 Check out the bookstore at IRBS.com |
|
|
|
|||||
|
||||||
From: Frank Reed (no email)
Date: Tue Oct 18 2005 - 19:52:49 EDT
SEXTANT TELESCOPE COLLIMATION
If the telescope of a sextant is not exactly parallel with the instrument's
frame, the instrument will measure large angles to be too large. The error
is
proportional to the square of the telescope's tilt and proportional to the
tangent of half of the measured angle. I've been testing a group of sextants
in the past two weeks, and all of them had significant telescope collimation
problems that could easily yield errors of one to ten minutes of arc. So how
do we test it, and how do we fix it?
To test for telescope collimation problems, measure a large angle --ninety
degrees or more. Any star-to-star distance will work for this, but they
should
be distinguishable (one considerably brighter for example). A terrestial
angle will work, too. Bring the two objects together in perfect contact near
the
center of the telescope's field of view. Then rotate the sextant so that the
two stars move across the field of view towards and away from the
instrument's frame. The stars will separate slightly. If the telscope is
properly
collimated, they will separate symmetrically. That is, you will find that
the stars
are a couple of minutes apart when the stars are on the right side of the
field of view, in contact at the center of the field of view, and a couple
of
minutes apart again when they are on the left side of the field of view. If
the test reveals that the telescope is not correctly collimated, then it
should
be adjusted.
To collimate the sextant telescope, you'll need a large room or hall twenty
feet or more in length. You'll also need a small straight telescope, like a
finder scope for a larger telescope or an old-fashioned sextant telescope,
or
alternatively a laser level (there are sufficiently accurate levels
available
in hardware stores for $10 to $15). You'll also probably need a couple of
blocks about half an inch high and as identical as possible to support the
leveling scope.
Set the sextant on its side on a table at one end of the room. Swing the
index arm as far out of the way as possible. Place the blocks on the
sextant's
arc (or frame if it's flat enough) and then set the leveling telescope or
laser level on the blocks. Now turn on the laser or look through the
leveling
telescope. Assuming this telescope has crosshairs in the field of view, you
can
now mark a spot on the wall at the far end of the room that will be your
reference point for collimating the sextant's telescope. If you're using a
laser
level, you've already got your reference point. Take a look at your sexant's
telescope and estimate how much farther it is away from your sextant's frame
than your leveling scope or laser level. Let's suppose it's half an inch.
Now
look through the sextant's scope at the far wall. The center of the field of
view should be a spot that is that same half an inch higher than the the
reference point. But since we already know that the telescope is not
correctly
collimated, it's probably aimed a considerable distance above or below the
correct level. If you find it difficult to locate the center of the field of
view, draw some horizontal lines on the wall and count off. Now adjust the
telescope's aim with the screws or knobs for this purpose on the sextant
until the
center of the field of view is exactly aimed half an inch (or whatever
height difference you find) above the reference point from the laser or
leveling
telescope. You should try to get this alignment to the nearest inch or so if
the far wall is twenty feet away. If you can get the alignment accurate to
this
level, the error in measured angles will be less than 0.1 arcminutes even at
measured angles of 120 degrees. That is, a tilt of one inch in twenty feet
(20 ft 4 inches, to be precise) is essentially perfect for all sextant
angles.
Generally, the error, dh, in the measured angle, h, from telescope tilt, T,
is dh=T^2 * tan(h/2). Or, if x is the aiming error measured on the wall and
D
is the distance to the wall, then the error in minutes or arc is
dh=3438*(x/D)^2*tan(h/2).
It turns out that this is a relatively easy procedure, and it's well worth
trying. Many sextants include small screws or knobs that allow the
telescope's
inclination to be adjusted. But if your sextant doesn't, you'll need to
improvise some sort of "shimming" system.
-FER
42.0N 87.7W, or 41.4N 72.1W.
www.HistoricalAtlas.com/lunars
|