Help / FAQ

[svadagio]

Hi
I am using Opencpn and a GPS puck to navigate. I have a small but persistent error in the position. It is always the same. I was wondering if someone has a suggestion or is there a place in Opencpn where I can put in a factor to correct this. Opencpn has worked fine for over 2800 miles. I would just like to eliminate the error. As I said it is a small error.
Bob

[bcn]

- How small (meters, cables,...) is the "small error"?
- Consistent, i.e. like an offset or random noise?

In case of an offset you might play with the ship parameters in the settings.

Hubert

[PjotrC]

Recording the incoming NMEA and showing the corresponding OCPN position would probably help.

[CarCode]

What kind of charts you are using?
If cm93v2 you can correct the offset.

Gerhard

[svadagio]

The error is about 125 meters at 200 degrees. It is consistent not random. That is the ship shows to be south west of where it really is. Can you give me a hint where the ships parameters are. That sounds like the answer. I am using charts made during the 2700 nga progect and charts bought commercal. The error is the same on all charts.
Bob

[nohal]

Bob...
Are you sure (as in really sure, confirmed against other GPS that is correct or a geodetic stone on land) that it's the position that's off, not the charts?
As it's very unlikely, that the GPS will be *consistently* off, and in fact it could kill someone in most of the cases, there is no simple mechanism for such correction.
With CM93, you can correct the individual cells. In KAP charts, you can modify the header.

Pavel

[bcn]

Originally Posted by svadagio

The error is about 125 meters at 200 degrees. It is consistent not random. That is the ship shows to be south west of where it really is. Can you give me a hint where the ships parameters are. That sounds like the answer. I am using charts made during the 2700 nga progect and charts bought commercal. The error is the same on all charts.
Bob

That sounds like an offset in the ship data you are not aware of

Settings (wrench symbol) --> ships --> own ships tab and there the GPS Offset

The other stuff to have a look at Pavel has commented already about.

Hubert

[cagney]

Originally Posted by svadagio

The error is about 125 meters at 200 degrees. It is consistent not random. That is the ship shows to be south west of where it really is. Can you give me a hint where the ships parameters are. That sounds like the answer. I am using charts made during the 2700 nga progect and charts bought commercal. The error is the same on all charts.
Bob

For an example of what unknown geodetic datum can do to a NGA chart, have a look at Chart Info | Official OpenCPN Homepage, at the bottom of the page.
What is the Geodetic Datum for your charts?
The NGA chart and the commercial chart are probably based on the same survey.

Thomas

[svadagio]

I am sorry but I can not find the gps offset. What version are you on?

I place the point of the curser on the lat long lines and get the right number to the third decimal place. As with most charts the feathers on land are off but I allow for that.

Bob

[bcn]

Originally Posted by svadagio

I am sorry but I can not find the gps offset. What version are you on?

I place the point of the curser on the lat long lines and get the right number to the third decimal place. As with most charts the feathers on land are off but I allow for that.

Bob

3.2.2 - the actual one

The offsets are to define where the GPS antena is placed with reference to your ship.
Measurements are placed at the GPS's antennas place - so if your ship is slightly larger - say 300x25m for example - it is interesting to reference this to the ship. In our cases, 12m, 40 ft, perhaps not - but as the precision of GPS today is about two meters it may already be worthwhile. The chart for our harbor is 25m offsetted for example.


Hubert

[svadagio]

Thanks I found the offsets. I have 0 in them so that is not the problem. I may try putting something in them to correct the error.
Bob

[bcn]

Might help, but we have not found the reason then.

Please consider the advices of Pavel, Thoma and Pjotr.

Hubert

[svadagio]

Well after further thought it might not work. If the offsets turn with the ship then it will not work. It seems they have to to be accurate.

[svadagio]

Here is something new. I got out my backup puck and with it there is no error or very small. I will have to test it some more to be sure. Never heard of a gps being off! Any ideas?

[cagney]

Originally Posted by svadagio

Here is something new. I got out my backup puck and with it there is no error or very small. I will have to test it some more to be sure. Never heard of a gps being off! Any ideas?

Back to the Geodetic Datum. On some GPS units this can be set, while on others it's fixed. The popular gps puck BU-353, for example, is locked to WGS84.
What brand is your first puck?
A different datum than WGS84 can very well explain what you see.

Thomas