GPS accuracy

Accuracy of GPS ... it would be nice if not only for the GPS, but many other items were as true as the books tell you. Compare accuracy of GPS position to car fuel consumption as an example. My Volvo is supposed to average out at about 9.5 ltr / 100 km. I wish !! It actually works out at more like 11 ltr / 100 km.
My GPS plotter, old 8 ch Magellan Meridien and eTrex seem to agree very closely ... eTrex hanging under sprayhood, Magellan in its cockpit bulkhead bracket, Plotter antena hanging on curtain runner inside cabin. Average reported estimated error by the machines at about 30m. Sometime less .. occasionally more. I do have tree cover to contend with at mooring. Once I exit mooring reported accuracy increases to ~20m or so.
But I'm happy with that as when I need to be closer or more careful than that - it's proceed with caution using every available means possible.

Someone mentioned Decca ... I liked Decca, not only used it on ships but also had the early Nasa one on my boat. Decca has always had its 'knockers' - but was considered good enough by Europoort Pilots / Traffic Control to be used to guide super tankers into and out of Europoort / Maas Channel. The channel was aligned on a specific red lane and the pilot carried a box that plugged into ships decca unit. It gave him X track error on that lane. It worked excellently rain or shine, night or day.

GPS has developed into one of those items that has revolutionised our way of 'navigating' any vessel, car, plane ... even to fitted inside missiles to find that X. We all have the "don't forget the old methods" comments thrown around. Daresay similar comments hundreds of years ago when methods changed then from Lunar tables to time derived Chron / Sun - stars methods .. Evolution marches on. I'm from the old school ... still have all my sight work books .. brother pinched my sextant ... but am very happy to have that plastic brain ticking away telling me where I am ... /forums/images/graemlins/wink.gif
 
[ QUOTE ]
... That's not my GPS finding the same place on the same day, it's someone elses GPS fixing the spot weeks/months previously and me finding it again. Other folk had found it too, they had signed the book.

[/ QUOTE ]
The error could be consistently wrong in one spot giving a precise fix relative to the GPS system without giving the correct coordinates...
 
The accuracy of the systems are stated according to 1 or 2 standard deviations (SD). It is not an absolute figure but flexes /forums/images/graemlins/smile.gif

When GPS came out (Transit really) at 120 metres accuracy it appeared to be more accurate than Decca. Decca then immediately improved their accuracy overnight. They then quoted their accuracy on a 1 SD basis. Before the change it was on the 95 per cent (2 SD) setting so the accuracy was quoted at about 150 metres (IIRC) then they said it was 50 metres but based on about the 57 percent (1 SD) basis. Nothing had actually changed except the way it was measured.

Decca was great fun to use and could be improved on by using various plotting techniques when moving fast in confined waters. Knew exactly where the machine said you were, but everybody had conveniently forgotten that the basic systemn geometry had not changed so it was an illusion of greater accuracy rather than the real thing.

Transit showed us that a lot of the differences between morning and noon sights were not errors in our accuracy but the affect of tides in the deep ocean. We had always assumed they didn't have any affect that far from shore but it made a difference in the actual and assumed distance run. With the novelty of the new system we recorded the satellite derived position when ever it was obtained and the satellite DR position on the hour. No instant fix until GPS came about.
 
I would check two things:
Signal strength/no of satellites: to see if the antenna (etc) is degrading
Chart Datum: The unit may have got set to the wrong one by another 'user'
Otherwise, in open conditions I would expect an 'estimated accuracy' of better than 20m for a modern GPS. This is defined in various ways, but is usually along the lines of 90% probablility you are in a circle of that radius from the given position.
It may be worth leaving it switched on for a long while (>day) to allow it to recalibrate fully.
Other possibilities include interference either radiated or supply born.
I accept the points about 'we don't need that accuracy' but if it is not working well with plenty of margin in good conditions, it may fail completely when the going gets tough, due to extra loss from heavy rain for example. Or the problem may just get worse with age, if its something like corrosion in the antenna reducing performance, as happened with my GPS120.
 
[ QUOTE ]
The accuracy of the systems are stated according to 1 or 2 standard deviations (SD). It is not an absolute figure but flexes /forums/images/graemlins/smile.gif

When GPS came out (Transit really) at 120 metres accuracy it appeared to be more accurate than Decca. Decca then immediately improved their accuracy overnight. They then quoted their accuracy on a 1 SD basis. Before the change it was on the 95 per cent (2 SD) setting so the accuracy was quoted at about 150 metres (IIRC) then they said it was 50 metres but based on about the 57 percent (1 SD) basis. Nothing had actually changed except the way it was measured.

Decca was great fun to use and could be improved on by using various plotting techniques when moving fast in confined waters. Knew exactly where the machine said you were, but everybody had conveniently forgotten that the basic systemn geometry had not changed so it was an illusion of greater accuracy rather than the real thing.

Transit showed us that a lot of the differences between morning and noon sights were not errors in our accuracy but the affect of tides in the deep ocean. We had always assumed they didn't have any affect that far from shore but it made a difference in the actual and assumed distance run. With the novelty of the new system we recorded the satellite derived position when ever it was obtained and the satellite DR position on the hour. No instant fix until GPS came about.

[/ QUOTE ]

Dunno why you replied to me !! I was in MN through Omega, Loran A + C, Decca sets with flicking needles, and then lat Long displays, on ships trialing Magnavox Transit Sat Nav ... RDF ... and of course the tried and trusted Sextant.... as well as Syledis and various other Seismic systems etc.

GPS was a godsend really - it put paid to all the other cr*p that we wasted our time on trying to get sense out of.

As to Transit ? I hated it ! the bunching up of sats over 14 day periods so a rush of positions and then nothing for ages ... only real use of the thing was compass errors !! Did those a treat. In China Sea / Palawan Passage - it was only an aid that was often rubbish - we still had echo-sounder running and turn on a reef mark etc. - these were 85,000 cu.m Liquid Gas Carriers ... Transit - yeh - tell me about it !!
 
All this season I have had a chart plotter running from my elderly Garmin machine, and have plotted my track through the marina lock.

It clearly shows the track through the entrance (about 8 meters) on every trip. You can even see the different track leaving and entering.

Not differential or anything clever just the basic GPS. I think that is remarkable accuracy from something costing less than 200 squids.

Unless you had a very specialist application I can't see anyone would require anything better.

Like some others I started with compass and tidal diamonds and yes I too have look at the name on buoys - just to make sure. Would'nt like to have to go back to that now although I do occasionally turn the electrickery off- just for the hell of it, and to see if I can still do it manually.
 
[ QUOTE ]
All this season I have had a chart plotter running from my elderly Garmin machine, and have plotted my track through the marina lock.

It clearly shows the track through the entrance (about 8 meters) on every trip. You can even see the different track leaving and entering.

Not differential or anything clever just the basic GPS. I think that is remarkable accuracy from something costing less than 200 squids.

Unless you had a very specialist application I can't see anyone would require anything better.

Like some others I started with compass and tidal diamonds and yes I too have look at the name on buoys - just to make sure. Would'nt like to have to go back to that now although I do occasionally turn the electrickery off- just for the hell of it, and to see if I can still do it manually.

[/ QUOTE ]

My boat sits at back of the house on private mooring ... this on a wide river and gives me a good days sailing or mast down under bridges .. mast up and out to Baltic. Anyway - if I leave plotter ticking away at mooring (boat is secure and only moves 1 - 2m on the bow to the buoy, stern steady to pontoon. I get quite a jump around of the position plot. It becomes a track in fact.
But once I start moving into the river ... the track then seems to follow without the jumps of quite a few mtrs ... it actually steadies out.

On mooring ... (yes chart is my creation as no marine chart exists electronically for plotting .... )

ventaestate2.jpg


The purple / magenta lines are the tracks ...

ventangen.jpg


appventaest.jpg


And for those interested ... here is the chart I created for the river so I could work out my "home eta's" etc. >

venta-north.jpg


venta-mid.jpg


venta-south.jpg


Anyway - post is to show that GPS can take you into those uncharted regions ... /forums/images/graemlins/wink.gif /forums/images/graemlins/wink.gif /forums/images/graemlins/wink.gif /forums/images/graemlins/wink.gif /forums/images/graemlins/wink.gif
 
To be slighly pedantic....

Decca had very good repeatability.

But was not accurate in inshore / river environment!

My old Decca could put me back on my berth every time.

But if I plotted its coordinates, they were several hundred yards out.

But would that matter, repeatability is sometimes a good vice!

GPS is good for 95% of the time. In practice better than that.

It's all mans/womens magic for certain!!

PC to the end!!
 
I sail from Royal Quays - 2 miles up the River Tyne - if you'll forgive the expression.

Not quite sure I follow.

By repeatability rather than accuracy I understand you to mean if I stop at one specific spot a number of times then the machine always gives the same position but that position given may not be the true position.

ie The machine always gives the same (wrong) answer.

Is that it?

So if the machine always takes me to the same spot on the chart is it that the machine thinks the spot is in a different place to its true place. Or is it the chart that shows the spot to be where it actually - isn't?

My brain hurts now

I think
 
As I understand it, with Decca, the same machine would get you back to the same position to within something like 5-20metres, depending on where you were in relation to the ground stations. A different receiver would have slightly different errors, so would indicate a slightly different position. Perhaps 50m or more different.
BTW I believe that Loran is to be switched off soon, one off Obama's efficiency drives.
 
You must log in or register to reply here.

Other threads that may be of interest

N
Integral Battery Electric Outboards… with MORE POWER!
Replies
11
Views
744
ontheplane
O

Share this page

Top