted ?
  • SCT : the tilt of the secondary mirror
  • some SCT : adjust centering the corrector plate and thereby the secondary mirror (AFAIK it can't adjust tilt of the corrector plate)
  • SCT : usually cannot adjust the tilt or offset of the primary, relative to the OTA
  • SCT : needs extra collimation adapter to correct tilt between primary & focuser -- I don't have one for my 925HD and therefore no first-hand results
  • RCT : the tilt of the secondary mirror
  • some RCT can adjust the spider to center the secondary onto the optical axis (a RCT has no corrector plate to adjust,)
  • RCT : adjust the tilt of the primary relative to the optical axis & OTA
  • RCT : also needs extra collimation adapter to correct tilt between primary & focuser. Yeah, I got that together with the motorized focuser upgrade :-)

Making sure all the optical planes are parallel is the most important. I have to see what's the impact of any off-axis errors. As I see it (but I may be wrong), the remaining error would be between the OTA and the (combined & collimated) optical axis of those three.

To perform collimation, you have several ways to do it:
  1. Use a real star on a clear night
  2. Use an artificial star (day or night)
  3. A "simple" laser collimator projecting a dot, used instead of the eye-piece
  4. A "holographic" laser collimator instead of the eye-piece, projecting a pattern
  5. A laser collimator shining into the front of the telescope (plus a mirror in place of the eye-piece
  6. Combining webcam, software and (artificial) star
  7. ... any others ?

To select among those, I have a few extra constraints :
  • Want to do it in daylight
  • preferably in a room or small camp-site -- not needing 50...200ft distance (e.g. artificial star)
  • using an EQ-mount -- many procedures are easier done by Alt/Az to recenter the star

So far, my collimation of the RCT has progressed better than that of the SCT. The RCT also requires/allows more adjustments. Fortunately my Orion RCT has one useful extra, the 925HD lacks -- an "engraved" center dot to align the optical axis !! This is really important as it serves as a reference to determine the optical axis. And no, this dot doesn't affect your image because the center obstruction caused by the secondary mirror is much worse.

IMO, collimating the focuser / camera is an important part. If you add extension tubes, you can easily see how they can increase mis-alignments and how improper fastening causes unpredictable variations. It also helps to see which of the tubes are of better quality.
To reduce discrepancies, it is best to use a setup that's close to your imaging setup -- same extensions and possibly same focus position. My choice collimator is the "Howie 2" Glatter" with the extra circular pattern (a dot mask comes FOC). There are other vendors and I found that this one has a good price and many positive reviews to the accuracy & long-term stability.

Now let's get started with the actual collimation work :

First, you have to get familiar with the collimator as well as how to setup the telescope and be able to peek inside to see the center dot (on the secondary mirror). This is a learning curve and once you've done it twice, you don't spend much time on it. I found it convenient to place the scope on a table or a tall workbench -- when on my EQ-mount, I have to bow down a lot to look inside and make the adjustments. You also want to make sure you don't have bright, stray lights e.g. from windows pointing at your eyes or inside the telescope tube -- one more argument to do it indoors where you can draw the curtains to dim the lights when needed.

The adjustments start from the rear element / focuser and gradually move forward to secondary and finally primary mirror.
STEP #1 : Point the laser's center onto the dot on the secondary's mirror (dot mask). After you have finished this step all three elements will sit (ideally) CENTERED onto the optical axis. They still may be tilted but they are centered -- tilt is what will be adjusted during Step #2 & #3.

To perform the adjustments, you need to carefully turn the screws on the focuser collimation ring, go back & peek into the front of the OTA to see how the laser's position has changed and repeat that until the laser dot is dead on the center.
REMEMBER this adapter is an optional, extra accessory and not part of the telescope itself. You can choose smaller versions placed in front of or part of the camera adapter. Or skip this altogether.

STEP #2 : With the collimator's laser now representing the optical axis and hitting the center of the secondary mirror, it is time to adjust the TILT of the secondary mirror. For this you need a dimly lit environment since you have to look inside the OTA to see the pattern on the primary mirror. If the surroundings are too bright you won't be able to see those ring patterns or at least it will take time for your eyes to adapt.
The adjustment procedure is easier than #1 since you will look into the OTA while turning the adjustment screws of the secondary mirror. No need to move back & forth. Collimation of the secondary is achieved when the (circular) pattern on the primary is perfectly symmetric. In case of the circular pattern, you can move the focus back & forth till the edge of one ring exactly matches to outer edge of the primary. Once the pattern illuminates the entire edge of the primary, the secondary mirror is collimated.

STEP #3 : By our assumption (and lack of adjustments), the primary mirror always sits dead-center on the optical axis -- but it may be tilted. Now with secondary mirror is projecting an image parallel to the collimator / camera plate, we can use that, to correct the tilt of the primary mirror.
This requires a little bit of extra preparation : Preferably place the telescope in front of a screen or wall and make sure it is at a right angle (horizontal & vertical).
The laser projects a pattern onto the secondary and that gets reflected by the primary. The primary in turn reflects it again and you can see the projected pattern on the outside. You also see the shadow of the spider and the secondary mirror. And that's the key !!
While you adjust the primary mirror, you can observe the changes of the symmetry of the shadow. The primary mirror is collimated when the shadow of the secondary is symmetrical in that projected image. (NOTE : projection surface / wall should be parallel to the OTA's mirrors. If they aren't, you don't see circles but still can recognize or measure any asymmetry)

RCTs, by design, have a fixed connection between primary & focuser ==> once you change the primary's tilt, the focuser 's axis won't be pointing to it's original target anymore and you have to restart at STEP #1 ... #3. In most cases adjustments are very small, if any.
CORRECTION : It seems not all RCTs are constructed the same way -- some models have the rear-cell attached to the OTA instead of the primary mirror.

Compared to a (natural or artificial) star -- why should I use a laser when I get natural stars for free or artificial ones for a lot less money than a laser-collimator ? For me the reason is CONVENIENCE & TIME.
Using a laser collimator, I can prepare well in advance. And even do a check on the collimation before the observation TIME.
And if you recall Steps #2 & #3, there is a lot of CONVENIENCE if you can instantly see the changes and don't have to move back & forth. And especially with a natural star, all adjustments involve kneeling down to see through the eye-piece and getting up to adjust the secondary. Plus the need to re-center your (alignment)star. And it moves and you have to make sure your tracking is good. Given the maginification, touching the telescope can point it away from the alignment star.
Step #1 you can do without a laser, using the Chesire eye-piece that usually is included with a good RCT. As per Orion's instruction you should be able to perform collimation steps, especially adjusting the optical axis, BUT YOU NEED DAYLIGHT and an evenly lit surface or cloudless sky. That's okay for the first round but once you adjust the primary, you cannot verify the optical axis with the Chesire EP (unless you setup some lights & screen).
And once you start observing during winter time, the CONVENIENCE of doing all these adjustments beforehand and not in the freezing cold is yet another valid argument in favor of a laser-collimator.
In my location at home, space constraints also aren't favorable for use of an artificial star since there you often need 50...200ft distance and I don't have that amount of space available.

There are other, technically intriguing options -- the HOTECH Advanced CT Laser Collimator is one of them. Instead of plugging the laser in the EP's position, that collimator shines light into the front of the telescope, onto a mirror (in place of the EP) and uses the reflected image to do the adjustments
A few comments got me intrigued -- especially the brief sample on how to center the SCT's CORRECTOR PLATE. I haven't seen a tool to perform this. HOWEVER -- the youtube demonstration also mention the need to recenter the alignment star and show the procedure with an Alt/Az mount. IMHO, this recentering will be a lot more time-consuming if you have to do this manually using an EQ-mount.


© Copyright 2013, All rights reserved -- post a link to this page or contact me, if you want to use it in your blog or publication or if you want a (large) print