NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Sextant index error measurement
From: Paul Hirose
Date: 2006 Nov 02, 15:42 -0800
From: Paul Hirose
Date: 2006 Nov 02, 15:42 -0800
I'm just a bubble sextant owner, but based on what I've read it seems that the normal method for measuring a marine sextant's index error is to bring the direct and reflected images of an object (horizon etc.) into coincidence. There is also the method of splitting a celestial body and bringing opposite limbs into contact. Trying to think outside the box, I looked at another approach, utilizing the eye's ability to accurately bisect the gap between two closely spaced lines. The method I propose uses a test target of several parallel, equally spaced horizontal lines. The lines should appear closely spaced but clearly resolved at *double* the distance at which the target will be used. (The reason for that will become clear.) The white spaces between lines should be should be considerably wider than the lines, perhaps three times wider. Set up the target at a measured distance and sight it through the sextant. Bring the reflected and direct images into coincidence. Due to the parallax between the two lines of sight, the sextant will be set to a slight negative angle ("off the arc"). From the vertical distance between the telescope and index mirror, and the distance to the target, you can compute the parallax. Add parallax to the sextant reading to obtain index error. That's a preliminary value. The point of coincidence is hard to judge when attempting to put black lines on top of each other. It's easier to bisect a narrow white space with a black line. That's the reason for the multiple lines on the target. A slight movement of the micrometer knob will split the images so the black lines of one will bisect the spaces of the other, thus doubling the number of visible lines and halving their spacing. The additional small angle to attain this condition can be calculated from the line spacing and distance to the target. Additional bisections occur at multiples of that angle, allowing an average index error to be calculated using more than one point on the worm. (Doing a full revolution would require a large target, though: five feet at 100 yards.) Many variants of this method are possible. For example, the simplest test target would be a pair of lines plus a single line to bisect the space between them. If the pair and the single are spaced to match the vertical offset between the direct and reflected lines of sight at the sextant, parallax is eliminated. To check parallax, make another measurement at half the original distance. It would be best to construct another target so the sight picture looks the same at the shorter distance. The results from this second test will reveal any residual parallax, since IE is constant with distance while parallax varies in inverse proportion. Unfortunately I don't have a marine sextant to try any of this. -- I block messages that contain attachments or HTML. --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to NavList@fer3.com To , send email to NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---