OziExplorer and Map Calibration

Code Developer said:

A couple of weeks back I bought Ozi Explorer

Good choice.

Quoted message said:

The map section I used was SD 580540 to SD 670450 and I used these two
points to calibrate the map corners as 53deg 58.822', 2deg 38.515'W
and 53deg 54.008', 2deg 30.222'W. (2 point calibration). I used a free
conversion utility called NGR84.EXE by Specialist Systems to convert
the OS map co-ordinates to the Lat/Long co-ordinates required by Ozi.

Good grief! You are making things difficult for yourself aren't you?
Just set Ozi to use OS coordinates!

In the configuration (alt-C), on the Maps tab, set the Alternate Grid to
BNG, and blank map datum to Ord Srvy Grt Britn.

When calibrating a map, set Map Datum to Ord Srvy Grt Britn and map
projection to BNG. Then simply enter the OS grid coordinates. Best to
place the calibration points on grid line intersections.

Also, in the configuration, on the System tab, set the Data File Datum
to WGS84. Also, on the GPS tab, set the GPS upload/download datum to
WGS84.

This may sound confusing, but remember this, the GPS *communicates*
using WGS84, but you want to *display* the information in OS BNG.

Quoted message said:

Whilst I was there, the GPS almost showed the correct position, but
for the duration of the walk constantly showed me (on the map) as
being around 50 metres south of my actual position.

A common problem, you probably just got the settings wrong. Set them as
above and all should be okay. Scan the map area around your house then
go for a quick stroll to check that it's correct.

An easy way to check that your map calibration is correct is this.
Click on Map - Grid Line Set Up, then click on the Other Grid tab and
tick the Grid On box. This will overlay a grid over the map. It should
match perfectly. If it doesn't, you've set something wrong. I always
do this, every time I calibrate a map. Check that the grid line numbers
are correct too. It's easy to enter a wrong number or letter in the
calibration process.

Quoted message said:

I know the GPS will probably not give my *exact* position on the map,
but I thought it would be slightly more accurate than it was.

It's a helluva lot more accurate than that.

Quoted message said:

- Was my conversion from OS co-ordinates to Latitude / Longitude
above correct?

Don't know, but who cares? There's no reason to do it that way. You
could easily check with the grid overlay anyway. Lat/long is just
asking for trouble anyway, because OS lat/long is not the same as WGS84
lat/long.

Quoted message said:

- Ozi-explorer can calibrate maps using up to 9-point calibration.
Would calibration with more than two points give noticeably more
accurate positioning? After all, the map wasn't skewed or distorted
in any way so I thought two points would have been enough.

If there's no distortion then no. If there is then yes. Just overlay
the grid and you'll soon see if there are any distortions.

Quoted message said:

- Are there any other reasons why the positioning may have been out
by around 50 metres?

Wrong settings.

Paul
--
http://www.wilderness-wales.co.uk
http://www.wildwales.fsnet.co.uk

Code Developer <[email hidden]> wrote

Quoted message said:

Just a quick question regarding map calibration.

A couple of weeks back I bought Ozi Explorer and decided to try an easy walk
with a scanned map to see how reliable it would be.

The walk was around Mellor Knoll, (Dunsop Bridge in Lancashire), down
through Lower Fence Wood, across to Whitewell and back to Dunsop Bridge. I
scanned a small section of my OS map, just large enough for the walk and
calibrated it, using opposite corners of the map as calibration points, as
follows.

The map section I used was SD 580540 to SD 670450 and I used these two
points to calibrate the map corners as 53deg 58.822', 2deg 38.515'W and
53deg 54.008', 2deg 30.222'W.

Whoa!

I calibrate my Ozi maps to OS grid refs, as do most of us here, I'm
sure.
On an A4 size scan I only use two points, on the 1km grid intersections.
The usual reason for small position errors which you describe is due to
incorrect calibration of the map. ie - I use OSGrid, selected on the
Calibration screen.

Inconsistency between calibration of OziExplorer and the GPS is another
cause of similar errors.

I'm sure others will give a more detailed rundown, but that's my quick
response.

The help file in Ozi spells it out really, but ask again in here if in
any doubt.

--
Gordon

The message <[email hidden]>
from "Paul Saunders" <[email hidden]> contains these words:

Quoted message said:

Quoted message said:

- Are there any other reasons why the positioning may have been out
by around 50 metres?

Quoted message said:

Wrong settings.

Don't forget that a 50 m error is about what you get when you set the
GPS up with a mismatch of OSGB and WGS84 instead of BNG. BTDTGTT. :-)

--
Roger Chapman so far this year 69 summits
New - 47 (Marilyns 13, Sweats 5, Outlying Fells 35)
Repeats - 22( Marilyns 7, Sweats 16, Wainwrights 16)

"Code Developer" <[email hidden]>
mentioned:

Quoted message said:

Just a quick question regarding map calibration.

A couple of weeks back I bought Ozi Explorer and decided to try an easy walk
with a scanned map to see how reliable it would be.

The walk was around Mellor Knoll, (Dunsop Bridge in Lancashire), down
through Lower Fence Wood, across to Whitewell and back to Dunsop Bridge. I
scanned a small section of my OS map, just large enough for the walk and
calibrated it, using opposite corners of the map as calibration points, as
follows.

The map section I used was SD 580540 to SD 670450 and I used these two
points to calibrate the map corners as 53deg 58.822', 2deg 38.515'W and
53deg 54.008', 2deg 30.222'W. (2 point calibration). I used a free
conversion utility called NGR84.EXE by Specialist Systems to convert the OS
map co-ordinates to the Lat/Long co-ordinates required by Ozi.

Whilst I was there, the GPS almost showed the correct position, but for the
duration of the walk constantly showed me (on the map) as being around 50
metres south of my actual position.

I know the GPS will probably not give my *exact* position on the map, but I
thought it would be slightly more accurate than it was.

Before I do any more serious walks using OziExplorer I'd like to know :

- Was my conversion from OS co-ordinates to Latitude / Longitude above
correct?

- Ozi-explorer can calibrate maps using up to 9-point calibration. Would
calibration with more than two points give noticeably more accurate
positioning? After all, the map wasn't skewed or distorted in any way so I
thought two points would have been enough.

- Are there any other reasons why the positioning may have been out by
around 50 metres?

- And finally, can anyone recommend any other good walks up around the
Forest of Bowland? It's the first time I've been up walking around there
(apart from a couple of walks near Hurst Green) and I really enjoyed it.
Especially the magnificent views from the top of Mellor Knoll.

Regards,
Shaun.

Have you tried calibrating Ozi directly in Ordnance Survey - both map
datum and map projection? My scanned OS map segments of that area,
calbrated in OSGB, match well when tracks are superimposed.

I usually use 4 points for calibration.

Even when it's as wet as this there are lots of great loop walks on
the back roads in the area.

Brian
(change closed to open for email)

"Paul Saunders" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

In the configuration (alt-C), on the Maps tab, set the Alternate Grid to
BNG, and blank map datum to Ord Srvy Grt Britn.

When calibrating a map, set Map Datum to Ord Srvy Grt Britn and map
projection to BNG. Then simply enter the OS grid coordinates. Best to
place the calibration points on grid line intersections.

Also, in the configuration, on the System tab, set the Data File Datum
to WGS84. Also, on the GPS tab, set the GPS upload/download datum to
WGS84.

This may sound confusing, but remember this, the GPS *communicates*
using WGS84, but you want to *display* the information in OS BNG.

OK, done.

<snip>

Quoted message said:

An easy way to check that your map calibration is correct is this.
Click on Map - Grid Line Set Up, then click on the Other Grid tab and
tick the Grid On box. This will overlay a grid over the map. It should
match perfectly. If it doesn't, you've set something wrong. I always
do this, every time I calibrate a map. Check that the grid line numbers
are correct too. It's easy to enter a wrong number or letter in the
calibration process.

Done, also. Perfect match. I haven't tested the map with the PDA yet, but
the calibration looks good on screen. I'll try this calibration method next
time I do a walk, hopefully this weekend, depending on the weather.

Now, to find another walk up near Bowland.

Cheers Paul.
Regards,
Shaun.

"Code Developer" <[email hidden]> wrote in
message news:[email hidden]...

<snip ozi map calibration problem>

Thanks to all who replied.

I've tried the suggestion by Paul, and I'll see if this gives any better
results this weekend.

Regards,
Shaun.

Code Developer said:

Cheers Paul.

Good luck!

Paul
--
http://www.wilderness-wales.co.uk
http://www.wildwales.fsnet.co.uk