|
|
|
|
|||||
|
||||||
From: George Huxtable (no email)
Date: Sun Dec 04 2005 - 07:23:29 EST
Willem Piccer wrote, about dip-
>In an old study-book I found this method of finding the dip.
>I sent an attachment to Robert Gainer at
Comments from George are interpolated below.
It's good to see that Robert Gainer's Nav-l blackboard works. As we get
accustomed to it I can see it being more widely used.
That description of a dip-meter is new to me.
Willem doesn't quote a reference for his "old study book", or his diagram.
Where have they come from, please, in what language, by which author, and at
what date?
>Prism for the measurement of the dip
>
>On some sextants a device can be placed, which gives an opportunity to
>measure the angle that the directions towards two points on the >horizon
>diametrical opposite to each other make
>Supposing that in both directions the dip is the same, one finds ( after
>apllication of the I.C.) double the value of the dip.
>For this purpose a rectangular prism A is put in front of the non-silvered
>part of the horizon mirror.
The "rectangular prism" is explained if you look at the drawing. It's a
piece of glass, with a rectangular cross-section, which has its two ends
made as reflecting faces bevelled at 45 degrees, so as to reflect light back
toward where it came from (and displaced). It could just as well be a pair
of mirrors rigidly and accurately located to each other, rather than a prism
To work as proposed, those two reflecting faces need to be very precisely
made at 90 degrees to each other: (but see below)
>This prism has a foot with a square pin which is slipped into a box which
>is placed in the frame of the sextant
>
>The prism can easily be placed or removed.
>To measure the dip, a prism B is screwed on to the telescope or the
>eye-tube, by which it is possible to see through the telescope the pictures
>which are formed in the silvered part of the horizon mirror and in the
>prism A
That seems rather similar in effect to a "star-diagonal", as used with the
eyepiece of an astronomical telescope, allowing the eye to look in a
direction at right angles to the telescope, and to the plane of the sextant.
That gets the observer's head out of the path of light that will enter the
prism A. Inevitably, it moves the eye further from the telescope, and that
may
limit the field of view, depending on the telescope's design. B could be a
45-degree mirror rather than a prism. No great precision in its alignment is
called for.
.
>To be able to keep the eye during the measurement on some distance from he
>plane of the sextant ( which is desirable, because otherwise the head of
>the observer will intersept too much light) sometimes a tube is mounted on
>the prism B
That makes sense, to fix the eye in the right place.
>One keeps the instrument vertical such that one has the telescope left and
>the horizon mirror right, meanwhile looking sideways through prism B
>The instrument is kept in such a position, that the picture of an on the
left side lying part of the horizon through reflection is seen in prism A
>( in the picture K2 gives the direction of the for the observer on the left
>side lying horizon)
>By moving the alidade one brings the double reflected image of an on the
>right side lying part of the horizon in line with the first mentioned image
>( K1 shows the direction of the horizon on the right side; K1 does not go
>through the prism you will see it next to it!)
>Through the rectangular prism A the beam K2 will be reflected parallel to
>itself, so that the reading of the instrument, after I.C. has been applied,
(see below, about the index correction, IC)
>gives the angle between K1 and K2. This angle is double the dip.
>In general one finds a negative reading.
>I have never seen this and hope the translation is good enough to give you
>a good understanding how it works.
It certainly is.
Presumably, you are meant to observe the angle measured between the horizons
in the two opposite directions, then remove the prism and measure the index
correction (IC) in the ordinary way. For that to work, the prism must bend
the light very precisely through 180 degrees.
I think there's a better way to do that job. Don't measure an IC at all.
Instead, after the first observation, turn the whole instrument upside down,
and measure again the angle between the two horizons. If it was a negative
angle before, it will be a positive one now. Subtract the two readings, and
the result will be four times the dip. Done that way, there's no requirement
for the prism A to bend light through exactly 180 degrees; any such error
disappears in the subtraction. Does that make sense?
Thanks to Willem for bringing this to our attention.
George.
contact George Huxtable at
or at +44 1865 820222 (from UK, 01865 820222)
or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
|