NavList:

   

Reply

Alex wrote, in Navlist 2647, for which I've changed the threadname.

"I am impressed with your list of error in Moscowitz papers."

Well, part of the stuff Moskowitz covers is rather a hobby-horse of mine;
the place of Mayer and his mariners' circle, in the development of the
sextant. On that topic, I had the benefit of a bit of special knowledge,
that clearly Moskowitz didn't.

But I have this weakness in my character, which I have tried, but failed to
suppress: that I can't read anything without doing my best to probe it for
errors, or pick holes in its arguments. It's a negative, and rather
destructive, trait, that you can see evidence of in some of my Navlist
postings. It's just the same when I pick up a newspaper, which never fails
to provide fertile ground.

Those Mostowitz papers, in spite of their (presumed) factual  errors, are
well worth reading, however, and I should have made that clear.

Alex wrote-

"Cassens-Plath claims 10" and SNO-T claims 12". But this is in their
advertisements, not verified by any independent testing."

Presumably, that difference related to the way they were marked, as much as
the precision of marking. I mean, that the Cassens-Plath was divided into
sixths of an arc-minute, and the SNO-T into fifths. Is that guess correct?


I presume that the paper Alex mentions,  about the sextant testing equipment
at Kew was one that appeared recently in the Bulletin of the Scientific
Instrument Society. Going from memory, those collimators were originally
placed according to a theodolite at the centre of the rig, so everything
depended on how well that was divided. Which leads to further questions
about how that was calibrated...

The Vernier sextant that I have (labelled Sewill, but I reckon it to be a
badge-engineered clone of Heath) happens to have a Kew certificate dated
1920, which showed the error, at all points at multiples of 15 degrees
across its scale, to be 0 minutes and 0 seconds. It doesn't tell me the
limits of that measurement; that is, how big a deviation has to be to be
shown as greater than 0' 0". I am guessing that it means they have detected
no error greater than 5", and would like to be more certain about that. That
was the tested error then, but what is it now, I wonder? I don't regard that
zero deviation as rendering the instrument any more precise than another
calibrated instrument which has non-zero deviations, as long as the
navigator, when doing precise work, takes heed of those deviations pasted
into the box.

As for the Dutch octant that Nicolas de Hilster kindly posted, it's
remarkable, in my view, in three respects.

First, that an instrument could have been so crudely divided as to show
errors of twenty-odd minutes of arc, which is worse than you would expect
from a modern schoolboy's protractor.

Second, that its maker should then have been so up-front as to actually
provide a table of those deviations. Many makers, I suggest, would have been
ashamed to do so. It shows that for all those enormous errors, it wasn't out
of line with what mariners of the time expected from such a wooden octant.
At least, that table rendered the octant usable for getting latitudes.

Third, that Holm himself created a printed form for entering such details,
very similar to later calibration certificates. I wonder how his checks were
done?

George.

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


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

   

Reply