NavList:

   

Reply

On Sat, 06 Mar 2010 03:01:26 -0700, Douglas Denny
 wrote:

> This paper would seem to be the final and definitive answer to the
> lunar/star issue of irradiation. It is particularly exciting to me, as
> it is the experimental method I used myself all those years ago, but in
> a highly refined and systematic fashion - done properly in other words.

Umm . . .  Maybe not.

       From that paper:

**********long quote********
"The subject's task was as follows: The planetary disk and the reference
star appeared in the sextant telescope. The star was offset a few minutes
of arc from either the upper right limb or the lower left limb.  The
subject rotated the sextant vernier knob to bring the star image to the
limb and oscillated the star to achieve a more accurate star-limb
tangency. When he was satisfied that this had been achieved, he pressed a
button that automatically recorded the angle to which the sextant had been
set.

"He then backed the star away from the limb and repeated the process. When
he had completed eight of these sightings, he moved the star image to the
opposite limb and again took eight sightings.

"For each subject the star always moved in the same direction, i.e.,
toward the limb from the extradisk position. This was done in order to
circumvent gear backlash errors in the sextant. This effect was minimized
when a subject went past, or overshot the limb by requiring him to back
the star image away and start the sighting over.
The eight sightings on each limb were averaged. By subtracting the mean
for the upper from the mean for the lower, the apparent angular subtense
was computed ."
******end of long quote****

It seems to me this is equivalent to asking a surveyor with a theodolite
to take a set of eight readings approaching a well prepared point target
 from the right, and then a set of eight readings approaching the target
 from the left.  This is guaranteed to show the cumulative effects of
backlash in the readings, not circumvent the effects.

Always take the final reading *against spring force*, which is usually
clockwise on the setting knob, no matter where you find your initial
crosshair location.  In fact, some of the older texts even asked that you
rotate the instrument almost 360 degrees if you started off on the "wrong"
side, in order always to turn the setting knob in the same direction to
avoid backlash in the rest of the system.

Try it yourself with the best of targets: the difference of the average of
readings from each side of your readings on a point target *will be* the
measure of the backlash, and not the Zero expected by the author.

Further to that point; I think Bill (engineer) gave some figures (on the
effect of backlash in finding index error of a sextant) quite recently.

As to the star to star training, the procedure seemed to be different:

"All subjects, therefore, trained for 5 days. The training task was to
measure the angle between two simulated stars by superimposing their
images in the sextant telescope's field of view."

I see no mention of irradiation error or effect mentioned there, but only
when the different simulated magnitudes of stars were compared to a bright
disk.

Also, you have neglected to mention one or two important sentences after
your own quote:

"In the region of brightness comparable to that of the moon seen from
outside the earth's atmosphere, about 900 foot-lamberts (ref. 8),the error
is almost 40 seconds of arc.

"According to figure 5, a 2.5 log unit neutral density filter would reduce
this to about zero."

As to the sentence after that:

"The temptation to use a bright, high contrast planetary disk as a target
in navigation sightings apparently should be avoided."

. . . This would seem to be the reason for the common practice of shading
the sun and moon disks appropriately when taking sights.  I believe these
are available on the sextant used:

"The sextant used was the Plath micrometer marine sextant with the six
power (6~) monocular. The sextant was gimbal-mounted and the measured
angle was detected by a digital shaft encoder and recorded on a printer
(4). The encoder is described in reference 3 and shown in figure 4. The
subject's head was held steady by using a dental bite mold and plate,
adjusted for comfortable monocular (right eye) viewing."

I hope I haven't muddied the waters excessively by all this quoting.

--
Richard . . .


Using Opera 10

   

Reply