NavList:

So let's talk about scope collimation (telescope parallelism). Modris and others, yes, I have written about scope collimation many times before in NavList messages, but I am not going to tell you 'yeah, whatever, it's all in the archives...' and then point you to some message from a decade ago. Why not? Well, first of all, it's a bit rude. We are here and now, and our discussions should be, too. But perhaps more importantly, because 'go read the archives' is one of the poison pills that kills online communities. I am strongly tempted to block or paywall the NavList archives for messages older than, maybe, three years. I've already reduced their "presence" on the NavList website. But they do still serve a useful purpose, and some frequent readers are seriously addicted to the archives (looking at you, A.S.!), and I wouldn't want to cause any withdrawal crises. :)

The scope on a sextant is attached by a mounting point, often a vertical "riser" perpendicular to the frame with a large thumbscrew that holds the scope in place at some offset height above the frame. The scope or sight tube or observer line-of-sight must be parallel to the frame of the sextant. Actually, to make a fine point of it, the scope or line-of-sight should be perpendicular to the geometric plane to which the mirrors have been made perpendicular. For most sextants and under normal adjustment, that imaginary plane is the plane of the frame of the sextant itself. The sextant scope must be collimated so that it is parallel to the frame. An interesting counter-example is a "Bris" sextant which has no actual frame but collimation of the line-of-sight still matters (how do we achieve that??). Collimation is a fancy word, common in astronomy generally, meaning accurate optical alignment. It's probably un-necessary jargon, but since it is common in optics, we need to know it.

Why worry? If the scope or line-of-sight is not parallel to the sextant frame, you'll get small errors. You'll also discover that you get differences in "contact", for example, the contact between the Moon and the Sun in a lunar, or the contact between a star and the horizon in common altitude sight, at different locations in the sextant's field of view. Note that these differences of contact can also be caused by optical anomalies so you should be aware that you may be seeing two or more phenomena in combination. If the contact changes as you look up and down in the field of view, you probably have slightly warped mirrors or other optical components (I have seen this many times with plastic sextants). If the contact changes from left to right across the field of view, then you should check your scope collimation.

Modris Fersters, you posted an image (pdf) showing how the contact can change from one side of the field to the other (left to right): good contact on left, tiny gap at center, good contact on right. This is an example of a well-collimated scope, and in my (long!) experience, very few observers can detect that "tiny gap at center" once the scope is collimated. Some careful observers with good eyes and steady hands can see it, but for most observers, this residual is not visible. Historical navigation manuals recommend looking for this gapping as the primary method of testing scope collimation. That's a bit nuts, and they were almost certainly just copying from each other. This scope collimation can be done instead with a nice tabletop test (see below).

The optical test, looking for a gap that changes across the field of view, can be useful "in the field" (someone hands you a sextant on a ship and it has not been adjusted) especially when using an old instrument. Modern sextants do not often suffer from scope collimation problems because manufacturing has become far more exacting and rigorous (usually!) in the past 300 years. It's not a rare issue to discover a little misalignment. It's also not all that common. But it's worth using the optical test to detect a gross error of collimation. As in the image from Modris, you bring together a pair of stars separated by a large angle, 90° or more. Bring them into contact so that they are visible on the far left of the field of view. Then, without changing the angle (don't touch the micrometer), rock the sextant so that the two images shift to the far right of the field of view. If the star images are now separated, your sextant scope is out-of-alignment. It's tilted. I needs to be "collimated".

On short notice, "in the field", bad scope collimation may be directly visible just by looking at the scope's tilt as attached. Hold the sextant so you are looking down on the instrument from above the index mirror. You can see the axis of the scope relative to the sextant frame. Sometimes you can actually see that it's tilted. In that case, shim it! This really works. Just get a bit of card stock (maybe fold it once) or perhaps a bit of a toothpick (yes, really!) and stick it into the mounting point between the riser and the scope. Visually align it as best you can, and you may well discover that the scope now passes the optical test. If you have more time and finer tools, you can mess around with adjustment screws to get the tilt to zero.

Historical accounts talks about those delicate wires in the field of view of the scope. They're not necessary for any of this, but the authors of most of those historical manuals cribbed each other's work, and they always seem to mention the wires. If there's no appreciable change in contact across the field of view, then you're all set. You don't any wires for this.

The old optical test was included in sextant manuals seemingly without any thought. The authors copied each other for over a century in that period when a tenth of a minute of arc mattered to the lunarian observer. But this scope collimation is done more accurately and more easily in a tabletop test. Find a long-ish hallway (5m or 15 feet or so should be enough). Affix a target, like a piece of graph paper, at a convenient height on the wall on the far end. Place your sextant on its side on a table at the near end of the hall. Sight along the sextant frame and find the exact spot on the target that is aligned with the frame (there are some tricks here, which we can talk about in follow-ups). Mark that level on the target with a nice, easily visible horizontal line. Now look through your scope. The exact center of the field of view of your sextant at the target should be offset by exactly the height of the scope's physical central axis above the frame of the sextant. That is, if the scope's axis is two inches above the sextant frame, then you should be looking dead-center at a spot on your target that is two inches above the horizontal line you drew.

A nice feature of a tabletop test like this, as well as a tabletop adjustment for side offset (erroneously termed "side error") is that these are good "rainy day" tasks. You can do them indoors. It's raining here again today, but of course writing these NavList posts is also rainy day work. :)

Frank Reed