View Full Version : gps and ground speed

1st Jul 2007, 11:42
I always assumed that gps would give accurate ground speed. Is this the case? I have a new gps in the motor and it shows a -4mph discrepancy with the speedo. New motor (SAAB) so I would hope that the speedo is accurate.
I am interested both from the aviation point of view and the motor. 4 MPH ground speed over 2 hrs flying is not too much of a discrepancy I know but it is interesting.

1st Jul 2007, 11:58
The speedo is wrong. They are desisgned to overread as this puts you on the safe side. I do not think they are allowed to under read.

1st Jul 2007, 13:10
A consistent 4mph over read over its range is a bit large. I take your point though.

green granite
1st Jul 2007, 13:16
Max permitted error I believe is 10% (well it used to be anyway) so @ 40mph 4mph would be right on the limit, @ 70mph well within the permitted tolerance.

1st Jul 2007, 14:08
The permitted tolerance is -0%, +10%. Since the accuracy is affected by rolling radius (tyre wear and air pressure) as well as inherent variations within the tachometer itself, the manufacturers generally aim at the centre of the permitted range, i.e. +5%.

1st Jul 2007, 14:40
I take it then that the GPS can be taken as accurate. There isn't a variation betweet types. Logic tells me that GPS must ba accurate. Makes me woner how accurate the other prameters are such as MPG as expressed by the on board computer.

1st Jul 2007, 21:20
Keep in mind also that the speedo knows nothing about hills or turns, which will affect your groundspeed somewhat, especially the former.

FE Hoppy
1st Jul 2007, 21:36
Bit of blind faith in GPS on here.
As the GPS position accuracy is at best 35meters and gps speed is relative to a geometric spheroid based on the centre of the earth I would stick with your taco.

1st Jul 2007, 21:44
You may well find that the GPS will give you some G/S when you are stationary - for "raw" GPS this is normal given the accuracy... the true value is probably somewhere inbetween the GPS and taco values.

1st Jul 2007, 22:53
Keep in mind also that the speedo knows nothing about hills or turns

Ah that is the thing that was niggling me. It is obvious now I see it.:ugh:

Still the error was consistent.

Jeremy Caldwell
2nd Jul 2007, 12:18
Interesting- I purchased a hand held gps which is stated to be accurate to 0.1 knot,to check ground speed when sailing to take advantage of tides and current flows.Out of interest i thought I would check my nissan's speedo. To my surprise it was reading 5mph over on the car. the car is also fitted with sat -nav but only gives average speed. The average speed calculations for both gps was indentical.I have checked my colleages cars and all vary between 1-5 mph over read which I understand the manufactures adjust to protect liability.

2nd Jul 2007, 14:19
"A consistent 4mph over read over its range" is actually pretty typical.

The EU accuracy spec is as follows:

2.3.7. the speed displayed must never be lower than the actual speed. Between speed V1, read on the speedometer and actual speed V2: there must be the following relationship with the test values specified in item 2.3.5 and between those values:

0 <= (V1 - V2) <= 0,1 V2 + 4 km/h.

As Sallyann1234 says, manufacturers will always aim for mid range of such a spec, but modern speedometers are remarkably accurate and so cluster quite closely around an overread of about 5 mph.

3rd Jul 2007, 11:43
You may well find that the GPS will give you some G/S when you are stationary

Really? The Garmin 1000 I use regularly not only accurately records taxiing around the apron, it re-orients as the aircraft turns (track up, so map rotates during turns) while taxiing at walking pace. I'd tech it if it showed map shift while stationary.

GPS over tachometer any day.

3rd Jul 2007, 11:51
The Garmin 1000 I use regularly not only accurately records taxiing around the apron, it re-orients as the aircraft turns (track up, so map rotates during turns) while taxiing at walking pace. I'd tech it if it showed map shift while stationary.There are 2 issues here... The GPS "raw data" and the software processing of that data by the Garmin.
GPS raw data will move around within the error tolerance, caused by inherent errors, receiver quality and signal strength. When stationary it will give +-2-3Kts, and the Track outputs rather random - in short unreliable.
It now comes to the software as to how to handle "smoothing" these errors. "Over smooth" and you get lag and essentially best guess information - as your Garmin is showing. "Under smooth" and you get small shifts, track / GS jumping around. The latter is more accurate if you really want the know what is going on - the former more pleasing if you just like to believe it ;)

3rd Jul 2007, 12:01
GPS will generally give very accurate speed readings, though by definition, it displays an average spped over a short period of time. So it will under-read when increasing speed, and over-read when slowing down. It is also subject to various errors from time to time, which will affect the reading, especially when the signals are weak, or the the geometry of available satellites give rise to significant errors. These errors have become significantly reduced since Selective Availability was switched off in the late 90s.

I have yet to see a car speedometer which does anything but over-read by some 2 - 10%. However car speedometers are usually very consistent at given speeds, over time.

So the best approach could be to calibrate a car speedometer at various speeds when at a steady speed using GPS, then use this knowledge of the error to derive an accurate speed in any situation (much like a deviation card).

4th Jul 2007, 08:33
Fair enough NigelOnDraft. I bow to your technical knowledge. Perhaps most users don't want to see velocity vectors jumping rapidly in direction and magnitude, it being easier to read time-averaged digital outputs.

No Country Members
4th Jul 2007, 08:58
GPS raw data will move around within the error tolerance

True of course, but it's good enough for government work.

4th Jul 2007, 13:19
Some of you guys need to get out to play more!

I have a Garmin 430 in the aeroplane and a Garmin 296 portable that I use in the aeroplane, the car and the boat.

A few observations about GPS.

1) Steady state GS is very accurate and at 100 kph is spot on with the digital speedo in my car (GMH Commodore SV8) - this has been calibrated against a number of police radars - at some expense !!!!!!!

2) If you have a residual GS while stationary, your GPS is stuffed. I have ever never seen this. When stationary, my GPSs read 0 GS.

3) Although they say GPS is accurate to 35m in my experience it is closer to 1m. I regularly use the 296 to locate underwater structure when fishing. I can navigate to previously marked waypoints to within 1m.


5th Jul 2007, 04:59
My understanding is that GPS errors are (mainly) systematic:

This is why there are augmentation systems (Differential GPS, WAAS) built around the concept of a ground station at a known location providing offset information as to the system error, allowing further calibration of receivers in the area.

This also would explain the ForkTailedDrKiller's observations - the *accuracy* may well be 35m, if comparing the reported position to the actual position. However, the *repeatability* is rather better if all you want to do is hit the same place / waypoint again, rather than finding a spot from a given lat/long.

This would also suggest that the velocity measurement (rate of change of position) is somewhat more accurate than the absolute position.

Interestingly enough, I also used to drive a saab (9-5 with an electronic speedo), and found a pretty consistent 4mph underread accross the range - suggesting that the spedo is pretty accurate, and has a built in factor.

Other cars including older saabs with mechanical speedos tend to show a percentage error.

5th Jul 2007, 06:39
Being involved in a business that does GPS mapping/survey that guarantees a data accuracy less than 4cm and aims for 2cm whilst moving, I have a little understanding of this issue.

First of not all GPS's are created equal. The gear we use, uses the Russian Glonass as well as U.S. satellite system and is also conected by radio to a base station to give a real-time correction. We consistently achieve 2 cm accuracy. Hand helds, car systems and basic aircraft GPS all vary from 1m to 35m accuracy and worse in some cases.

The point he Mark1234 raises is very valid though. It is more about 'repeatability' rather than variability. Even the 35m accurate systems will probably get you back to the same spot, pending clarity/accuracy of the screen read-out, within a short time period. 2 hours later, next day, change in the weather, satelites in view and position can very quickly see that position change however. What they will not do is reliably take you to a pre-detirmined spot on the earth with any accuracy. Thus the position of a 10m accurate system can vary 10m radius from the given point or up to 20m from a point marked previously.

Given all that I would generally trust the GPS more in a steady state on level ground. Tyre pressure and wear does have an impact on the speedo reading in my car as well. As they wear the difference tends to go from an indicated 2kpk faster than the GPS to 4kph when needing replacement. If you were on a 45degree incline then the GPS would obviously be under-reading drastically.

If you wish, it is easy to use a stopwatch to check your speedo. I once had a car that when I first drove it I seemed to be passing cars all the time even when driving at the limit. A police radar fine a few months later alerted me to the fact it under-read by 12 kph. This was traced back to the wrong drive gear fitted to the speedo gearbox drive. Still had to pay the fine though.

My thought anyway. ;)

FE Hoppy
5th Jul 2007, 11:59
The points about Differential GPS, WAAS are very interesting as this disscusion is moving towards RNP/EPU and RNAV operations. Inorder to fly approaches with RNP < 0.3 the aircraft requires and Augmented GNSS system as raw GNSS cannot guarantee accuracy. in some places RNAV procedures are published with (GNSS) as the stated nav aid required to acheive the accuracy in others it is now being removed. The bottom line is the Aeroplane needs to be able to evaluate the nav accuracy and alert the crew if its not able to meet the RNP.

Getting back to Hand helds or in car systems. I have a unit that knows where the speed cameras are and I use the distance coundown to judge the accuracy of the GPS position. Often as I pass the camera there is still some meters to go or the position is already passed before I get to the camera. The shift from too late to too early can happen in a short distance so that from being too early by 20 meters it can bacome too late by the same distance after just 500 meters travelled. If this same position uncertanty is used to calculate the speed without any smoothing then I wouldn't trust it over the tacho.

8th Jul 2007, 21:56
You can check this for certain. Al you need to do is observe whether the GPS knows where you are. If it does then it must give you an exact ground speed as its clock is extraodinarily accurate. Its true it is using a simplified model of the Earth's surface but that is unlikely to make any noticeable diiference under normal circumstances.

8th Jul 2007, 22:15
"I have a unit that knows where the speed cameras are and I use the distance coundown to judge the accuracy of the GPS position."
FE Hoppy
Your position indicated by the GPS unit is accurate to within a few tens of metres.
It is the location of the speed camera that is in doubt, because the lists of locations are compiled from a variety of sources including amateur users and are frequently inaccurate.
If you repeat the same journey you will find that the indicated error at each camera is consistent - always the same every day.

9th Jul 2007, 02:28
Two points that seem to me missing in the above responses:

1. Consumer GPS generally gives Doppler a considerable weight in the displayed speed--so it is not just a matter of position error deltas for speed error. Generally Doppler does better, which is why they use it. It also responds quite quickly, though not instantaneously, depending on what sort of smoothing the manufacturer puts in for post-processing.

2. Confidence from displayed zero velocity is a bit misplaced. Consumer units generally clamp the displayed velocity to zero when the initially computed value is below some threshold. I noticed they lowered the threshold on the ones I use when SA got turned off. I think I recall it used to be about 1 mph. A few minutes walking about extremely slowly with my Garmin GPS V just now with WAAS turned on and eleven well-positioned satellites in view never once displayed 0.1 km/h, though 0.3 and 0.0 appeared many times, and 0.2 often. I surmise the clipping limit in the current firmware in that condition was probably a little over 0.2 km/h.

Despite those two points, I agree with those who suggest that a _consistent_ difference between an automobile speedometer and even a very cheap consumer GPS is almost certain to find the GPS in the right of things.

Disclaimer: I'm not a pilot, but am an electrical engineer, and have used GPS a lot since 1991.

10th Jul 2007, 11:58
Oooh I'm so glad I started this one. Thank you all for such a stimilating discussion. :ok:

10th Jul 2007, 12:15
Maxter, I hope you're still reading this.

The professional GPS equipment you're using. I've always assumed that such equipment, used for mapping purposes, is mounted on a tripod or something and has a button or something that allows you, as the operator, to tell it that its position is now fixed. From that point on, the receiver can start collecting GPS positions and average these positions out over a longer period of time (minutes, hours, perhaps weeks)? This, combined with a differential GPS broadcast device, set up locally, in a known position, will then give you 2cm accuracy, not?

But an average consumer GPS (up to and including the G-1000) will not have the benefit of a "steady state" button, so it has to assume that it's moving all the time, making averaging over a periode more than a few seconds impossible, and will not have the benefit of a locally produced DGPS signal so has to rely on WAAS, if WAAS is available for that area at all. Hence the 6-35 meter accuracy for these devices?