Swinging a compass... with GPS?

bobdoughty

bobdoughty

Active Member
Joined
16 Nov 2007
Messages
49
Visit site
This may be a dumb question but I've tried and failed to work out the steering compass deviation by using transit lines and hand bearing compasses so why can't I make comparison between the GPS heading and the compass heading and use that figure to produce a rough deviation card? I know it doesn't take account of variation but this is only about one degree and falling where I am. This seems too simple - could someone explain the fault in my logic. Thanks
 
Your GPS will not give your heading, only your track. Your compass shows the way the boat is pointing while the GPS shows the way it has been moving - these may not be the same.

Alan
 
Jack
What part of the world are you in? 1 deg variation and getting less ~ down here in the S.E we have 5+ degs.

Your basic logic isn't flawed and you should be able to come up with a reasonable GPS based correction table for the deviation on the compass, but I would have to ask you why you need it so much?

Most yacht compass's are marked in 10 deg segments (the needle is probably 3 degs wide on it's own). Most helms can't steer better than a 10 deg course. Then you have leeway/wind to take into account.
In the (distant) past when I, like loads of other people on here, navigated by pencil and paper, log readings and depth soundings rather than GPS chart plotters with interlinks to autohelms ~ we were more than happy to fetch-up with +/- 10nm of our aiming point after a 60 nm passage.

Hands up all those true sailors that have made a landfall and 5 nm off have made a major course correction to actually hit their waypoint!!!!!

Peter.
 
Probably a dumb answer. Find a skipper whose boat has an accurate compass. Bribe him with bottle of wine/beer. Ask him to sail/motor the points of the compass while you follow. Communicate on the radio - ie. as soon as you have noted each deviation he changes course for the next etc. etc.
 
[ QUOTE ]
Find a skipper whose boat has an accurate compass. [ QUOTE ]


Put crew in pulpit with h-b compass clear of all metal objects (not hanging onto forestay and drinking can of beer!) then steer round points of the compass with crew sighting back along ship's centre line and noting reciprocal bearings. That's all the info you need to compile a deviation card.

But if worried about deviation it's well worth paying a pro to adjust/compensate your compass. Bad news is that Robertson in Southampton now charges over £100.
 
I diasgree with your description and logic.

Most boat's compasses are marked in 5 degree intervals in my experience. In the past I never asked anyone to steer to anything in between as its a bit of a waste of time, however, I have heard the argument (from Tom Cunliffe himself) that its possible to steer to within one or two degrees, but I remain sceptical and usually still give out courses to steer to the nearest 5 degrees. Please note that I still plot to the nearest degree when I add or subtract deviation and variation, leeway estimated helmsman's error etc.

I do claim to have done plenty of navigation without electronic fixing aids such as Decca or GPS. When I was doing it regularly I would expect to be within much less than 10 miles of my EP having crossed the channel. Taking careful consideration of the tidal stream and leeway, I seem to recall being within a couple of miles or less on most occasions. However, navigating in this traditional way certainly reminds you to 'aim off' and aim to be upwind of time and tide as you make you approach. I believe that course accuracy is never as good as log accuracy, but errors of less than a couple of miles over every sixty are fairly easy to achieve with careful plotting and good understanding of the tidal streams etc.

If the original poster wants to produce a compass deviation card, then I suggest that one of the easiest ways is to find a quiet patch of water with a conspic object several miles away and circle slowly watching the bearing of the object on a hand bearing compass. If the bearing stays the same from within your (small) area as you circle then you have shown that the position in the boat that you are holding the handbearing compass in is free from deviation. Obviously, if the bearing changes choose somewhere else and try again but it's normally possible to find a place (often up by the backstay) which is deviation clear for the handbearing compass.

All you then need to do is to take bearings of the fore-and aft line of the boat from the same position that you were in for the first part of the exercise and compare the bearings with the main compass to draw up a deviation chart for the main compass.
 
[ QUOTE ]


Most boat's compasses are marked in 5 degree intervals in my experience. In the past I never asked anyone to steer to anything in between as its a bit of a waste of time, however, I have heard the argument (from Tom Cunliffe himself) that its possible to steer to within one or two degrees, but I remain sceptical and usually still give out courses to steer to the nearest 5 degrees. Please note that I still plot to the nearest degree when I add or subtract deviation and variation, leeway estimated helmsman's error etc.



[/ QUOTE ]

At last some common sense.

Its sensible to have a deviation card but the likelyhood of you having to use a compass (as opposed to chosing to use one) on a well equipped modern boat are very small indeed, and if push comes to shove you could use a deviation free hand bearing compass anyway. So I wouldnt (and dont) worry about not having a deviation card.

The easiest way to derive one is to forget transits, have a second person on board and then simply steer the boat on autohelm in different directions. Best steer by autohelm since its more steady, then one of you shouts out the main compass bearing and the other checks the hand compass and notes down the readings of both. Change course by 20 degrees and repeat. Before doing all this make sure there is no ironmongery near the main compass, and after doing it turn to the position with the worst deviation then either switch the engine on if its been off or vice versa. Note if this chganges the deviation significantly and if it does repeat the whole process withg the new engine state.
 
I would have thought an easy way to check the compass and make a deviation card/ table is to put a local bouy into the GPS as a waypoint.

The GPS will give you the bearing to it. (The bearing will either include deviation or not depending how you set the GPS up)

If you stay a mile away from the bouy, point the boat towards it visually, you will get a reliable bearing to it and you can check / note the compass bearing. This can then be repeated on as many points of the compass as you wish.

You might just steam past the bouy to check its position first or to log its position.

Voila!
 
[ QUOTE ]
I would have thought an easy way to check the compass and make a deviation card/ table is to put a local bouy into the GPS as a waypoint.

The GPS will give you the bearing to it. (The bearing will either include deviation or not depending how you set the GPS up)

If you stay a mile away from the bouy, point the boat towards it visually, you will get a reliable bearing to it and you can check / note the compass bearing. This can then be repeated on as many points of the compass as you wish.

You might just steam past the bouy to check its position first or to log its position.

Voila!

[/ QUOTE ]

I like the sound of this method. My plotter does show a heading but I think this would be more accurate. I've tried the hand bearing compass method but I'm usually single handed and it's nigh on impossible to steer a steady course and get in a position where you can sight the compass and see the steering compass. I completely agree about the accuracy of steering, I just want to be aware of any gross errors. My reasoning for doing this is that I much prefer to steer by a compass rather than the GPS
 
[ QUOTE ]


I like the sound of this method. My plotter does show a heading but I think this would be more accurate. I've tried the hand bearing compass method but I'm usually single handed and it's nigh on impossible to steer a steady course and get in a position where you can sight the compass and see the steering compass. I completely agree about the accuracy of steering, I just want to be aware of any gross errors. My reasoning for doing this is that I much prefer to steer by a compass rather than the GPS

[/ QUOTE ]

I'm not sure how good are the tables of variation buil;t into GPS sets - do they take into account local variations as shown in the charts or not? So you are likely this way to end up with a mix of deviation and variation on your chart.

Even single handed I dont dee the problem with getting the autohelm to steer, which they do much better than you or I, and then simply standing there and comparing hand bearing compass with ships compass. Press the +20 buttons on the autohelm and repaet.

Chose a reasonably calm day

If you did try the buoy approach you would have to circle the buoy at 1 mile radius. Often wont work with a channel marker.
 
We did an 'initial' swing of our compass when we first acquired our latest boat by using the dGPS, (one with differential built in, pre EGNOS) and that initial card is still the one in use 8 years on! We did ours in Studland bay at slack water neaps on a flat calm day, motoring with pilot steering at a steady speed of about 4 to 5 kts and comparing compass heading to GPS heading. Our GPS was set to magnetic, but the automatic variation figure was checked first and agreed with local charts. We calibrated our log at the same time by comparison with GPS SOG. We then swung the fluxgate compass on the pilot electronically as per manual and compared this also to the GPS swing, with a degree on all headings so more confidence gained.

It does work, but any tide will screw it completely!

For the record, we actually have a proper Pelorus, to which we have added a shadow pin to allow measurement of sun's bearings and this by comparison with accurate time (from GPS in our case) and a mini computer with astro programming can give really accurate results. We did our last boat compass with this over several weeks when we collected the data as we found opportunities to be on a stable platform on many different headings. A bit nit picky geekish but the Pelorus cost just $30 from West Marine USA and I already had the Astro Nav progam set up on a Casio programmable calculator.
 
Don't believe that the handbearing compass will always be deviation-free. It is affected by any fields just like the ships compass.
A distant charted object, like say the Nab tower from Hayling bay, let the boat slowly turn in the wind, as the distant object comes into line with sighting lines, eg straight across shrouds, fore and aft, shout 'now' and blonde assistant reads compass.
Then repeat with second distant object about 45deg different. That gives you 8 points.
Or set up more sighting lines with pelorus, 30/60 set square etc.
Ideally get data at 15deg increments and plot a graph then check out any anomalies.
Use GPS to give true bearing to objects.
 
You must log in or register to reply here.

Share this page

Top