NavList:

At this point I am somewhat better off than I was going in, but the
experimental results are puzzling at best. Thanks to all of you for your
input.

Fred made a good point when when he mentioned his 4sd's were over, and he
trained his eye to lower it and bring it in line.  The limiting factor in
separating (on the arc) and merging (off the arc) is the bright area that is
visible when overlap occurs. I have always been more comfortable/confident
of separation for that reason.  When merging, I will occasionally take a
reading, nudge it another 0.2' towards merge, and determining if overlap is
achieved.  If so, use that initial reading.  If not, too far away and repeat
the nudge.

George also makes some excellent points.  Size does not matter.  If you
doubt that, change scopes.  I submit that even a change of focus will affect
size.  George, Frank et al also point out refraction is not a constant.

At this point I feel I can pretty well rule out a shift from shades.
Further testing with sun/shades and moon/no shades 4SD will be needed to
confirm that.

I cannot be certain there is no frame/mirror gravity effect from my trials.
I can be certain that the shift from one eye to the other shows opposing and
roughly offsetting results.  To totally rule out flex, both of my eyes would
have to be almost identical, the only difference being the astigmatism being
rotated 90d. Or the trials would require one perfect eye.

I do feel confident that the astigmatisms play a large role in the shift
from tangent horizontal vertical to tangent horizontal.  That I can see, and
to some degree measure.

Regarding Fred's training of the eye, and George's correct IMHO statement
regarding size, I agree.  But let's assume the eye is not accurately
reporting the information from the scope.  As an analogy, let's assume I am
trying make the yolks of two eggs tangent.  If I can see the yolks, no
problem.  But if my eye can only see the egg whites due to astigmatism, then
I will see tangency too late (figure to high) on the arc, and too soon
(reading too low, 60-reading too high) off the arc.

Unless I am missing something, what happens on the arc due to eye distortion
must also happen off the arc. Here is where the data falls apart. If my left
eye is distorting the scope input (making it too large along the Y axis, and
smaller by comparison) in a vertical position both on and off the arc, that
should not affect IE, and 4SD should be too large vertically.  Depending on
any x-axis distortion, 4SD could be too small, just right, or too large.

For example:

IE/IC as determined by horizon, building tops, a star etc = 0.0'

4SD = 63.2

In a perfect world, I would read 31.6' on the arc, and compute 31.6 minutes
off the arc (direct reading 28.5' off the arc, 60 - 28.4 = 31.6)

Now lets assume my eye sees tangency at 31.9' on the arc due to distortion.
An overshoot. It should also see merge happen prematurely (28.1'), so the
direct reading was too low and the 60-direct too high (60 - 28.1 = 31.9).
Since IE is the midpoint, it should remain at 0.

4SD should increase by the sums of each overshoot, or 0.6' ( 0.3 + 0.3).
Therefore 63.8'.  The data does no support that.

From a practical standpoint, I believe I can stop worrying about eye-to-eye
differences and any frame/mirror flex between horizontal and vertical where
IE is concerned.  That can be removed with a constant.  For example, if I
make a round(s) of solid IE checks with a natural horizon etc., and a solid
set(s) of IE checks with the sun and there is any difference due to flex, I
can compare and arrive at a correction factor.

My concern at the moment is the discrepancy between the data.  I feel I need
to go back with one eye, and redo in the horizontal mode very carefully and
deliberately, using every trick and sanity check I know.  Wait until
conditions are better than 90d F. While I had been careful to bring the
temperature of the instrument up to ambient temperature in the shade, 10-20
minutes in the sun may cause problems.  As might refraction from 30.05 to
30.14" BP and 90d F. Keep the body high, and no more marathon runs.

Agreement between on-the-arc figures from trials correlate very well, so
off-the-arc arc seems more likely to be the problem (but has limits on how
much I can overshoot the off-the-arc direct reading before I see overlap).
That however is the kind of thinking that gets one into trouble.

In short, check mirror alignment, slow down, and double check each
measurement.

Bill



--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to NavList@fer3.com
To , send email to NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---