NavList:

George quoted from Peter Ifland's book:
"Tubular spirit-level artificial horizons, whether straight or curved, have several serious limitations. Two tubes are required - one parallel to the plane of the instrument to establish the horizontal and one at right angles to deal with tilt of the plane of the instrument. ln the early designs, the observer had to look at three places at once: at the reflected image of each of the two bubbles and at the image of the celestial body reflected in the horizon glass."

I'm not convinced that this was really a problem. According to other descriptions I've seen, the navigator did not really have to look in "three places at once". The image of the Sun or star would be seen nearly coincident with the image of the longitudinal bubble (the bubble for determining altitude) for much the same reason that we can see the Sun superimposed on the sea horizon in a standard marine sextant --you get partial reflection all across the horizon glass. The transverse bubble was slightly below but I don't see that it would be harder to keep an eye on that than it is to swing the arc in a traditional marine sextant. It might even have been a little easier.

Also, it's worth mentioning here that Gago Coutinho, according to the pages from the book by Rogers posted by Gary, was aware of circular bubble levels and decided by design that a pair of tubular levels was sufficient for his purposes at that time.

Also from Ifland:
"A special lens was required to keep the eye focused on the bubbles at a few inches from the eye and at the same time, focused on the celestial body at infinite distance."

Apparently, such collimating lenses came later. The main reason for putting the bubble out by the horizon mirror in the "Gago Coutinho" system was so that no special focusing was required (for younger eyes, I suppose). Also, the original Gago Coutinho sextant apparently had a sight tube only, not a telescope, which made it that much easier to dispense with a collimating lense. Presumably the Plath implementation of the design added a collimating lense since those sextants had telescopes.

Apparently, the Gago Coutinho bubble system worked quite well. In an article I found on the web (in Portuguese, which I can read enough for this sort of thing), it's reported that Gago Coutinho found an accuracy of +/-10' for individual sights and +/-3' for averages of 7 sights. However, since this sextant also permitted a direct view of the horizon, it's unclear whether these are "pure" bubble observations. It's possible that the combination of the bubble nearly superimposed on the somewhat vague sea horizon visible at altitude was better than either one alone at the altitudes at which airplanes were flying c. 1920.

-FER

----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------