NavList:

   

Reply

Having been on a steep learning curve with the restoration of this sextant I 
have to confess to having been barking up the wrong tree with the varibility 
I found in readings of up to 25 minutes of arc.  Though it makes me feel 
foolish to admit publicly , it is of importance to others to know this is 
possible,  so report it here.

The one factor which I assumed was an impeccable reference - the bubble - is 
the cause of the variability.

This was proven in calibrating the sextant after dismantling completely, 
repairing, cleaning, oiling and re-assembly.  The end-float in the worm 
assembly which blamed, which in fact was only slight, was removed completely 
with a shim washer and thus all engineering possibilities of the variability 
were eliminated.

I am now bemused at not thinking of the obvious all along - the bubble itself 
is sticking slightly,  which is surprising as it appears to be very lively 
indeed, and much moreso even than my utterly reliable MkIX sextant which has 
a telscope magnifying the bubble too.

This sticking could be seen when calibrating, for when the sextant is tilted 
forwards and the collimating target examined, one reading is obtained; and if 
the sextant is then rocked backwards and the target examined again there is a 
slightly different reading.  Tilt the sextant forwards again and a similar 
reading (but not necessarily the same)to the first is obtained.

The bubble size I was using on this sextant was most probably too small. It is 
a magnified image of the bubble one sees, so quite a small bubble fills the 
viewing circle aperture making you think it is an adequate size. I was using 
a bubble diameter about one sixth of the diameter of the aperture, or perhaps 
a bit less. (i.e. one could use six side by side across the diameter of view.

The Hughes IX sextant has 'tramlines' in the centre and recommends a bubble 
slightly less than these which is about a quarter of the aperture.
Increasing the bubble diameter in the A10 to around a quarter of the view 
circle has improved matters considerably.  Averaged over four minutes, and 
with care, sun shots now result in position lines within a couple of nautical 
miles or so. (where they should be!).
The down side is of course, as the bubbe is increased, assessing the centre is 
more difficult and allows for more variability in this.  A compromise is 
required.  I find a quarter of the view diameter is about optimum, which with 
the A10 means a physical bubble size of about 4.5 mm.

When the bubble chamber glass was being cleaned I noted the power of the minus 
(concave) lens it produces in air was 6.5 dioptres.  With crown glass of 
index 1.52  this means the radius of curvature of the concave surface of 
about 52 mm. (a fairly steep curved surface).

It has been an intersting experience restoring this sextant as I have learned 
quite a lot about potential error related problems involved, both engineering 
and physical with the bubble.  I have to say I am annoyed with myself for not 
even thinking of the obvious straight away, but one expects the bubble 
reference to be an absolute, and utterly reliable - but not so if it is too 
small.  That is the main lesson I have learned.

Douglas Denny.
Chichester. England.


--~--~---------~--~----~------------~-------~--~----~
Navigation List archive: www.fer3.com/arc
To post, email NavList@fer3.com
To , email NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---

   

Reply