NavList:

Modris,

Sorry I do not find any mentioning of SNO-T in the book. So I probably confused it with some other book. The experiments to determine accuracy are discussed on p. 13-14, 36, 42-43, 46-47.

Here is what I find in another book of the same author (1989):

According to the research of Krasavtsev, all instrumental errors of SNO-T are 2-3 times smaller than those of SNO-M. Random errors of altitudes for the new sextant under normal conditions are characterized by the average quadratic value 0'4 for the Sun and 0'5 for stars....
Average quadratic value of systematic error with SNO-T is 0'4 when using dipmeter and 0.7 and more when using tables (of dip+refraction). The old sextants have 1.5-2 greater errors.

This leaves an interesting question when exactly they started to produce SNO-T.

This other book "Determination of the ship position" is 218 p. And it is not about cel nav, but covers all methods of navigation, including [dead] reckoning and [radio direction-finding]. The only interesting statement about sextants is the one I cited.

Alex.