NavList:

   

Reply

Brad has been asking detailed questions about how to obtain or correct the 
index error of an octant (=quadrant) at sea, when a backsight was being 
taken for measurement of angles greater than 90º.

To which Frank responded, in 19 March,

"... this problem of getting the index correction for the back sight is 
actually identical to the problem of finding the arc error at some odd spot 
on the arc of a normal sextant. And therefore any of the solutions that 
we've discussed would also work. So, for example, you could do a star-star 
sight for any angle in range for the back sight."

===============

The kindest words I can find for that proposal is that it's a landsman's 
notion.

Measuring star-star distances isn't easy, even on land, due to stars being 
so numerous, and one star looking so much like another. It isn't easy to 
locate the intended star-pair.

And measuring stat lunars isn't easy, either, especially at sea, but at 
least there is one unmistakable object that can be kept in view; the Moon, 
which the observer can swing his view-line about, to locate the other.

But we come, next, to the proposal for measuring star-star distances, 90 
degrees apart, at sea, where nothing is stable. Has anyone ever done that? 
Has Frank? I certainly haven't, and wouldn't even attempt it. Has anyone, 
on this list? In the whole of navigational history, has anyone ever claimed 
to measure a star-star angular distance, from the deck of a vessel at sea?

I don't claim that it's inconceivable, but it would call for such 
extraordinarily calm conditions, that it isn't a practical navigator's 
technique. If Brad proposes to try it out, I wish him well.

==================

I've pondered about this, when evaluating some of the Lewis and Clark 
observations, made inland. Measurement of star-star distances would have 
been quite beyond their powers, in my view. But on land, there were other 
options open to them, depending on the details of the local topography, at 
the time. A procedure, which would call for a rather open landscape, would 
call for two distant landmarks such as identifiable trees, a mile or more 
away from the observer to minimise parallax errors. These trees would need 
to subtend around 90º at the observer, which he could adjust be approaching 
or receding from them. Then a comparison of horizontal angles measured 
between them, by foresight and by backsight, would provide the desired 
information.

At sea, this difficulty of checking index error in back observations 
provided a serious limitation to the use of backsights for measuring 
large-angle lunars, which the Dollond modification was intended to bypass. 
Navigators, who had attempted lunar distance observations using only their 
octants, would find its limitations restrictive. That's an important 
reason, in my view, for rapid acceptance of the sextant, for those 
navigators who indulged in ocean travel,  calling  for knowledge of 
longitude. For those voyaging shorter distances, an octant remained 
perfectly good enough for altitude navigation.

George.

contact George Huxtable, at  george@hux.me.uk
or at +44 1865 820222 (from UK, 01865 820222)
or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
----- Original Message ----- 
From: "Frank Reed" 


Brad, you wrote:
"I think I may have the answer to the index error for backsights.  Consider 
NAV1268 at the National Maritime Museum in Greenwich.  It is a backsight 
Octant.  As is typical of Octants, it is a vernier type, with the arc 
from -2 degrees to +101 degrees.  Due to the length of the vernier itself, 
however, the octant can only measure to 95 degrees.  The measurement beyond 
90 degrees is the key.  Using a FORESIGHT, measure the altitude of a star 
whose apparent altitude is greater than 85 degrees.  Why greater than 85 
degrees?  There is a doubled region between 85 degrees and 95 degrees, in 
which we can measure the altitude of a star with EITHER a foresight or a 
backsight observation.  Since it is possible with either method, we must 
perform the observation with BOTH methods.  In knowing what the altitude is 
with a foresight observation, we therefore know what the vernier must read 
for a backsight observation, given the same star.  Set the octant's vernier 
to the arc for a backsight observation and adjust the backsight horizon 
mirror until the altitude is correct. "

That would work, but it wouldn't be easy, and good luck getting a star in 
the right place! If you haven't already said so in a post I've missed, this 
problem of getting the index correction for the back sight is actually 
identical to the problem of finding the arc error at some odd spot on the 
arc of a normal sextant. And therefore any of the solutions that we've 
discussed would also work. So, for example, you could do a star-star sight 
for any angle in range for the back sight. Or, if you have three 
"lighthouses" in a row (far enough out on the horizon so that they fall on 
a great circle), let's call them A, B, and C, and you can measure the 
angles less than 90 between A and B and then the angle between B and C. 
Then of course the large angle between A and C is known. So you measure it 
as a back sight and any error is the index correction.

-FER

   

Reply