Which is more accurate the spedometer OR GPS?

optiplexgx270

Ok so a Ford Focus and a Garmin 370...

On the speedometer it reads:

120 +/- 1 and on the GPS it reads 112-113.
65-67 and on the GPS it reads 60-61.

What i want to know is which is closer reality.

I measured these on fairly flat surfaces so the GPS hill inaccuracies may not count.

It's mainly out of curiosity i want to know.

Opti

Edit: Damn spelt speedometer wrong in title

TwoShedsJackson

The GPS is more accurate, the calibration on a lot of speedos can leave a bit to be desired and they frequently read high, as yours seems to do.

cantdecide

Do you have aftermarket big alloys fitted?


The discrepency between new and worn tyres is substancial also.

daedalus2097

Speedos have a tolerance which they are supposed to meed for their lifetime, and it's one-sided only, meaning that they can read your current speed, plus a margin of error, but never less than your current speed. So most manufacturers leave them a bit optimistic, in case with age they drop a little, and to allow for wheel differences etc. GPS is mathematically correct to within a few feet, so the speed reported there is a very accurate moving average over the last second or two.

peasant

The GPS measures actual distance travelled and time used doing so, so it is fairly accurate.

The speedo measures wheel or gearbox revolutions and then multiplies by fixed factor "X". As wheel circumference differs quite a bit with make, age and profile of tyre, "X" never really reflects reality plus there is an additional "error margin" calculated in.

Golferx

cantdecide wrote: »

Do you have aftermarket big alloys fitted?


The discrepency [sic] between new and worn tyres is substancial [sic] also.

If you take a 16 inch wheel, 6 inches wide, and 205/45 tyres. (Purely for calculation purposes.) And assume the difference between the diameter of a brand new tyre/wheel and a fully worn one is approximately 20mm.

This would translate into an approximate difference in circumference of 1.8% between a brand new and fully worn tyre.

optiplexgx270

Its on stock Ford 16" alloys.

I figured the GPS was more accurate but just wanted to be sure. I'll use if from now on to guage my speed + its on the windscreen so is almost easier (not really i guess )t o flick the eyes over and read the digital readout.

Eoin

Here's a handy site for measuring the difference that new wheels make.

http://www.miata.net/garage/tirecalc.html

murphyp2

Yeah - but if it's your mother's Focus and you have a provisional license, you shouldn't really be at 120kph in the first place...

You should also subtract 1.4% tolerance for being a girly-girl.

BTW, IMHO, etc.

colm_mcm

murphyp2 wrote: »

if .. you have a provisional license, you shouldn't really be at 120kph in the first place...

Are learners allowed to do 120 in a 120 zone that's not a motorway?

JHMEG

colm_mcm wrote: »

Are learners allowed to do 120 in a 120 zone that's not a motorway?

Ooo, got on a minor technicality there! AFAIK there is only one stretch of road in the country that is not motorway that has a 120km/h limit?

Biro

colm_mcm wrote: »

Are learners allowed to do 120 in a 120 zone that's not a motorway?

Such a zone doesn't exist!
Motorway is the only 120kph zone. Learners not permitted to drive on it.

GPS is the accurate one. Wheels will throw your speedo off, but it'll be off anyway as they're supposed to over read.

littlejukka

Biro wrote: »

Such a zone doesn't exist!
Motorway is the only 120kph zone.

the N2 to ashbourne is a national route and the limit for a large portion of it from the m50 to ashbourne is 120kph.

cantdecide

cantdecide wrote: »

Do you have aftermarket big alloys fitted?


The discrepency between new and worn tyres is substancial also.

My mk1 focus (bog standard) over reported on the analogue dial by up to 10 mph at speeds around 70mph. It gets progressivly worse as you head for 70.

Interestingly enough in test mode the digital speedo displayed the speed nearly spot on.

optiplexgx270

Its a '99

what_car

optiplexgx270 wrote: »

Ok so a Ford Focus and a Garmin 370...

On the speedometer it reads:

120 +/- 1 and on the GPS it reads 112-113.
65-67 and on the GPS it reads 60-61.

What i want to know is which is closer reality.

I measured these on fairly flat surfaces so the GPS hill inaccuracies may not count.

It's mainly out of curiosity i want to know.

Opti

Edit: Damn spelt speedometer wrong in title

on my accord auto with 17 inch wheels, driving at a constant 30kph on speedo, garmin gps reads, 27kph, its a garmin gps with the ultra sensitive antenna, and is fairly accurate, the gpsmap 60csx.

macnab

My GPS shows that my car speedo is correct at 40mph. But at at 60mph on the GPS the car speedo shows 67mph.

pete4130

Considering GPS in a car is only 1 receiver, and not receiving correction calculations from a base (another stationary GPS unit), and being shielded in a tin box travelling at differing speeds which has to encounter multipath signals (signlas being bounced off buildings and shielded by surrounding buildings/trees) coupled with the unknown satelite geometry overhead (evenly, well spaced out satelites produce more accurate results than those grouped together or in a line). Take into account the GPS unit takes readings at certain time intervals (so a reduction and increase in speed can't be taken into account so it gets averaged out) and the fact that handheld GPS receivers are accurate to approx 5m with no interference from multipath signals and having very good geometry I wouldn't put more faith in it than a speedo. Also GPS is up to 3 times less accurate for heights than it is for eastings and northings so ascending or descending might influence how fast it says you are travelling.
Take into account too that proper GPS kits for mapping & surveying that can measure to mm accuracy cost the guts of €40k, how accurate is the GPS in your car for €300?

I'd personally go with my speedo over the GPS anyday.

Biro

The map doesn't matter to the GPS. I have a GPS unit for plotting coordinates, it has no map, so it's just a receiver. Like them all it's accurate to within 5m or something like that. If you hold a car at a steady 60mph on a level road for a good few seconds, then the GPS even if it's out by 5 metres at the start and at the end of your run will still give a fairly accurate distance travelled and therefore mathematics are giving you the exact speed. It can't be more than 1mph out at 60mph. 5 metre inaccuracy's over maybe 500 metres travelled is fairly neglegible. Your theory is wrong in this count.
In any case, ask your local Guardi about what they see on their speed dectector. It'll always show the car going a bit slower than what the driver thinks they're doing. The GPS does the same.

Alun

True. The variations in the factors affecting positional accuracy, such as satellite geometry and upper atmospheric conditions, over a small distance are negligible, thus making them cancel each other out in two positional fixes only 1 second apart. Add to that that GPS modules (i.e. the actual electronics of the receiver, not the computer that does all the fancy display stuff) also use doppler information in their determination of speed, and most GPS modules will offer a speed accuracy of between 0.1 and 0.2 m/s.

unkel

Alun wrote: »

most GPS modules will offer a speed accuracy of between 0.1 and 0.2 m/s.

I knew they were accurate but I've never seen figures as to how accurate. Thanks for sharing that, Alun

optiplexgx270

From the manual of my Nuvi 370:

Update Rate:
1/second,continuous

GPS Accuracy:
Position: < 10 meters, typical
Velocity: 0.05 meter/sec RMS

GPS (WAAS) Accuracy:
Position: < 5 meters, typical
Velocity: 0.05 meter/sec RMS

http://www8.garmin.com/manuals/1184_OwnersManual.pdf

milltown

Just checked my speedo with the help of a mate's GPS. I was hopeful of being granted an extra 10 miles or kilometres an hour over my indicated speed, but no joy. My speedo only over reads by 2 or 3 mph at various speeds from 25 up to 90 mph*.

A lot more accurate than I had expected and, he reckoned, the most accurate he's ever seen from a car.

*Test conducted on a private test track using a professional driver.