NavList:

Yep! They're great sextants. From some of the recent comments, I almost have the impression that somehow this is news to some NavList members. So let's say it plainly: the Astra IIIB is an excellent instrument. If you can buy one new, or equivalently "nearly new" on ebay, for example, you will almost certainly be very happy with your purchase. Plus you can show off to your family, neighbors, and so on by comparing yourself to Robert Redford in that movie. :)

You mentioned you didn't worry about side error, and really you don't have to worry about it. In fact if the Sun images came together reasonably well in your IC tests, then there is no "side error". The name is really a misnomer. There is no actual "error" from "side error". If it's large, it can be a nuisance, but otherwise it's really never an issue.

You should consider other methods for measuring index correction. It's very easy to screw up the alternating Sun-limbs method when reading the off-arc angle and with a relatively small IC, you would not notice the error. I have known people who have read off the angles for this test incorrectly for decades...

For your intercepts off an artificial horizon, you got:
"+1.7', +1.2', +1.9', +1.0' and +1.8'"

See, those are really not that good. We could blame a little of it on inexperience with a new instrument, but you already have sextant experience, so I don't think that's a good excuse. With this sextant and an artificial horizon, you should expect intercepts (altitude errors) on the order of half a minute of arc or better straight out of the box. These numbers look like the result of an incorrect index correction. They average to +1.5' so that may well be your residual index error. In fact, that implies another, perfectly legitimate way to determine index error: take a nice run of a couple of dozen sights at various altitudes with the Sun both rising and later falling and find out what the average offset (intercept) is when you intentionally leave out IC. If everything else has been eliminated, then index error is what's left over. There are statistical limits to this sort of test, but you can worry about that later.

-FER

PS: In your original message, you used this notation: +1!7, +1!2, etc., employing an exclamation as a sort of diacritic representation of the decimal plus the minutes of arc hash mark. I don't recommend this. This sort of typographic "hack" dates from the typewriter era, and it's very common in books from the 1970s especially. Today, it confuses newcomers to the subject, adds nothing for those who are familiar with it, and has the unfortunate consequence of interfering with machine-reading of text (which matters more and more today). I used to use this notation, too, but the faults out-weigh any benefits. In fact, I can't think of any benefits anymore. :)

----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------