NavList:

   

Reply

My own experience confirms everything that Brad wrote. I posted many messages on this on
this list in 2005-2007, but let me re-tell the story shortly.

When I purchased my SNO-T it was "new" from the factory (in fact 12 years old 
but in factory packing and never used).
It had all documents including the factory "arc correction table" which made absolutely no sense:
it showed constant +10" for the whole arc. It is clear that the person who 
filled this table just did not understand what she was doing:-)

I also have this Russian book which recommends to calibrate the arc with 
star-to star distances, and gives a short table
of stars for this. However the book does not even mention that to use this 
table one has to correct the distances
for refraction, which leads essentially to the same formulas that are used for 
Lunar distances. (All this was explained to me on this list).
I conclude that the author of the book just did not test his own advise:-)

So I programmed these refraction correction formulas on my computer and 
started to measure star distances, which I did systematically
during 3 years (many of my results were posted on this list).  My distances were
generally within 0.2'-0.3' of the true ones, but sometimes I had larger 
errors. No conclusions about the arc errors could be
derived from them.

At the same time I had my sextant calibrated by the two places which still did 
this at that time: Freiberger and Cassens-Plath (I visited their pants).
Cassens-Plath produced a certificate for my sextant which looks like their 
standard certificate: "This instrument is free from error for practical use".
Freiberger produced a certificate filled with some numbers (I was actually 
present at the process of measuring at Freiberger, while Cassens Plath did 
not permit me
to see the process). The table that Freiberger produced looks like this
0    10     20     30      40      50      60       70     80       90     100     110       120
0    -4"     -4"   +10"   +6"     -1"     +4"    +5"    -5"      +4"    +4"     +3"       +14"
I measured enormous number of star (and Lunar) distances, but the results show 
no correlation with this table.
Of course, the table does not contradict the Cassens-Plath statement, and on 
the other hand it has no practical use.

The general conclusion is that modern manufacturing process produces arcs 
which are "free from error for practical distances",
and you cannot detect any non-uniformity by observing star-to-star distances. 
On the other hand, from historical literature I know that
in 18th century people indeed were able to correct the arcs by observing star 
distances. My conclusion is that 18th century arcs were
just inferior, and had larger errors, at least some of them.


Alex.

   

Reply