Making a sextant: Help appreciated.

[richardandtracy]

I wonder if someone may be able to help me.

Up until now I've been using a home made octant that's of very simple design with no telescope (uses 2 fine alignment wires to ensure I'm looking from roughly the correct position) and a vernier engraved on the end of the arm. In order to get the vernier to read an acceptable level of accuracy the whole thing is rather large (15" x 9") and heavy.

I want to have a bash at making a lighter, more advanced sextant with a telescope and micrometer vernier, but find there are a few gaps in my knowledge which I'd appreciate it if someone could help me fill.

My existing octant's vernier has a resolution of 2 minutes of arc. As currently envisaged the new one will have a resolution of 1 minute of arc on the micrometer scale and 0.2 minutes of arc on the vernier beside it. Is this enough? I could easily make it go to 0.1 minute of arc by extending the vernier, but is it worth it - basically can anyone guarantee that level accuracy in their reading?

I have to confess I'm mostly ignorant of optics (despite having made one lens from flat glass), so making a telescope for the thing will be somewhat novel. I know the dimensions & can calculate the magnification I want, but don't know how to use this to size & specify the lenses. Can anyone point me in the right direction (web sites, books or anything)?

Are there sextants with semi-transparent mirrors (rather than half silvered mirrors) so that the image of the sun can be seen imposed directly over the horizon? If so, how effective is it?

Finally - has anyone got, or know of, a photo of a sextant with an artificial horizon? I think I can envisage one (a vial part filled with glycerine), but seeing a photo would confirm my idea, or point me in another direction.

As to why don't I buy an Ebbco plastic one?
I know it would be much easier, but I am stubborn.
And I think this home made one will no worse, and probably better.
And it will be a fascinating exercise.
And winter's coming up so there is more time for things that don't involve getting soaked in spray.
And.. I won't go on [any more].

Thanks in advance for any help.

Regards

Richard.


<hr width=100% size=1>

 

[kilkerr1]

Hi Richard

I'm afraid I know nothing about optics or sextants and I don't know if this will be any help, but earlier in the year someone posted a link to a home-made sextant how-to-do-it-yourself website. He's got plans and sizes and all sorts - looks like all you need is a CD and some lego pieces. Amazing!

Have a look here for his do-it-yourself octant project.

Cheers.

<hr width=100% size=1><A target="_blank" HREF=http://kilkerr.members.easyspace.com/>Santa Teresa and other t'ings</A>

 

[alex_rogers]

My Astra sextant has a semi-transparent mirror as you describe and it works very well. Some people argue that they aren't needed, claiming that they make a sun sight which is already easy, easier and a start sight which is tricky, more difficult, as the mirror isn't quite as bright. I haven't found this and think the full mirror is much easier to use for both star and sun sights.

As for accuracy, I've never bother reading the sextant or doing the sight reduction with fractions of a minute. Offshore I regularly got fixes within a few miles from a small boat and the error was nearly always due to estimating the current or course badly when shifting the earlier sunsight - when you're that far from anywhere, a few miles doesn't matter anyway.

As for the artificial horizon, I think the easiest approach is to sight the sun reflected off the surface of a pool of liquid - apparently engine oil works very well. I had an artificial horizon which fitted on instead of the telescope and had a spirit level type bubble inside it - tricky to use and never seemed very accurate at all ( 10-20 minutes ).


<hr width=100% size=1>Alex Rogers
<A target="_blank" HREF=http://www.YachtsAtSea.com> www.YachtsAtSea.com </A>

 

[richardandtracy]

Thanks for the links, but I have seen these before - my octant is a metal one inspired by the ideas shown in the octant site you pointed to (along with a bit of wide interpretation and modification).

Regards

Richard.


<hr width=100% size=1>

 

[richardandtracy]

Thanks, that's interesting.
Transparent mirror. I do like the idea. Now all I've got to do is source one (fairly easy I'd think).
I'll probably stick to a 0.2 arcmin vernier, so that I can use it only if provoked by a wild desire to be wrong with extreme accuracy.

I'd not seen the reflective artificial horizon until a websearch this morning, and I can't quite get my head around how they work.
My glycerine idea was to use the under surface of the liquid between the meniscus' (in a vial at the lower mirror position) as a substitute horizon. But thinking further on it, unless the vial is held perfectly vertical, the level will be incorrect and cause an index error. This is soluble, but not easy.

Regards

Richard.


<hr width=100% size=1>

 

[AndrewB]

Sounds good. My experience is strictly as a user, plus building one out of cardboard from a National Maritime Museum kit, with the assistance of my son!

Regarding accuracy, unless you are skilled at machining in brass, I'd be amazed if you can do better than 1 minute. That's the best to expect from even good plastic sextants (despite the 0.2' vernier). To be honest, because of the greater effect of minor faults, I'd expect a homemade sextant to be less reliable than an octant, and that in turn less reliable than a back-staff.

The telescope isn't strictly necessary and if you can't cannibalise a 3:1 opera glass, a sight tube would be quite adequate.

Sextants where the images are directly superimposed generally use a dielectric beam converter (see for example a <A target="_blank" HREF=http://www.davisnet.com/product_documents/marine/manuals/mark25.pdf>Davis 25</A>). They are certainly rather easier to use than traditional half-silvered mirrors. Is this what you mean by a 'transparent mirror'?

Bubble sextants are cumbersome and not that commonly used at sea. A dodge for taking sun sights ashore where the horizon is blocked, is to reduce the sight to a reflection of the sun in a basin of water.

<hr width=100% size=1>

 

[alex_rogers]

The reflective artificial horizon is quite clever - the surface of the liquid ( originally mercury ) is horizontal due to gravity ( you need a reasonably sized surface so that it is flat - not curved due to surface tension ) so if you line up the reflected sun through the horizon mirror and superimpose the sun through the index mirror, you end up measuring twice the normal angle (you ignore the height of eye correction).

I've seen them for sale but as far as I can tell, they are just a fancy plastic wind shield. I've used engine oil in jam jar lid and it works quite well.


<hr width=100% size=1>Alex Rogers
<A target="_blank" HREF=http://www.YachtsAtSea.com> www.YachtsAtSea.com </A>

 

[richardandtracy]

I have a very nice rotary table I use for machining (brass included). It has a resolution of 10 arcmin out of the factory & a test cert verifying initial manufactured repeatability better than 0.25 arc min. With backlash elimination and improved verniers (which I've made) I can get a resolution of 1 arc min when machining. When being really careful about everything I reckon I can get a guaranteed positional accuracy of +/- 0.4 thousandths of an inch (+/- 0.01mm) on the linear dimensions, so little errors shouldn't build up much.

I won't claim to be skilled at machining brass, just bloody minded about making sure every possible error is eliminated where possible. This approach seems to get reasonable results (in the end). I have a feeling my new sextant won't be too inaccurate.

I'd rather have a telescope - they're something I don't know much about, and I'd rather like to remedy the situation.

Regards

Richard.


<hr width=100% size=1>

 

[richardandtracy]

Thanks, I may try one of these.

Regards

Richard.


<hr width=100% size=1>

 

[AndrewB]

Great! Will you do me one too? (nm)

/forums/images/icons/wink.gif

<hr width=100% size=1>

 

[MainlySteam]

At it again Richard!

Just on the micrometer part of your question - reputable metal sextants were available with resolution of 0.1 and 0.2 minutes of arc on the micrometer vernier. I say "were" because I don't know if any 0.2 minute ones are made now.

The compact Frieberger "Yacht" sextant was 0.2 minutes of arc and was considered plenty adequate and was in common use. While I would not ever claim to be an experienced sextant user, the little I have done makes me think that the comments that the accuracy of a sight will be limited by the boat's environment rather than the sextant itself are well founded. That is assuming you can make an accurate one, however, putting aside a minor deviancy associated with junk rigs, I have come to the conclusion that it is likely that you will make an at least reasonable go at it!

Regards

John

<hr width=100% size=1>

 

[Sandy.]

Hi Richard, Here is a site that I had a look at some time ago. Hope it still works. Site is www.tecepe.com.br/nav/xtantproject.htm. Good luck.
Sandy


<hr width=100% size=1>

 

[richardandtracy]

Re: Great! Will you do me one too? (nm)

Erm.. I think I'd better get mine right first!

Regards

Richard



<hr width=100% size=1>

 

[richardandtracy]

Thanks.
0.2 minutes of arc it is.
As for the vote of confidence, very nice thanks. My wife, however, always comes up with the accidental putdown of 'It looks like its been made by someone who knows what they're doing' in a surprised tone of voice..

Regards

Richard


<hr width=100% size=1>

 

[snowleopard]

as a dyed-in-the-woll computer user, i am well aware of the dangers of spurious accuracy. e.g. weigh 7 people, total is 1000lbs. therefore one weighs 1/7 of that, i.e. 142.8571428lbs.

well that's the answer on my calculator so it must be right!

when taking sights on a small boat you're lucky to get within a full minute of the 'correct' answer, then if you use the sight reduction tables the working is in places to 0.2 mins at best.

some claim to be able to get a fix within a mile but i'm happy if it's within 3m (i.e. 3.0 minutes).

as for the artificial hrizon using the reflection of the body in a bowl of oil or whatever, it works well but the angle you measure is double the altitude and you don't use the allowances for dip and refraction. index error is corrected before halving the angle.

<hr width=100% size=1>

 

[Mirelle]

Fred Rebell crossed the Pacific upwind from Sydney, Australia to the San Pedro, California, between 1931 and 1933, using a sextant that he had made himself - being skint and without any sort of workshop he happened on the good idea of using a hacksaw blade as the arc, reasoning that the division was likely to be reasonably accurate! (His boat incidentally was a retired Sydney Harbour 18 foot skiff, re-rigged and decked.)

May I differ very slightly from Andrew B and recommend that you use oil (does not much matter what, olive oil, 10W40, whatever!) rather than water for an artificial horizon (halve the angle on the sextant) because water ruffles in the slightest breeze.

My sextant, a Plath, has an artificial horizon attachment which is basically a 3 x 30 telecope with a bubble in the middle of it - you get the body in the middle of the bubble and that's it. I have absolutely no idea how it works and having just seen the price of a new one (far more than I paid for the sextant!) I have no intention of finding out!

<hr width=100% size=1>

 

[Mirelle]

The dish full of oil type artificial horizon cannot be used at sea, of course, whereas the Plath bubble version can (it's a toy for the perfectionist navigator who wants to do stars in the middle of the night!)

My copy of Lecky's Wrinkles describes the procedure for recovering GMT to correct your chronometer by observation from a known position ashore using a cast iron trough with a pool of mercury in it

<hr width=100% size=1>

 

[Gunfleet]

JohnM\'s Sextant Wrinkles

You could use a gps to give GMT. I'm told they're extremely accurate clocks ;-)

<hr width=100% size=1>

 

[Mirelle]

How absolute the knave is! ;-)

We must speak by the book, or equivocation will undo us!

I was talking about artificial horizons for using a sextant when ashore, not recommending the use of a century-old means of establishing GMT! ;-)

STS Lecky, were he alive today, would certainly be recommending that you navigate by GPS, but rate your chronometer off the GPS and keep a record of the chronometer rate, just in case you ever needed it.





<hr width=100% size=1>

 

[qsiv]

I hadn't truly appreciated the precision - I've just dug out my Old Man's sextant. The original NPL certificate inside the lid certifies it to 0.1, with zero error at all points except one wher there is a -0.1 correction. Like most sextants (I suspect) it hasnt been recertified - but I do recall being mighty impressed as a child at the accuracy of his landfalls, and recall well the ritual of sun and star sights, and listening for time signals whenever possible.

<hr width=100% size=1>