A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Modris Fersters
Date: 2020 Oct 15, 11:23 -0700
Problem: the best method to find the index correction with maximum accuracy during night time using Star.
The usual method, published in many suorces, is very simple: the obsever must bring the direct image of the star on top of the reflected image. But there is a problem with accuracy, because the two images seams to be exactly on top of each other even when the observer turns the micrometer for same fractions of minute. As a result here arises an error.
If the micrometer is rotated only in one direction (to achieve the maximum accuracy avoiding backlach error of the micrometer), the problem described above becomes more actual, because it is not possible to check the moment when the direct and the reflected image are exactly above each other and there is no possibility to move micrometer in opposite direction.
As lunars and star to star distancies are my favorite, I tried to find some method how to get index correction with better accuracy (by star). I tried following method (see picture): I bring the direct and reflected image in very little distance (when I can visually see a little gap between the two images: see picture). I write down the on or off arc reading (depending on wich dirrection I turn micrometer). Then I continue to turn the micrometer untill the images change their places and I can see the both images in the same small distance. Again I write down the reading. Then I calculate the index correction using the same principle as we do it using the Sun.
If one reading is on arc, but other off arc I use the following formula: IC=(60-a-b)/2; here “a” and “b” are the readings (it is not important which reading is on arc, wich is off arc, the result will be the same); if the result is negative, then sextant “0” value is on arc; if the result is positive — the sextant “0” is off arc.
The results with this method are very good (compering with tradicional method). Usually I take some 5 observations and then calculate the mean value.
Of course this method is nothing genius, but very practical. Maybe the navigation experts can suggest some better way to find index correction during night time. I only want to emphasis that I speak about VERY ACCURATE measurments (on the border of sextant limits). Of course I understand that for tradicional moder navigation this is not actual. But for lunars I think it is not good practice to unnecessarily increase the total error only because the index correction contains needless error of for example 0,3’.