GPS Inaccuracy.

  • Thread starter Thread starter bob_tyler
  • Start date Start date
B

bob_tyler

Guest
Yesterday, there we were under sail going up the River Fal looking forward to a good lunch at the Sumglers' Cottage when I gently ran aground on a mud spit, only the first time accidently in many years sailing. I knew I was cutting a corner inside a green buoy marking the deep water channel which is used by big ships. According to my Plotter on large scale I was in water with a minimum depth of 3.3mtrs and I only draw 1.3 mtrs.

After getting afloat again I continued up river keeping well to starboard in the deep water channel. According to the plotter, as I passed up the river near the King Harry Ferry I was well over to the port side!

Later in the day I was passing the red and green markers to the Mylor Channel entrance. According to the plotter they should have been about 50 mtrs to port but they were, in fact 50mtrs or so to starboard.

In all the above events the plotter was constantly showing about 100mtrs out!

Any GPS experts have an explanation please! Is it my plotter or what? Reception was good with the plotter using either 7 or 8 satellites for the fix and a low HDOP (whatever that is) of around 1.01 to .89.

I had previously noticed that the Black Rock marker and other nav points appeared slightly out of place but put that down to a good night out the night before and didn't bother about it. /forums/images/graemlins/confused.gif /forums/images/graemlins/confused.gif
 
GPS has built in inaccuracy of up to 100metres as I understand it the US and other military have more accuracy.

Up by the king harry ferry the trees and sides of the estuary are steep - this also doesn't help.

Might be completely wrong though

Scopas
 
the deliberate random offset has been discontinued by the yanks for a few years now, so a gps has an accuracy of about 5 m

with a differential correction applied its a few cms

second, have you set up your readings for wgs 84, all admiralty charts around the uk use this now, (but check the wgs84 offset which might have to be applied to the chart) then make sure you know if your readings are in degrees, mins and seconds, or degrees mins and decmals of a minute ....... otherwise this also can cause errors - especially on large scale charts

finally, keep a lookout for 3d or 2d signal reception, 3d is best for rockdodging

loads of info on google - wgs 84 etc ....... /forums/images/graemlins/cool.gif
 
Why use a GPS for navigation when you were obviously on pilotage? No paper charts or pilot book?

How would you have managed in the dark days before GPS, when 100mteres you have been very acceptable accuracy?

I'm not having a pop, I just don't see the point and it does sound likethis proves the reliability of Mk1 eyball.

Okay, so I'm a luddite!
 
Sometimes visibility is not that good, and at these times it is good to know how accurate your GPS and navigation system is.

Quite by chance I discovered (in the fog) that one of my charts was not WGS84 - nearly aground I was 200m from where I thought I was.

Quite right to question and work out why it is not giving you the right answer.

I agree though, when the visibility is good, use eyeball to keep you in practice, although I am not sure you say how you were navigating in your post?

Pre GPS ships and boats went aground and many were lost due to lack of good fixes. GPS is a good thing.
 
Is it true that 3-D is more accurate than 2-D mode as stated by Landaftaf earlier in this thread? I thought that 2-D was used for traversing the surface of the earth such as us boaters do and 3-D was used in aircraft where elevation has also to be considered. Now I'm confused.

Any other knowledgable opinions out there?
 
Duncan, I said I was sailing but perhaps drifting lazily in the sunshine at about 1 kt would be more accurate and I had knowingly and deliberately cut the corner to save gybing! There wasn't really a problem as the tide had turned and was on the make and we were in no hurry..

My navigation was solely by eyeball with a coffee in hand - no chart in use just lazing. I only turned the plotter on once we were aground to see what it showed. That's when I noticed the apparent discrepancy so watched it with interest for the rest of the day. Yes, the plotter is on WGS84.
 
Pelican Pete, I think my plotter is only 2D as I don't take my boat up mountains and it is not accurate enough to measure wave height.
 
Your plotter is on WGS84 but is your GPS? They both need to be singing from the same hymn sheet which doesn't have to be WGS84 but that is the easiest option unless you are taking data from the GPS to plot on a paper chart as well that is in say OS36 datum. 3D v 2D is usually selectable as an option from the set-up menu for the GPS. Which plotter, GPS and chart system are you on?
 
the difference between 3d and 2d is all to do with the quality of the satallite fixes, no good having 6 sats all on the same approx brg /forums/images/graemlins/grin.gif

4 good 'crosses' with a good signal sterngth will give you a 3d fix ........ 3 is dodgy and less than 3 is poor

if you are getting only 2 d fixes you ought check the antenna and connections, cos you ought get 3d with a decent piece of kit

and no - nothing to do with mountain climbing /forums/images/graemlins/cool.gif
 
[ QUOTE ]
Duncan, I said I was sailing but perhaps drifting lazily in the sunshine at about 1 kt would be more accurate

[/ QUOTE ]

and while you were doing that I was stuck in a baking hot office!

Envious, me? You bet.

I do take the point that you took the oppurtunity to check the GPS against reality and believed the reality. I have heard of people believing the GPS when reality is obviously different!

There again, I don't really understand or trust these electronic gizzmos anyway.
 
The GPS signal will place you on the surface of its mathematical model of the earth to within a few meters - in good reception conditions. The plotter part then has to overlay that position with the chart detail - but that detail also has its own inaccuracies. Then there is the issue of time lag - the GPS give you a position a bit behind where you now are, with a bit of help from clever calculations of which way you were going etc.

All these inaccuracies can add up. So the GPS bit might have said where you were to within a few meters but the river wasnt where the plotter said.
 
[ QUOTE ]
Then there is the issue of time lag - the GPS give you a position a bit behind where you now are, with a bit of help from clever calculations of which way you were going etc.

[/ QUOTE ]

I'm afraid that that has no bearing on the situation. When aground, for example, I did not move for about 15 min.

When going up river my course was on the east side of the river, not on the west side as plotted. So much so that I suspect if I had gone over to the other side of the river the plot would have shown my course to be overland. I do not have an amphibious craft.

The two nav. buoys were on the opposite side from that indicated, the plotted position being about 100 mtrs adrift!

Nobody, as yet, seems to have come up with a satisfactory reason for these errors. Could there be a fautl in the maths calculations in the plotter? Could the C-chart have been inaccurately converted from an Admiralty Chart or was the survey on which the Falmouth chart is based incorrect?
 
... what charting system does you plotter use? One commonly used digital chart package has put me "ashore" at the same time that UKHO based raster charts had me in the right place...

Non-differential GPS accuracy will be better than 10m for about 90% of the time but you can get excursions out to around 15m. Assuming you don't have antenna obstruction issues you should be getting that sort of accuracy, perhaps with 5m accuracy at least 50% of the time.
 
the time lag is a red herring ..... gps updates itself about every second, in fact a little less than every second


as I have alluded too - if your chart was an old edition you must check for offset corrections for wsg84 and apply them .... etc etc

from my experience with gps plotters, if set up properly and interfaced with the chart correctly they are extremly good around europe, I use them with an ecdis (electronic chart display) around the uk and they are accurate to a few metres

if I were you I would get out the instructions again /forums/images/graemlins/cool.gif
 
[ QUOTE ]
if your chart was an old edition you must check for offset corrections for wsg84 and apply them .... etc etc

from my experience with gps plotters, if set up properly and interfaced with the chart correctly they are extremly good around europe...

[/ QUOTE ]

The chart and plotter are both set to WSG84 and my Imray Chart is the same. (I did not take bearings etc and try them on my Imray chart as I wanted to get to lunch. That may have been an interesting excercise)
 
Yep I've had similar accuracy in the Fal, but my gps was working only 3 satelites and not regularly on each update, poor signal from some I assumed due to the steep valley and trees as others have said. They will give a position with 3 but it gets better the more they pick up.

I have never found gps reliably giving better than 50m accuracy where there may be obstructions. When you have horizon to horizon views they should be better.
 
In my car I have had the GPS showing me 2 miles west of my actual position. The garage took the simple step of replacing the sat-nav under warranty. In summary, I used the sat-nav to get me where I wanted to go - all fine. Got back in the motor to go home, the sat-nav couldn't find a route. Puzzled, changed to map mode, found that sat-nav thought I was in the middle of a moor 2 miles west. Fiddled with connections to no avail. Drove home on 'manual'. Problem not resolved until garage replaced sat-nav.

I would start by checking the connections between the antenna and receiver, then reset the receiver, then try another receiver with your antenna.

FWIW

Steve
 
You must log in or register to reply here.

Other threads that may be of interest

T
Notices to Mariners - Week 50
Replies
2
Views
680
tillergirl
T

Share this page

Top