NavList:

I had an interesting experience while taking  the sun sights with home made artificial horizon. I used motor oil as a reflecting surface. But after calculations I found that all the results of the measurments are quite inaccurate (I hoped to achieve the accuracy of fractions of minute, but I could achieve the accuracy only some minutes of arc).  Initially I thougt that there where problems with my sextant. But the lunar distance and star to star distance measurments showed that the sextant is OK. This was a great puzze for me untill I found some 19. century author book (Raper; 1840). The author emphasis that only quicksilver and water are only materials proper for perfect artifical horizon.

Then I tried water as liquid surface. And all the problems disappeared. The results where perfect.  Again I experimented with an oil and found that in relatively small container (about 10cmx15cm) the measured angle differs significantly dependent on the point where I looked at the reflected image of the sun (at the center or near to the edges of the container).  

It is interesting that even in proffesional books (including old books) the oil is mentioned as one of the liquids suitable for the artifical horizont. Maybe it is because usually the requirements for extra accuracy where not so high.

Modris Fersters