I remember somebody telling me that speedo's can be anything upwards of 10% inaccurate. Hence the reason the boys in blue will only prosecute at 34mph - 45mph - 78mph.

.... but it could be one of those urban myths :err:

 

The GPS speedo is usually as accurate as the speed cameras... As far as I know...

 

legal requirement for a speedo in a car is -0% +10% of the actual speed. Most car manufacturers err on the side of caution to ensure they meet the -0% error. So a lot of cars read fast. The GPS will be accurate if you are travelling straight and have good triangulation with 2 or more satelites.

so in short both are probably as accurate as they are meant to be.

 

How do you think calibration factors will effect the accuracy of both the cars speedo, and the GPS system. Is GPS a system which would require calibration?

Dunno, just my ignorance!

Fcek aye,
Nutz.

 

GPS requires no calibration. The positioning aspect of GPS is usually only accurate to one or two meters, less if Differential GPS is used. However, the speed measuring of GPS is usually super accurate, As Dave says, as long as you have good triangluation off 3 or more satelites. There can be up to 8 satelites insight at once, but not all devices can cope with that many.

 

GPS requires no calibration. The positioning aspect of GPS is usually only accurate to one or two meters, less if Differential GPS is used. However, the speed measuring of GPS is usually super accurate, As Dave says, as long as you have good triangluation off 3 or more satelites. There can be up to 8 satelites insight at once, but not all devices can cope with that many.

Errm, not quite.

Standard GPS is defused in some way to stop a decent pin point accuracy. There is a decoder available (but only to the US Military seeing as GPS belongs to them) that can give much more accurate GPS, eg down to a foot.

However, to get around this the cheapy GPS manufacturers (eg everything under 10k), used more sats and a specific technique to increase accuracy.

My Garmin will take 12 sats at once, and give a measurement down to feet, but not a foot.

Technically three sats should pin point you, but with the encrption used and the method employed by the handheld you are looking at 100 to 200 foot resolution.

The other thing that improves resolution is movement. But only slightly.

Also the speed readout is useless unless you are cruising. As a layman I would think that it only plots the speed every nth second and not your speed in real time, so the GPS always lags behind the speedo in a big way.

Once you cruise then the GPS is pretty accurate.

My TF is around 5% out. I checked.

Tin

 

Thanks for the advice guys....... Excellent info as usual

 

Errm, not quite.

Standard GPS is defused in some way to stop a decent pin point accuracy. There is a decoder available (but only to the US Military seeing as GPS belongs to them) that can give much more accurate GPS, eg down to a foot.

However, to get around this the cheapy GPS manufacturers (eg everything under 10k), used more sats and a specific technique to increase accuracy.

My Garmin will take 12 sats at once, and give a measurement down to feet, but not a foot.

Technically three sats should pin point you, but with the encrption used and the method employed by the handheld you are looking at 100 to 200 foot resolution.

The other thing that improves resolution is movement. But only slightly.

Also the speed readout is useless unless you are cruising. As a layman I would think that it only plots the speed every nth second and not your speed in real time, so the GPS always lags behind the speedo in a big way.

Once you cruise then the GPS is pretty accurate.

My TF is around 5% out. I checked.

Tin

The inaccuracy is called Selective Availablility which can be selected and deselected by the USA Military. Currently it is switched off, and as such the GPS should be accurate to within a metre or so (though this will depend on how many satelites are actually being tracked and used for triangulaion.

A GPS is far more accurate than a speedo would be, and even when accelerating it can easily adjust within half a second or so, unless you have a really cheap GPS, even the ones costing a couple of hundred pounds will record an accurate speed even when braking and accelerating.

I have used one on many occasion and have found that the increase in speed of the GPS generally matches that of a speedo.

Steve (who used to install, setup and fix GPS receivers )

 

Errm, not quite.

Standard GPS is defused in some way to stop a decent pin point accuracy. There is a decoder available (but only to the US Military seeing as GPS belongs to them) that can give much more accurate GPS, eg down to a foot.

However, to get around this the cheapy GPS manufacturers (eg everything under 10k), used more sats and a specific technique to increase accuracy.

My Garmin will take 12 sats at once, and give a measurement down to feet, but not a foot.

Technically three sats should pin point you, but with the encrption used and the method employed by the handheld you are looking at 100 to 200 foot resolution.

The other thing that improves resolution is movement. But only slightly.

Also the speed readout is useless unless you are cruising. As a layman I would think that it only plots the speed every nth second and not your speed in real time, so the GPS always lags behind the speedo in a big way.

Once you cruise then the GPS is pretty accurate.

My TF is around 5% out. I checked.

Tin

to use your own phrase errm not quite.

Standard GPS is not defussed in any way, it's simple (well ok quite complex physics) that affect the accuracy. Its true that the forces are the only ones to have access to the signals on the L2 carrier, but there are other ways to get around this (differential GPS) and get accuracies to within 10 cm2.

General accuracy without these clever circuits and trilaterating with only 3/4 sats should still give 2 - 6 meters accuracy depending on the sats accessed. Of course really cheap GPS ignore the GDOP principle and hope for the best and this can lead to some strange results.

Movement only improves accuracy on really cheap GPS, otherewise like most movement it makes accuracy worse as it adds another parameter and more doppler effects.

speed is a side effect of the positional calculation, but as you say, it is not usually available in real time on cheap systems. Hence my original comment about travelling in a straight line, but I should add steady speed helps.

 

even the ones costing a couple of hundred pounds will record an accurate speed even when braking and accelerating.

obviously mine is too cheap then.

using the golf recently I have easily accelerated and been around 10-12mph ahead of the GPS, and in a couple of cases was braking hard when the GPS was still going up.

there is a delay on mine which makes it virtually impossible to give a true reading when accelerating and braking. only when you cruise does the gps give the true accuracy.

Tin

 

Maybe it is not that cheap, but the refresh is not that fast. I have used a Garmin and it was fine for speed measurement. There will be a slight delay, but when I was checking mine, it was around half a second.

Anyway when you are accelerating it doesn't matter much as you would not really be interested in the delay whilst accelerating. I would only compare when I have got to your desired speed.

Steve

 

My MGF is quite a bit out too - at 90MPH Speedo reading, my iPAQ with a Navman GPS sleeve tells me I'm really doing 83MPH...

 

The difference in speedo accuracy is about 3% just on the difference between new and old tyres....

So 2 cars with speedos reading 100mph (1 with old and 1 with new tyres)....
There will be 3mph difference between the cars...quite noticable...