A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Frank Reed
Date: 2020 Jan 14, 17:22 -0800
Hey great! Of course, it's usually something simple like that. I don't know if I said it during the workshop that you attended, but in any celestial navigation manual calculation, I recommend putting an X or a line through any quantity that you no longer need. After you have pre-cleared the LD, you no longer need the original observed LD, so run a line through it, or X it out. This is just a way of avoiding that common mistake of picking up the wrong value at a later stage in a manual computation.
Now that you have worked this by the basic series method, you should definitely try the short method (Bowditch-Thomson) since you already have half the work done doe that, and what's left is relatively easy.
I cleared each of your individual observations a couple of hours ago using my web app. The error on each individually is 0.9, 0.9, 0.4, 0.1, 0.5, 0.3 (all minutes of arc). There's a possibility here that you have under-estimated your index correction. If you had used -2.5' instead of -2.0' for your I.C., the average error would be zero. Just something to consider as you shoot more lunars. If there's always this positive bias, then your can correct for it. Shooting lots of lunars will narrow down uncertainty in your index error as a byproduct. Note that the standard deviation of these errors is about 0.32' which is really good. I find that good observers with good sextants and just a modest amount of experience shooting lunars can get a standard deviation on individual sights of 0.25' and after averaging sets of four, the random error is reduced by a factor of two to 0.12', which is as good as it gets. This, of course, depends on eliminating any residual systematic error such as that potential error in the index correction.
Congratulations on fixing that small error! And see? Lunars are easy. :)