NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Freiberger Trommelsextant errors
From: Alexandre Eremenko
Date: 2012 Apr 18, 17:56 -0400
From: Alexandre Eremenko
Date: 2012 Apr 18, 17:56 -0400
Wolfgang, I am interested in seeing the raw data of the observations of Mars you mention, if it is not too much trouble to tipe them:-) > Just another cumbersome > correction in the sight reduction. It is not so sumbersome indeed:-) Just glue the correction table to your sextant box, and add to each observation together with Index correction. > however frequently use an artificial mirror horizon. Self-made or Freiberger (or else)? > I made a sine approximation of the error curve and > calculated a correction table in 5? increments in > decimal minutes, to simplify handling. I do not see how sine (or any other) apprixomation simlifies handling. All you have to do is to add some number of minutes to each observation, which can be easily done mentally. What is the difference between adding a true number or an approximation? > A couple of weeks ago I took a series of observations > of Mars with the Freiberger and with my SNO-T. About > 10 shots each, changing instruments after every > observation. Why did you switch the instrument fater every observation? (To my understanding a natiral thing to do is to observe the whole series with one instrument, and then with another). > Any > deviations at random, normal observation inaccuracy. Yes, I would like to see these deviations. Or the row data and corrections you applied. > I just wonder why Freiberger is not consistently > producing "good" instruments, given the long > experience and likely high quality machinery they are > using. I don't know. The high quality machinery can be worn out. I can also imagine economic reasons. The sextant is probably tested once, after it is made and assembled. Suppose that 10% have errors of the order 1' while the rest goal is < 20". (This must be something like a normal distribution after all:-) What would you do in such cituation? Throw these 10% or 20% away? Of course a sensible solution would be to give them some ratings. Say, classes A and B, as the British did. But this is not really necessary as soon as a certificate showing errors is attached. Alex.