NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Norman Baatz
Date: 2025 May 9, 01:01 -0700
Hi Frank, here is a plot of all my sightings vs. the ratio of my measured altitudes and the altitudes shown in your app (set to Centered / AH). You can see I'm using an artificial horizon. It includes a few more recent sightings, including a group of four around 30°, the left most four points. It might be hard to get much lower angles with the Davis artificial horizon. (I see that I'm showing the ratios in the European not the American format, with commas instead of periods.)
The red line is a best fit through all the sightings, the green line just considering an average of the four sightings I'm always taking in succession (and possibly excluding a few outliers). For both lines I've added the (obvious) constraint that they have to pass through (0, 1.000), and, as I wrote, I do check the index error before every single sighting. (Grey is a previous line I had been using, just for reference for me.)
So I'm using your app to determine the line, which I'm then using to create the arc correction, which I'm then applying before I plot the LOPs. The lines from last week and from today are not much different, and I'm still adding measurements, and also trying to get more low angles. My recent LOPs, after applying the arc correction, are still looking good.
OTOH when I plot straight from sextant, it is quite obvious that something consistent is happening. If anybody else sees that their Mark 3 is also out of whack (and mine seems to be a lot more than Greg's), I would be interested to see their LOPS and their distance from the actual location, to see if they also have what looks like a consistent problem.
I am assuming that people with the Mark 15 or Mark 25 won't be having as much of an issue, because they can better fix perpendicularity errors, by having correction screws on both the index and the horizon mirrors. I for one have now gotten over my initial frustration with the Davis Mark 3, thanks to the post (I think it was Frank's) that the perpendicularity error can mostly be solved by an arc correction. I decided to put some effort into getting the best out of it, as well as learning something on the way, instead of throwing it into a corner and buying a better model.
Norman






