Forum: General Topics

Forums / General Topics / Speed issue

 

Subject:Speed issue 

diego

13:45
Thursday
10-Jul-2008

The other day I had a chance to ride the AVE high speed train from Barcelona to Madrid.

Different from planes, you can run all electronic devices throughout the whole trip, so it's a nice occasion to test a gps device along.

I found a consistent difference of about 80 kms/h between TMJ speed of 220 km/h and the train's speed panel, which showed around 300 km/h most of the 2.30 hours trip. I also used a laptop plus various GPS programs showing also 300 km/h, so I tend to think they may be some issue with GPGGA or GPRMC message handling at higher speeds?

I have also flown TMJ across Europe several times, but the airliner was not showing speed over ground, so I had no way to detect any speed difference.

Any other user has detected that?
 

Stephen

18:08
Thursday
10-Jul-2008

Location:
Surrey, UK

Phone Model:
BB 9800 Torch
BlackBerry 8900
SE W910i
Nokia 5800

Hi Diego,

You're lucky - last year I tried to record a track on a train journey from Valencia to Madrid, but the 'insulation' in the carriage was so good that I wasn't able to get a GPS signal at all for the whole journey - very annoying!!

If this is a bug I certainly need to look into it - I don't suppose you created a log file from the journey (NMEA ideally) that I could have a look at?

I'm just looking into a train log from a couple of months ago from London to Leeds to see if I can see a similar error (perhaps it only shows itself at higher speeds - our trains only get to 125mph here!), but so far haven't noticed anything too odd. Track distances certainly seem to be pretty accurate (I'm comparing with values calculated by GPSU).

The speed readings are calculated on the fly by working out the distance between subsequent points and dividing by the time difference (assuming that the Settings/Use Speed & Bearing from GPS option isn't selected). Theres a chance that this might result in small rounding errors, but I'd be surprised if its 20% out, as you seem to have found!?

I'll keep an eye on this myself, see if I can recreate it...

Cheers,
Stephen

 

diego

8:06
Friday
11-Jul-2008

Hi, Stephen.

Well, I had to stick the GPS dongle to the window for some ten minutes before getting the 1st fix. Once done, I could track the signal from even my front table. Similarly on a plane.

No, I am sorry I did not save any log file. I will do it next time.

Discrepancies at lower speeds (walking or car) may be easily neglected. And at higher speeds (jet liners) I have no mean to compare unless I get into the pilot's cabin.

I will let you know any further finding.
 

diego

7:57
Wednesday
6-Aug-2008

Hi again, Stephen.

I just remembered there's an autotrack of this trip last 3rd, July on Devices/My Mobile at this very same webpage, showing an average speed of 247 km/h and a maximum speed of 276 km/h. At that time, train was running at around 300 kms/h.

Let me know if that that can be of any help, and how could you get it.

Regards,
 

Stephen

19:31
Wednesday
6-Aug-2008

Location:
Surrey, UK

Phone Model:
BB 9800 Torch
BlackBerry 8900
SE W910i
Nokia 5800

Hi Diego,

I've had a look at that 3rd July trip in your Devices folder - hope you don't mind! I've exported it to GPS-Utility and analysed the speeds between the first few trackpoints, results as follows:

# -TMJ- GPS-U
1 ----- -----
2 275.8 275.1
3 256.1 251.1
4 262.8 268.8
5 263.5 264.3
6 247.0 246.6
7 240.8 240.5
8 240.7 240.5
9 233.6 229.5
(Units in km/h)

Average speeds:-
TMJ: 247km/h
GPS-U: 246.841km/h

As you can see these figures do seem to be pretty similar - the individual trackpoint speeds vary slightly, but this could be due to the slightly different ways that each program deals with Time (TMJ takes milliseconds into account, but the GPX file imported into GPS-U only contained whole seconds)...

Are you able to check the track calculations in any other software - at the moment I'm tending to side with TMJ!!

One other comparison might be to look at two individual trackpoints, eg points 6-7:

#6 Lat: 41.284843, Lng: -1.900922, Alt: 706.90
#7 Lat: 41.274608, Lng: -1.923513, Alt: 735.99
Time between points: 33.0 seconds

Distance between points (calculated by TMJ Website): 2.2081897 km
Distance between points (calculated by GPS-Utility): 2.2 km
Distance between points (calculated by this website): 2.204 km

Again, the figures all seem pretty similar - they're certainly not out by anything that could account for the 300kmh/274kmh discrepancy!

All-in-all these comparisons don't necessarily rule out a bug in TMJ-Mobile since the data on this website had already been recorded/filtered by the program. An NMEA log is really the only way to do a completely fair comparison (recording the NMEA log in TMJ is fine though since it logs the entire unfiltered data from the GPS).

Any thoughts?

Regards,
Stephen
 

Alec

15:01
Thursday
18-Sep-2008

The speed a gps reads will never be 100% accurate, there are too many factors.
an example of one of these factors is the atmosphere thickiness above you, as the signal (an ellectromagnetic (light) microwave i belive) has to go from a less dence medium, to a more dence medium, it refracts, in the case of earth and the conditions stated, towards the earth, shorting the path that the signal has travaled and thus changing the distance the signal has to traval, chaing the time taken, making the gps reading inaccurate.
another thing which can cause the gps to be inaccurate is the clocks abourd the satalights, even though they are attomic, they are only accurate to few billion billionts of a second, (doesnt sound much but when you concider that light travles 300,000,000 meters per second it can be alot) adn makes it inaccurate.
it is due to the inaccuracy of time that more statalites = more accurate position, it averages them.

these are just some of the resons, i hope this helps you be alittle more forgiving to this still new technology :)

alec

hope this helped
 
 

diego

16:21
Tuesday
16-Dec-2008

Hello.

Of course, I don't mind you taking a look at my tracks. I hope they are of any help.

I also took a second look at this issue on a trip I made last 10.12.2008 on board a high speed train from Barcelona to Madrid, for over 264 kms.

The maximum speed reached over this track was 295 km/h, and I compared from time to time the speed indicated in the cabin with the one showed in my TMJ-powered mobile.

I saw no significant differences. So I guess it was a glitch or something, maybe even at the cabin indicator. Again, feel free to take a look at the data on my Auto-Track folder 10.12.2008 (user diego).

Thank you very much for your interest, and keep up with the good job.
 

(You must be logged in to post a reply to this thread)