NavList:

   

Reply

Alexandre Eremenko wrote:

> I also doubt that the rigorous "least square" method
> is of any use in practical navigation.
> I think the last chapter of Chauvenet is aimed at astronomers,
> who have a lot of time, and make VERY precise observations.

Well, the US and UK governments seem to think otherwise. A least square
based sight reduction algorithm is the only one that's offered for use
with calculators in the Nautical Almanac  (Concise sight reduction
tables for pencil and paper computation are also included, but they
don't preserve the full 0.1' nominal accuracy of the ephemeris).

Chauvenet wrote in 1860 and the appendix is clearly aimed at astronomers
(as is the whole book). But then, he does not mention averaging of
altitudes either, does he? The quote according to which Chauvenet
recommends averaging up to six observations presumably comes from the
text that George Huxtable made available to us. This is clearly in
reference to lunar distances.


> P.S. Celestaire has a wonderful device called datascope.
> According to their description (I don't have it) it records
> automatically the time and altitude.

Presumably, Alexandre is referring to the KVH datascope. This is indeed
a very useful device. Basically, it is a fluxgate compass with
non-inverting 5X30 telescope. I normally have it on me for look out, and
use it particularly to resolve "constant bearing" situations with other
vessels. I prefer it over binocculars with integrated compass, because
it is not as heavy, I can operate it easily with one hand and I can
store the bearing for later comparison. The latter feature is
particularly useful when you need to keep track of several vessels at
the same time.

Although I have the mounting bracket for my Astra, I have used it in
this way exactly once. It just makes the sextant heavier and bulkier and
I am more comfortable with the standard scope. The advantage of being
able to record the times is marginal. It does NOT record the altitudes
(it does not know them). You still have to jot them down and later match
up the altitudes with the stored times, which is error prone.

Its use in finding the stars is more than questionable. I see no point
in precomputing altitudes and azimuths, as they change rapidly and it is
difficult to anticipate the exact moment at which the star will be out,
the horizon clear, the clouds out of the way and the Genua in the right
position. Like every hand bearing compass, the fluxgate has its error
margin as deviation changes in different places of the boat. So, the
azimuth is not too reliable and you still have to scan the horizon for
the object. It's just as easy with the conventional telescope to pre-set
the altitude as best as you can and search the horizon until the object
comes into view. (I once shot Jupiter in daylight that way, just to show
off for the rest of the crew. They could not figure out what I was
doing, staring into the empty sky. Needless to say that this kind of
exercise is of limited practical value: It helped a lot that I had my
GPS position.)

If you need to confirm the identity of a star, it's better to take a
distance to a known one. This is a little more work, but more precise
and independent of time. Practically speaking, all this is useless
busywork. Day after day the same  four or five bright stars (or planets)
come out in the evening in exactly the same sequence in exactly the same
place. After the third day out on a passage, you are able to identify
the stars by which sail is in your way when you try to shoot them.

Herbert Prinz

   

Reply