Reader does informal plastic sextant testJan 1, 2003
To the editor: In the recent article on buying a sextant ("Choosing a sextant," Issue No. 93, Nov./Dec. 1998), the author pointed out that a problem with plastic sextants is a wandering or changing index error (IE) that was related to temperature.
Several years ago I taught navigation class for the local Power Squadron, and student questions got me interested in this phenomenon. I had the sight records for 14 students representing a little more than 200 individual sights. About half of the students used plastic sextants and half used metal sextants. When I analyzed this data I found that there was little difference in the accuracy of the two groups. If I used only the daytime sun shots for the comparison, the metal sextant users were slightly better. I finally concluded that not much could be learned by looking at data from beginning students, and decided to do some experiments myself.
I carefully adjusted the IE of a Davis MK25 to zero by sighting a star. I put the sextant back in its case and set it into a closet. A month later I took it out and checked the IE. As I expected it was zero -- no change. Then I put the sextant and case in the trunk of my car and left them there as I drove around for two weeks. Then I checked the IE against a star and again found no change. The IE was still zero. So far the checks were made at twilight with nearly constant temperature. Next I went to the beach at twilight on a hot day and checked the IE by sighting on a star. The case had been in an air-conditioned car, and the first checks seemed to show a small index error, but as I continued to check, the IE became zero again. Temperature changes from about 70° F to 95° F had no observable effect.
Then I checked the IE on the beach on a sunny day with a mild breeze. Here the plastic sextant lived up to its reputation. The IE was clearly not zero. It appeared to be about -6 minutes. The Mark 25 I was using had a "full view" mirror. To find the index error the problem was to exactly superimpose the direct and reflected images of the horizon. On this day, as on many on the Gulf Coast, there was a haze out over the water. Thus, both the direct and mirror images of the horizon were slightly "fuzzy." Trying to get one fuzzy image to exactly superimpose another is far from a precision operation. I was getting IEs that differed by as much as two minutes.
To sort out this problem I borrowed a plastic sextant with a split mirror. With the split mirror I had no trouble using the horizon to get an IE. The problem here is to line up the direct view horizon with the mirror horizon. This is easy to do, and repeated checks convinced me that I was getting a consistent IE reading. The IE obtained with the sextant in full sunlight was about -6 minutes. I was pressed for time and had to leave before I was satisfied with the results. The next day at the beach was very windy with substantial gusts. The ICs were shifting all over the place, from around four to as much as eight minutes. Repeated attempts confirmed the shifting values.
Finally, I recognized a pattern. If the wind was strong, the IEs were smaller, and if the wind was very calm the IEs were larger. The change in IE was the result not of temperature, but rather of temperature gradient. The sun's rays heated the sextant unevenly since it strikes some parts directly; others are shaded. A strong wind tends to equalize temperature to that of the air.
So I concluded that a plastic sextant will be stable for twilight shot, but the IE will be shifting and changing for sun shots. When taking sun shots with a plastic sextant, an IE reading should be taken before and after the sight. If the wind was not gusty and you have a split-view mirror, the sun shot will probably be okay. If it's gusty, good luck.