Welcome to the NavList Message Boards.


A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Position-Finding

Compose Your Message

Add Images & Files
    Re: What precision is required in cel nav?
    From: Bob Goethe
    Date: 2015 Jul 18, 17:20 -0700

    >>If others have actually tried celestial while offshore on a boat of, say, 30~50 ft LOA, I'd love to hear their opinions.<<

    Last summer I was navigator on a Dufour 45e that sailed from Hawaii to Victoria, BC, Canada.  The boat had been outfitted to sail in the Vic-Maui race, and so had a navigation station that was a delight to use:

    • GPS
    • AIS (https://en.wikipedia.org/wiki/Automatic_Identification_System)
    • Radar
    • High Frequency Radio
    • a Windows laptop PC running a software package called Expedition.

      The computer software interfaced with the HF radio and used SailMail to download GRIB files (http://weather.mailasail.com/Franks-Weather/Grib-Files-Explained)which.  The software could then integrate our GPS fix with the GRIB files, together with what it knew of the boat performance characteristics on various points of sail, as entered into the configuration section of the software, and produce an animated graphic of winds/waves along our route as they were forcast to change on an hourly basis for the next 36 hours.  Out of this, Expedition suggested an optimal course for the boat to sail to make the most of the forcast conditions as they unfolded over these 36 hours.

    My goal for this trip, in addition to just generally having fun, improving my seamanship, and learning to use this navigation station, was to learn what I could about using a sextant at sea.  I have fixed the location of my living room some hundreds of times over the last 5 years using a cereal bowl filled with water as an artificial horizon.  And though I had taken my sextant along on coastal cruises, this was my first time to use it on an extended trip out of the sight of land.

    My first sextant shot gave me a line of position (LOP) that was 92 miles away from where my GPS said I was.  My next effort was 36 miles off.  I pondered this for a while and inferred that I must be taking my shots from the trough of a wave, and having a horizon that was a wave crest that was potentially only 100 or 150 yards away.  I was looking *up* at a wave crest.  Based on my prior experience of doing living room fixes, I had not foreseen the importance of this consideration.

    I could see already that GPS was providing me a way to know that I had screwed up, and needed to refine my technique.

         Sextant navigators in the 21st century can know whether they have taken a good sight
         or not, and determine their accuracy to a degree that was beyond imagining a century 
         ago. I am betting that the most skilled sextant users (though possibly still not the best 
         navigators) in history are alive right now...and probably are members of this group. You can 
         dial in your sextant shots to a fare-thee-well today.

    By the middle of thge trip, I got to the place that if the waves were coming from ahead, astern or abeam-on-the-same-side-as-my-celestial-object, such that I could SEE the wave coming, I was able to take a sight that was invariably accurate plus or minus 3 miles, and 30% of the time was accurate plus or minus 1 mile.  The trick was taking your shot on the crest of the wave.  Doing so normally gave me 2 or 3 seconds to do the final dialing in of my sextant.
         I say "invariably accurate to such-and-such," but this was true only when I had a 
         clear, non-hazy view of the horizon.  So "invariable" is not actually the right word.

    The trick for me was to get the altitude as good as I could get it, and then - while holding the sextant in position before my face - look at the waves and wait.  If I had to wait for more than a few seconds, then it was back to the sextant to readjust it for the movement of my object in the sky.

    I have a Tamaya "Venus" sextant, which is a 7/8ths sized sextant designed for yachtsmen, and is pretty light for a metal sextant.  Still, by the time I could get four sights taken, my arms were often trembling.

    If the waves were coming from the side of the vessel that was opposite to where my celestial object was - this is to say, my back was to the waves as I sat amidship looking out - then my LOPs would regularly be out by 6 to 15 miles.  I could never get any sort of knack for *feeling* the rise of the boat.  If I couldn't see the crest coming towards me, then I couldn't get a decent shot.  I would look over my shoulder at the waves, but doing so would throw off my sextant positioning.  It took me too long to reorganize myself and FIND my object in the view finder.

    I always took my shots in sets of four, and then did a graphical sight-evaluation of those four shots compared to the movement (slope) of the sun during the same four minute period when I was taking my shots.  This technique I learned from David Burch, "How to Use Plastic Sextants", from Starpath.com.  I got an electronic copy of this book when I purchased a Davis Mark 3 plastic sextant from Starpath.  The book is helpful enough in this one respect that even if you don't need a Davis Mark 3, it is worth buying one just to get to read Burch's section called "How to Average Celestial Sights" on page 36. 

    Not infrequenly I found that no two of my four shots lay along the precise slope of the sun on my graph paper, so I picked a hypothetical sight (i.e. I took one of my existing sights and added or subtracted X.X minutes to it to come up with what I thought my observation *should* have been had I taken an accurate sight).  These hypothetical sights were often among my within-one-nautical-mile-of-GPS shots.  I came away quite enthused about slope-averaging.

    Coming back to waves, I had several guys on the boat who were intrigued by the sextant and were interested in working with it themselves.  I taught them to the extent that seasickness allowed, and we helped each other as assistants to time our sights relative to the waves.  I found that if the waves/swell were approaching my vessel from the opposite side as where I was pointing my sextant, that I could get good results by having my assistant watch the waves FOR me, and give me a countdown.  "5 - 4 - 3 - 2 - 1- CREST!"  Because of the unpredictable results of waves plus swell, the countdown was often more like "5 - 4 - 3.......3 - 2 - 1 - CREST!  No, wait.  NOW!"

    So, using David Burch's sight averaging techniques plus giving an enormous amount of attention to when we got up onto the crest of a wave, I was able to regularly get 1-3 nm precision.  I had two shots where my LOP was within the thickness of my pencil lead of my GPS fix, but I knew that had to be random good luck.  I used Pub. 229 for sight reduction.


    p.s. The skipper of the boat did not use a sextant often, but when he did was very enthused with noon-meridian sights.  Those are fine enough, but left me spending too much time staring through a sextant at a bouncy horizon.  The longer you spend doing this, the more - for me - feeling slightly seasick came into play.  I concluded that the most direct route to a St. Hilarie LOP was the healthiest way for me to do celestial.  Whether the resulting LOP happens to be precisely a parallel of latitude or is at a slightly different angle is of no consequence in getting an LOP you can cross with another in a running fix.

    So for me, any discussion about the relative merits of noon sights boils down to the best way to avoid feeling queasy.

    Browse Files

    Drop Files


    What is NavList?

    Join NavList

    (please, no nicknames or handles)
    Do you want to receive all group messages by email?
    Yes No

    You can also join by posting. Your first on-topic post automatically makes you a member.

    Posting Code

    Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.

    Email Settings

    Posting Code:

    Custom Index

    Start date: (yyyymm dd)
    End date: (yyyymm dd)

    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site