NavList:

   

Reply

Paul Hirose wrote-

>Thanks to those who replied to my query on shooting lunars.
>
>It was a little surprising to see Carl Herzog's statement that the
>choice of whether to use the horizon glass or index mirror for the
>Moon is mainly a question of physical convenience for the navigator.
>I'd expect the index shades to be much darker, since they have to
>make the direct rays of the Sun safe to view. (Remember, I own a
>bubble sextant but no marine sextant.)
>
>I've seen old engravings of quadrants, but never noticed the second
>sight line George Huxtable mentioned. But in "From Sails to
>Satellites" by J.E.D. Williams (excellent book!) there is a photo of a
>1750 quadrant by J. Bird. That instrument appears to have two horizon
>glasses, each with its own index mirror shade glass! I don't think
>this is what George is talking about -- the horizon glasses (if indeed
>that is what they are) are parallel as well as I can judge. Doesn't
>seem possible one could be used for a back sight.

Here's my reply-

Yes, J.E.D.Williams' book "From Sails to Satellites" (an excellent book, I
agree with Paul) was indeed one of the texts I had in mind when I wrote-

"In many books on the history of navigation, the Hadley quadrant is shown,
either as a photo, an old engraving, or a modern diagram. In all cases, a
second peep-hole and horizon mirror is shown, placed below the normal line
of sight and facing in the opposite direction to the "normal" horizon
sight-line, with the second horizon mirror tilted to look up at the index
mirror. Yet in none of these modern texts that I am aware of will you find
any description, or even any mention, of this second sight-line."

The other books I was thinking of when I wrote the above were-

Charles.H. Cotter, A History of Nautical Astronomy. (1968)
E.G.R Taylor, The Haven-Finding Art (1971 ed.)
Derek Howse, Greenwich Time and the Longitude. (1997)
National Maritime Museum / Royal Greenwich Observatory pub. "Man is not
Lost." (pamphlet, 1968)


There are some snags about the photo (from the library of the Science
Museum, London) of the museum sextant in Williams' book. The picture (or,
more likely, its reproduction in the book) is rather muddy, and its details
are unclear. The instrument appears to be incomplete, in that there's no
peep-hole or sighting-tube, for viewing the normal, forward, horizon, where
you would expect to find it, on the right arm of the A-frame. Instead,
there just appears to be a clamp screw, above the point where the sight
would have
fitted. At least, that's how interpret that rather indistinct image; I
wonder if Paul agrees?  I agree with Paul Hirose that as far as one can
tell, the components on the right arm of the frame do not look right for
observing a
horizon viewed from left to right across the sextant. I tentatively suggest
that this might be due to incorrect assembly of the parts by someone who
was unfamiliar with the use of the instrument. Perhaps someone else can
propose a better explanation. On my next visit to London I will try to take
a look, if it's on view at the Science Museum.

There's no reason to believe that every version of the Hadley quadrant,
which were produced by several instrument makers, was identical to the
Hadley original.

I think the best illustration of how a Hadley quadrant was assembled is
shown as a cover illustration of the "Man is not lost" pamphlet. This was
an engraving taken from G. Adams, Geometrical and Geographical Essays,
1797. Within the covers can be found a photo of a Hadley quadrant, made by
the same G. Adams, and the details of the photo and the engraving accord
remarkably well. Also, they agree with the text of the Ludlam pamphlet of
1790. So I think we can believe what they tell us. Unfortunately, there's
no detailed description in the text of the modern pamphlet.

With apologies to those who object to receiving attachments, I have
included a copy of the Adams engraving because it is so clear and useful
(and its parts are lettered).

In normal use with a "forward" horizon, the observer's head is to the
right, looking through the peep-hole I. Light from the observed body
arrives at the index mirror E, is reflected into the horizon mirror F,
where it reflected again to the eye. The observer also views the horizon
directly, past F.

With a "backward" horizon, the observer's head is to the left, looking
through the peep-hole H, toward the horizon at the right, past the mirror
G. Light from the observed body arrives at E (from over his head),
reflecting down to the mirror G (which is angled the opposite way to F) and
so to the observer's eye.

The "backward" horizon mirror G apparently allows its view of the horizon
via a slot. Why this mirror should be different from mirror F isn't clear
to me. Perhaps further study of Ludlam will explain it.

You can see that for the "reversed" view, the peephole H is placed much
lower (further from the pivot) that the corresponding "forward" peephole I.
This allows more clearance for incident light to pass over the observer's
head into the index mirror. It was only because these quadrants were so big
(18 or 20 inches radius) that the observer's head could be cleared in this
way.

>From the engraving, it's clear that the light path from E to G is angled a
bit differently from the light pathe E to F. This is necessary, so that
mirror F doesn't obstruct the view from G. There's no problem about this
difference in angle, as far as I can tell, because it can be compensated by
suitable angling of G. It must imply that the frame of the quadrant has to
be tipped slightly differently, in a fore-and-aft direction,for "forward"
and "backward" horizon observations.

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

Earlier in this interesting correspondence R Winchurch wrote-

"As I understand it one measures the angle between the moon and selected
planets (Jupiter and ?) and certain stars."

The answer is that historically the angle that was measured was between the
Moon and the Sun (but this was unmeasurable near New Moon and near Full
Moon) or between the Moon and certain bright stars or planets that were
near the Moon's path
around the sky (but these were unmeasurable near New Moon). There were
always several nights in each month near New Moon when the Moon couldn't be
seen at all, and so no Lunars could be taken; a factor that could prove
embarrassing if it occurred at a critical moment such as a landfall.


For a good modern text on Lunars (and much, much, else), which isn't afraid
of some maths, I strongly recommend Charles H Cotter, "A History of
Nautical Astronomy", 1967. It's hard to find, but possible.

Lunars were used in two rather different ways, which may be causing a bit
of confusion.

First, you could take a lunar sight just to give you a time-check, to allow
you to assess the error in your chronometer. It was always important to
measure the Lunar distance (the angle Moon-Sun or Moon-Star) with as much
accuracy as possible, because an error of 1 minute in this angle gave rise
to an error of about 30 minutes in the resulting longitude; this is the
great weakness of the Lunar method.

Then it was necessary to make certain corrections for the effects of dip,
refraction, parallax (this was called "clearing the lunar distance"), and
as these varied somewhat with the altitudes of the observed bodies it was
necessary to know those altitudes. But as the corrections were only small
ones, it wasn't important to measure the altitudes very precisely, or at
precisely the same time as the Lunar distance. For the purpose of these
corrections, the altitudes could even be predicted from tables rather than
measured.

 Perhaps a Moon-Star Lunar was measured at night to correct the ship's
Greenwich time. Then, the next day, Sun altitudes could be taken at various
times to obtain latitude and longitude. To do all this, it was necessary to
have on board a deck-watch that was good enough to keep time to within a
few tens of seconds over a few days. Not a chronometer, then, but a
reasonable sea-going clock or watch.

The second method was what you had to use if you didn't possess a
timekeeper at all; for centuries, mariners had nothing better than a
sand-glass. Then at the same moment (or nearly so) that the time was
obtained from the Lunar distance, you would derive a position by measuring
the altitudes of two bodies in the sky and crossing their position lines.
What bodies would you choose? Of course, there they are, the same two
bodies that you used to derive the time, the Moon and the selected star (or
the Sun). But now, the altitudes have to be measured precisely, because now
they are not just required for making a correction, they are needed to give
a precise position.

Several myths exist about the requirement to have three observers with
three sextants, carefully synchronising their observations of the two
altitudes and the Lunar distance. This is largely nonsense, because the
longitude measurement from a Lunar is so inaccurate anyway, that precise
timing is not going to help. In reality, it's a one-man job, if he's smart
about it .

George Huxtable.





Attachment converted: 320Mb HD:Hadley quadrant (TIFF/Cit2) (0000799C)
------------------------------

george@huxtable.u-net.com
George Huxtable, 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
Tel. 01865 820222 or (int.) +44 1865 820222.
------------------------------

------------------------------

george@huxtable.u-net.com
George Huxtable, 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
Tel. 01865 820222 or (int.) +44 1865 820222.
------------------------------

   

Reply