AusENC S63 large scale cells offset

[aboon]

Greetings all.
I have been using the Australian S63 AusENC charts with SeaPro Lite+ for several years and have just started with OpenCPN 4.8.4. I have a licence for the Coastal Pack which includes most of the charts (cells) for my home State of Tasmania and a few for the trip north (67 cells in all).
Everything is working as expected except that the larger scale cells are offset to the west, by about 0.024' (at latitude 43 S). While this is not a great distance, it does result in discontinuities at the eastern and western edges of the cell and means that the locations returned by an Object Query of lights, marks, etc are incorrect.
Has anyone experienced this effect and are you able to suggest a cause/remedy?

Andrew

[bcn]

Aboon,

which cell exactly? When you hoover over the chart bar, what does it say?
S-63 charts are presented exactly as they come, no offset applied.

Can you please show a screenshot where you think the chart is wrong?

Object queries do not depend from the presentation of the chart.
Chart objects carry their own parameters and those are getting presented.
Each chart object is placed on the screen depending on it's proper encoded position.

Different from raster charts - there the whole thing goes wrong when referencing or datum is bad.

Hubert

[aboon]

Hubert,
Thanks for your prompt reply.
Attached are 5 screenshots. The first (Pt Arthur offset OCPN) shows part of the western boundary of cell AU5174P3. I have added a short N-S 'route' along longitude 147 48.95' E, which is where the boundary should be (according to the cell information on the AusENC website).
The second image (Pt Arthur offset SeaPro) shows the same area on the SeaPro Lite+ display. The third image (Pt Arthur offset OCPN less plan) shows the smaller scale cell over which the P3 cell is laid.
Two images (Pt Arthur beacon object query and Pt Arthur query Inquire) show the data for a beacon within this cell from OCPN and SeaPro, respectively. As you can see, the location of the beacon differs by about 0.015' of longitude.
I have checked one other large scale cell (AU5174P5) and found an offset of 0.024' in longitude.

Andrew

[bcn]

Hmm,

the attachments didn't make it into the forum.
You might need to ask a moderator being a new member

Use the warning triangle down left at each comment ...

[aboon]

Sorry about that. Trying again.

Andrew

[bcn]

Andrew,

looking at a S-63 set of the region (dated three years aprox.) you are reporting from and the charts involved (AU5174P3 and AU444147) I do observe that

- objects are in their correct place (the OCPN marks and the lights or buoys are spot on)


- that the shorelines of AU5174P3 and AU444147 show a slight jump at their intersection. Which one is the correct one?

The only way would go to that place with some good GPS, surveying and then comparing. Same with some more points on the shoreline of AU5174P3 if you would have the curiosity to check at the spot.

So as far as I can see here, no issue of geo-referencing. Perhaps this has been corrected with a more recent edition of the charts.

Hubert

[aboon]

Hubert,
The objects you have shown from the larger scale Plans are all offset from their actual position. If you alter the waypoint lat/lon in Attachment 169478 to -43 9.291/147 51.911, which I believe is the actual location as shown on SeaPro, you will see the offset. All of AU5174P3 needs to be shifted west by approximately 0.015'.
The locations of objects in the smaller scale AU444147 chart agree, so it's only the larger scale plans that show the shift.
I found a Notice to Mariners which quoted the official location of a mark in AU5174P4:

"1110* AUSTRALIA - TASMANIA - Port Huon - Topmark.
MAST M121/15

Aus174 (plan: Port Huon) [738/2015]
Delete topmark 43° 09.740 S 146° 58.820 E"

The Object Query for this mark in OCPN returns
"Buoy, lateral (BOYLAT)
43 09.7409 S 146 58.8088 E"
The Inquire data on SeaPro is "Mark Position 146 58.8200E, 43 9.7400S".
So I am believing that the locations shown in SeaPro are correct. Note that the westerly shift in AU5174P4 is approximately 0.011' of longitude, slightly less than AU5174P3.
If the Plans were shifted to the west, the discontinuities at the boundaries would disappear.

Andrew

[bdbcat]

Andrew...

Trust me on this:
The object positions reported by Query on OpenCPN are not at all affected by any possible offset error in the display. The values come directly from the S63/S57 ENC cell data.

So, I think it more likely that OCPN is not picking up some updates to the chart in question, for some reason.

You may try to uninstall the OCPN S63 charts, and then re-install. The logfile will have information on the updates applied.

When I get a chance, I will take a look at the actual charts as well.

Dave

[bdbcat]

Andrew...

Considering only 5174P4:

I did some tests. In my chartset, there are 3 updates available for this cell.
If the updates are not applied, the location of the buoy you mention is incorrect. However, when the 3 updates are applied, then the buoy is reported to be at the "correct" position, namely 146 58.8200E, 43 9.7400S.

So, it seems that your chartset does not contain the updates, or OCPN is not applying the updates correctly.

Please check your charts and the logfile, looking at this particular cell as a reproducible test case. If we sort this one out, I expect everything else will fall into line.

Thanks
Dave

[aboon]

Thanks Dave.
I'm lost in a sea of data here. Can you tell me what I should look for?
Is it worth sending you the log file? Today, I have removed the OpenCPN program and all data then re-installed program, S63 plugin and chart data. I still have the same result (offset) but the current log file has the setting up of the sENC files.

In the AusENC Base Map\ENC_ROOT\AU\AU5174P4\2 folder there are 5 folders ('0' to '4'). I assume this is Edition 2 of this Plan with 4 updates. The 'Date modified' against the AU5174P4.000 file (in the '0' folder, 70 kB) is 17/07/2015. The date against AU5174P4.004 ('4' folder, 2 kB) is 13/01/2017.
The 'Updated:' date for AU5174P4.os63 reported on the bottom line of the OCPN screen is 2017-01-05.
There are no AU5174 folders in the Cumulative Update file for AusENC.

I don't know whether this is relevant. Please let me know what I can do to help track this down.

Andrew

[clownfishsydney]

Mine has the same error for the coastline.

[bcn]

Andrew,

just for getting everybody with same information:

- you are using the actual enc_base set from week 2-2018 and cumulative update from week 18-2018?

As you were stating, there are no updates for AU5147P3 in the updates and for the smaller scale AU444147 yes there is one.

Did you install be base set from scratch or updating a former already installed set?

We expect to have the latest AU charts here in a few weeks what will make it easier to analyze with the genuine base data. But in the meantime looking into your observations.
Btw.: any track going into Port Arthur to compare real positions?

Hubert

[aboon]

Hubert,
Yes, I confirm that I am using the base map from Wk 2 2018 and the cumulative update for Wk 18 2018. I have re-loaded the charts several times. This morning I uninstalled OpenCPN and all loaded plugins and started afresh, re-loading all permits, charts and updates. Same result as before.
Unfortunately I have no tracks in OpenCPN yet (I'm still using SeaPro on the boat and haven't been out this month yet). I may be sailing through a couple of the large scale plan areas this weekend (if the weather improves!) and will record my tracks.
I have noticed another issue with the 'tick-marks' on boundary lines. In the Symbolized view, some tick-marks point into the area bounded (no-anchoring, marine farms, etc) while others point away from the area. My understanding is that all should point into the area. One example is attached.

I found a Task (ID 1909) which could describe this but it referred to the Draw plugin. Should I start a new thread?

Regards,

Andrew

[bcn]

Andrew,

to you have perhaps the log when installing the charts?
Anything that might give us a hint there?

And no attachment for the tick marks...

Hubert

[aboon]

Hubert,
Here is the log file from today. In two parts to keep under 400 kB. I've also send again the boundary tick-mark image. Fingers crossed!
I loaded NZ raster files, CM93, a Test S57 cell for Australia and the AusENC files. I set up 4 Chart Groups, one for each of these.

Andrew