|
Cellular forums Home > Archive > Magellan GPS > October 2005 > Magellan SporTrak Color : inaccurate altitude calculation?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Magellan SporTrak Color : inaccurate altitude calculation?
|
|
| Anthony 2005-10-02, 5:49 pm |
| I can't get an accurate altitude reading from my Magellan SporTrak
Color. I compared two locations with considerable altitude difference,
but the reading was almost the same. I had good reception and I gave
the reading enough time to stabilize. Is this a general problem with
the Magellan SporTrak Color?
thanks
| |
| Sam Wormley 2005-10-02, 5:49 pm |
| Anthony wrote:
> I can't get an accurate altitude reading from my Magellan SporTrak
> Color. I compared two locations with considerable altitude difference,
> but the reading was almost the same. I had good reception and I gave
> the reading enough time to stabilize. Is this a general problem with
> the Magellan SporTrak Color?
>
> thanks
>
You need to look at the readings at both locations at many different
times to get a clear picture of the range of elevation readings
reported by consumer handheld receivers.
Average vertical error should be less than 22 meters 95% of the
time. See page 15 of
http://www.navcen.uscg.gov/gps/geni...andardFINAL.pdf
-Sam Wormley
http://edu-observatory.org/gps/gps_accuracy.html
| |
|
| Two things to keep in mind about GPS receiver vertical measurement
performance, and in particular performance of commercial civilian receivers
in common use.
1. Vertical dilution of precision, the vertical component of total geometric
error is, roughly, twice the total horizontal error. The problem is created
by not regularly having a satellite more or less directly overhead.
2. The altitude reading your are getting is probably an estimate of height
above mean sea level. This is probably done in order to have some
connection with map products. I have long wished that all civilian
receivers also would output the three ECEF coordinates, and height above the
"reference ellipsoid". Having ECEF would allow you to compute whatever
other frame you choose. Having height above the reference ellipsoid avoids
the vagaries of localizing mean sea level.
I had a brief exchange with both Magellan and Garmin people regarding the
subject of getting ECEF and other outputs, and their answer was an arrogant
"never happen". Early in my search, I thought NMEA messages would be
available with that information, but I was naive. I continue to hope that
measurements in their more raw forms will be made available. Interestingly,
some of the OEM boards provide a rich set of data.
On 10/2/05 12:07, in article PdW%e.384166$_o.13185@attbi_s71, "Sam Wormley"
<swormley1@mchsi.com> wrote:
> Anthony wrote:
>
> You need to look at the readings at both locations at many different
> times to get a clear picture of the range of elevation readings
> reported by consumer handheld receivers.
>
> Average vertical error should be less than 22 meters 95% of the
> time. See page 15 of
> http://www.navcen.uscg.gov/gps/geni...andardFINAL.pdf
>
> -Sam Wormley
> http://edu-observatory.org/gps/gps_accuracy.html
| |
| peter 2005-10-02, 11:48 pm |
| bcd wrote:
> Having height above the reference ellipsoid avoids
> the vagaries of localizing mean sea level.
> I had a brief exchange with both Magellan and Garmin people regarding the
> subject of getting ECEF and other outputs, and their answer was an arrogant
> "never happen". Early in my search, I thought NMEA messages would be
> available with that information, but I was naive.
Odd that you got that response since my Garmin receivers (III+, 12,
eMap, V) all display the estimated height above the geoid but also
indicate the ellipsoid vs. geoid correction factor used in the NMEA
$GPGGA message. That allows calculation of the height above the WGS-84
ellipsoid (and would allow determination of the ECEF coordinates as
well) and also a more accurate height above MSL if a better correction
factor is known. For example, I've found that when kayaking on the
Pacific near home with my eMap the elevation numbers were about 16' too
low which corresponds to the 5 m difference between the ellipsoid vs.
geoid figure used by Garmin and the one I found on the USGS site.
| |
| Eddy [UK] 2005-10-02, 11:48 pm |
| Hi Anthony,
I have a SporTrak Pro which is essentially the same as yours but with
a monochrome screen.
I have experimented with many ways of carrying my GPSr and analysing
the results. My unit will only give good altitude results when it
has a completely unobstructed view of the sky right down to the
horizon all around. E.g. if I hold it in front of me at chest level
to read the screen, my shoulders, neck and head are all obstructing
part of the sky and I get a reduced accuracy altitude reading. For
the best altitude readings I carry my GPSr on a short pole such that
the antenna is level with the top of my head at all times. Another
way of achieving the same thing is to use a re-radiating antenna sewn
into the top of your hat - I've known several people do that.
Given a completely unobstructed view of the sky my GPSr does give
excellent altitude readings, often within 5 metres and usually
within 10 metres of true.
[color=darkred]
What sort of altitude difference are you talking about?
All the best, Eddy [UK]
| |
| John Navas 2005-10-02, 11:48 pm |
| X-Newsreader: Forte Agent 1.93/32.576 English (American)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 29
Message-ID: <OyZ%e.102479$qY1.83052@bgtnsc04-news.ops.worldnet.att.net>
NNTP-Posting-Host: aae4fcb946b77dff29e4
1dbc76d20ed1
X-Complaints-To: abuse@worldnet.att.net
X-Trace: bgtnsc04-news.ops.worldnet.att.net 1128293678 aae4fcb946b77dff29e4
1dbc76d20ed1 (Sun, 02 Oct 2005 22:54:38 GMT)
NNTP-Posting-Date: Sun, 02 Oct 2005 22:54:38 GMT
Organization: AT&T Worldnet
Date: Sun, 02 Oct 2005 22:54:38 GMT
Xref: number1.nntp.dca.giganews.com alt.satellite.gps.magellan:4739 sci.geo.satellite-nav:400729
[POSTED TO alt.satellite.gps.magellan - REPLY ON USENET PLEASE]
In <1128293063.472681.159150@g14g2000cwa.googlegroups.com> on 2 Oct 2005
15:44:23 -0700, "Eddy [UK]" < one_computers@compus
erve.com> wrote:
>I have a SporTrak Pro which is essentially the same as yours but with
>a monochrome screen.
>
>I have experimented with many ways of carrying my GPSr and analysing
>the results. My unit will only give good altitude results when it
>has a completely unobstructed view of the sky right down to the
>horizon all around. E.g. if I hold it in front of me at chest level
>to read the screen, my shoulders, neck and head are all obstructing
>part of the sky and I get a reduced accuracy altitude reading. For
>the best altitude readings I carry my GPSr on a short pole such that
>the antenna is level with the top of my head at all times. Another
>way of achieving the same thing is to use a re-radiating antenna sewn
>into the top of your hat - I've known several people do that.
>
>Given a completely unobstructed view of the sky my GPSr does give
>excellent altitude readings, often within 5 metres and usually
>within 10 metres of true.
FWIW, my modest Garmin Foretrex 101 does a much better job of altitude
accuracy than my Magellan SporTrak Color.
--
Best regards, HELP FOR CINGULAR GSM & SONY ERICSSON PHONES:
John Navas <http://navasgrp.home.att.net/#Cingular>
| |
| Anthony 2005-10-03, 2:48 am |
| Hi Eddy,
I am talking about 50m difference. The horizon was not completely clear
as you described. I will follow your suggestion and let you know.
thanks
>
> What sort of altitude difference are you talking about?
>
> All the best, Eddy [UK]
| |
| Koen De Poorter 2005-10-03, 5:48 pm |
| On 02-10-2005 18:03, in article
1128272393.797026.290600@o13g2000cwo.googlegroups.com, "Anthony"
<amartin@whoever.com> wrote:
> I can't get an accurate altitude reading from my Magellan SporTrak
> Color. I compared two locations with considerable altitude difference,
> but the reading was almost the same. I had good reception and I gave
> the reading enough time to stabilize. Is this a general problem with
> the Magellan SporTrak Color?
>
>
I am under impression that the Sportrak Color also takes in account the
barometric height and under changing weather conditions strange things can
happen but after enough time things should stabilize.
I once cycled up a hill and the altitude shown dropped considerably, the
reading was under trees and I suspected that when you have bad satelite
reception, the unit switches to barometric height estimation.
Never had any doubts about the height reported in open fields though...
Koen
| |
| Eddy [UK] 2005-10-03, 5:48 pm |
| John,
I have read several forum messages indicating that Garmins are better
at altitude and trip distance than Magellans under less than ideal
satellite viewing conditions. I've never tried a Garmin myself.
If I carry my SporTrak in my pocket I get poor altitude and trip
distance results, on my rucksack shoulder strap they get better, and
carried with the antenna not below the top of my head they are
excellent.
Anthony,
Please let us know how you get on, thanks.
Koen,
I've just downloaded the SporTrak Color's manual(s) and there is no
mention of the unit using the barometric pressure to estimate the
elevation (altitude) when there is no satellite signal, so it would
seem that this is only for weather use. Please correct me if I've
missed something.
Eddy [UK]
| |
| Koen De Poorter 2005-10-04, 5:48 pm |
| On 03-10-2005 14:05, in article
1128344735.311425.219470@g49g2000cwa.googlegroups.com, "Eddy [UK]"
< one_computers@compus
erve.com> wrote:
> I've just downloaded the SporTrak Color's manual(s) and there is no
> mention of the unit using the barometric pressure to estimate the
> elevation (altitude) when there is no satellite signal, so it would
> seem that this is only for weather use. Please correct me if I've
> missed something.
I made the remark out of my head and based upon some practical experience
and observations. You're right it is not mentioned in the manual but some
others have suspicions too:
At gpsinformation.net I read:
> The relationship of the barometer to the GPS altitude isn't known at this
> time._ However, the barometer readings don't increase with altitude decreases
> (and vice versa) as the Garmin Vista (incorrectly) does._
So I could be wrong but I also could be right I guess.
Koen
| |
| Eddy [UK] 2005-10-04, 5:48 pm |
| Hi Koen,
>
> So I could be wrong but I also could be right I guess.
LOL! Reviews and manuals should explain something, not create
confusion. I'm none the wiser.
Eddy [UK]
| |
| Koen De Poorter 2005-10-04, 5:48 pm |
| On 04-10-2005 16:11, in article
1128438702.928435.232800@g43g2000cwa.googlegroups.com, "Eddy [UK]"
< one_computers@compus
erve.com> wrote:
> Hi Koen,
>
>
> LOL! Reviews and manuals should explain something, not create
> confusion. I'm none the wiser.
>
> Eddy [UK]
>
You're right, some more research shows that on the Sportrak barometric
pressure and temperature is not used to increase altitude readings.
(Some explorist models do apparently)
--- In sportrakgps@yahoogro
ups.com, "mhannebert" <mhannebert@y...> wrote:
> This is the message I send to Maggelan Suport :
>
> "I have a Sportrak color. I read the manual and I'm surprised that
> the barometer is not used to calculate more precisely the altitude.
> In fact this is done already by Garmin in the most advanced GPSr.
> Why isn't it done by Magellan ? Will Magellan provide an update of
> the firmware that includes this function ?"
>
> Here is the answer I received for Magelan suport :
>
> "We do not intend the barometer to be used for this feature. The
> barometer in the unit was designed strictly to provide weather
> pattern history for the sailor. This unit is mainly aimed at the
> marine market. Typically the altitude while at sea does not change.
> Best regards.
> Brian Oliver
> Service Support Supervisor"
Koen
| |
|
| I also asked the Magellan tech response if C/No estimates were available
somehow, and got the same answer ("Not gonna happen"). Later I used some
general purpose code to read the NMEA messages my Magellan Meridian Platinum
was sending. Lo and behold the C/No slots had data! In fact, if it they
were correct, the data answered a problem I was having that acted exactly
like cross-correlation. The numbers were quite high, reaching the low 50's
(assuming dBHz), which certainly could explain the problem. So, it sounds
like I was pretty much universally lied to!
On 10/2/05 13:51, in article
1128286278.492669.176030@g47g2000cwa.googlegroups.com, "peter"
<prathman@comcast.net> wrote:
> bcd wrote:
>
>
> Odd that you got that response since my Garmin receivers (III+, 12,
> eMap, V) all display the estimated height above the geoid but also
> indicate the ellipsoid vs. geoid correction factor used in the NMEA
> $GPGGA message. That allows calculation of the height above the WGS-84
> ellipsoid (and would allow determination of the ECEF coordinates as
> well) and also a more accurate height above MSL if a better correction
> factor is known. For example, I've found that when kayaking on the
> Pacific near home with my eMap the elevation numbers were about 16' too
> low which corresponds to the 5 m difference between the ellipsoid vs.
> geoid figure used by Garmin and the one I found on the USGS site.
>
| |
| Jack Erbes 2005-10-06, 5:48 pm |
| bcd wrote:
> <snip details> So, it sounds
> like I was pretty much universally lied to!
Don't you think that there a possibility that you happened to speak to
people that were not completely aware of every subtle detail about their
products and NMEA data?
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
|
| It's more than just a possibility, but so what? These were supposed to be
technical support people. If they didn't know, a simple "I don't know,"
would have been an infinitely better answer. Instead, I got very emphatic
(and I sensed a bit arrogant) answers that were wrong. That's inexcusable.
As far as knowing what's in your products' data messages, distributing
copies to the technical support crew would seem to be an obvious thing to
do. Even though the NMEA documents themselves are (were?) too expensive to
expect the average consumer to buy, having them available for technical
support folks seems a reasonable expectation.
On 10/6/05 07:00, in article 8KCdnfHvlflrsNjeRVn-uA@adelphia.com, "Jack
Erbes" <jackerbes@adelphia.net> wrote:
> bcd wrote:
>
>
> Don't you think that there a possibility that you happened to speak to
> people that were not completely aware of every subtle detail about their
> products and NMEA data?
>
> Jack
|
|
|
|
|