How Accurate Is The Miles Per Hour Function On A Gps?

[RexBloodman]

For the inquiring minds out there I am wondering how accurate a gps is according to how fast you are going?

 

Thanks for any help in this area.

[MouseFart]

I was fiddling with mine yesterday and showed 1mph slower than what my speedometer was showing. Which was right I don't know, but thats pretty close in my books.

[+ccindyybubb]

I would say "very" accurate. Much more so than your car, motorcycle or scooter.

[Neo_Geo]

GPSr's are MUCH more accurate (when you're getting a good lock). Speedometers typically show speeds that are faster than the true speed. I have an old Mercury ('93) that registers pretty well up to about the 55 - 60 MPH range, but then it gets progressively worse the faster you go. It'll show 70 when you're only doing 65, and 80 when you're doing 73.

 

I've gotten a few speeding tickets in my younger days, and I was always surprised that the speed written on the ticket was significantly LESS that what I thought I was doing.

[samsulli]

GPS is much more accurate than auto speedos as they(auto speedos) will almost always read high as that shortens the warranty period.

[+Ed & Julie]

My GPSr (Garmin Legend) is dead on with the calibrated speedometer in my patrol car.

 

Ed

[+ILBandit]

Both my GPS units have shown about 2 MPH slower than my speedometer in both of my vehicles. I would like to compare it to a radar gun sometimes to see if there is any difference.

Although compared to the speedometer on my bike, it was dead on the same as it shows.

[+timk54]

I put taller tires on my Silverado. When the speedodometer shows 70. My GPSr is showing 74.. The GPSr is correct

Chevrolet can't recalibrate my speedo for size of my tires which are 285 75r 16's. It's nice to know my real speed.

By the way that means I'm getting more actual miles covered by my warranty. Stick it to the Man.

timk54

[+shammond]

The speed on my GPSr usually shows 2-5 mph slower than my speedometer. I'd say the GPSr is probably right...

[+Renegade Knight]

Yup, the GPS is right, the cars are normally wrong. If I set the cruise on my Corolla at 85, I'm doing 75. At 29 its 25 roughly. I keeps me safe in school zones.

[+dodetrexer]

While traveling back and forth to AZ from Cali, I play with my GPSr... It seems to be right on for me and my truck...

 

dodetrexer

[+Airmapper]

In an airplane knowing your actual speed is very handy, as the aircraft's speed indicator runs off air passing over the plane as you go through the air. This means the GPS can accurately tell you the wind speed and direction when compared to the planes on-board air sensors.

 

Air speed vs. Ground speed (GPS derived) and magnetic heading (From gyroscopic compass) vs. Actual heading (GPS derived) Take these 4 readings and you will know exactly how the wind is affecting the plane.

[+cqedens137]

the way i check mine is with the electronic speed limit sign near town.these signs use radar gun to let you know how fast you are going. the gpsr has always been dead on with the sign. the speedo in my jeep is way off.

 

cqedens137

[+Team Cotati]

The car and gps indications are always within one mph of each other. The car is a Buick. The gpsr is a Magellan. Thus they are most likely both wrong. We only know how fast we are traveling when stopped by a state trooper.

[+Team Cotati]

The car and gps indications are always within one mph of each other. The car is a Buick. The gpsr is a Magellan. Thus they are most likely both wrong. We only know how fast we are traveling when stopped by a state trooper.

[+HaLiJuSaPa]

GPSr's are MUCH more accurate (when you're getting a good lock). Speedometers typically show speeds that are faster than the true speed. I have an old Mercury ('93) that registers pretty well up to about the 55 - 60 MPH range, but then it gets progressively worse the faster you go. It'll show 70 when you're only doing 65, and 80 when you're doing 73.

 

I've gotten a few speeding tickets in my younger days, and I was always surprised that the speed written on the ticket was significantly LESS that what I thought I was doing.

 

I almost wonder if they do that purposely to cut down on high speeding, seems strange that speedometers would suddenly read high after 55 or 60

 

My one experience is I had my cruise control set at 65 for about 10 miles and the GPS always read between 64 and 66.

[Vlad]

gps receivers are typically accurate to .1 mph

 

car speedos read higher than actual for several reasons but one of them is so that you don't go after the manufacturer because you got ticketed for speeding when the speedo said you weren't...

[+Fred Flintstone]

40 mph and below my 60 csx is the same as the speedometer. Above 40, the mph is changes. GPS shows 60 and speedo shows 58. GPS shows 72.8, speedo shows 70. Chevy s-10. My Monte SS is correct up thru legal speed and some illegal speeds.

[+UKGeoCyclist]

I use my GPS on my Bicycle, and it is always within 0.5kmh of my bike speedo, which is very accurate.

 

Bicycle speedos are calibrated by measuring the distance travelled in 1 wheel revolution, and I re-calibrate mine each time I change tyres.

 

AK

[+Sputnik 57]

The manual for the Garmin 60C shows velocity accurage to .05 meters per second, which would be 180 meters per hour. That means it is accurate to +/- 0.1118 miles per hour. Your car spedometer probably varies more than that due to changes in tire pressure!

[WR8Y]

By all means, if you GPSr is locked well, it's more accurate than that stupid dial on your dash! The speedometer is going to be a bit off, it's largely mechanical, afterall. Manufacturers 'fudge' them towards the high side, probably to avoid your complaining to them; ".... hey I got a ticket for doing 68 in a 65, but my speedometer said 65... "

 

We calibrate police radars here at work (well, we do a LOT of stuff here, that's one of the things we do...).

 

I tried an experiment last month, I took my GPSr and jumped into a sheriff car and went for a ride. Even his 'calibrated' speedometer was off by 2 mph at 70mph, but the GPS and radar were displaying 70. (The radar only displays in whole numbers, my GPSr in tenths of a MPH, so I can only say they are within a few tenths of each other.) I used a Lowrance Expedition Color on the dash with a good lock (EPE was 20-24 feet with WAAS on.

 

Now, if you want to hear about a grossly inaccurate speedometer, let me tell you about my 1979 Yamaha RD350!!!!

[RexBloodman]

Thanks everyone that is what I was expecting to hear. The car speedo is faster in most cases than the GPSr.

[+GlobalRat]

I've consulted to a number of vehicle manufacturers. They accept a 10% variance on the speedo, in the negative of course. Typically you you can expect a 10km/h variance once you hit 130km/h.

[Henny Pa]

The GPS speedometer is very accurate. I've checked it with my patrol cars radar. The GPS is right on.

[+Runaround]

GPSr's are MUCH more accurate (when you're getting a good lock). Speedometers typically show speeds that are faster than the true speed. I have an old Mercury ('93) that registers pretty well up to about the 55 - 60 MPH range, but then it gets progressively worse the faster you go. It'll show 70 when you're only doing 65, and 80 when you're doing 73.

 

I've gotten a few speeding tickets in my younger days, and I was always surprised that the speed written on the ticket was significantly LESS that what I thought I was doing.

 

I almost wonder if they do that purposely to cut down on high speeding, seems strange that speedometers would suddenly read high after 55 or 60

 

My one experience is I had my cruise control set at 65 for about 10 miles and the GPS always read between 64 and 66.

 

In a previous job I bought parts from a company that made speedo's for the big three. One of their engineers said it was purposely done because the manufacturers want the opposite, a driver going faster than what the speedo shows. Customers tended to complain in that case, especially due to the tickets. When the customer thinks they are going faster than they really are, no one complains.

 

FWIW, my Subaru is off by 3-4 MPH. The car will show 70, the GPSr 66-67.

[+HooloovooUK]

Dunno about the US, but in the UK the legal requirement for a speedometer is -0% to +10%.

 

This means that the speedometer must NEVER under read (say that you are going slower than you are) but it is allowed to over read by up to 10%. So when doing 70mph your speedometer could say anything between 70mph and 77mph.

 

In order to comply with this law, ALL manufacturers add a fudge factor to the speedometer reading to ensure that it never under reads. I checked my speedometer with my GPS and I thought it was over reading by more than 10% so I took it back to the dealer to check it out.

 

They plugged the diagnostic laptop into the car and drove off down the road. At 50mph on the digital display on the dashboard, the diagnostic computer read 46mph. That means an over read of 8% which is within the law so they won't do anything about it.

 

But don't you find that interesting? The car knows EXACTLY how fast it's going and hence the odometer reads correctly. But they fudge the reading before showing it to the driver on the display.

 

Now I can understand why they do it. Obviously the measured reading is between say -0.5% to +0.5% so they have to add an offset to ensure the error is always above 0%..... but why they have to add such a huge value as 8% rather than just 2 or 3% I don't know.

Edited May 19, 2006 by HooloovooUK

[+JSWilson64]

OK, dumb question time. If my GPSr is only accurate to 30 feet, for example, how can it peg my speed within 1 mph at 60 mph?

 

I think the answer is that the "Accurate to: 30 feet" display is relative to the displayed coordinates. The GPSr says, "You are within a 30 foot radius of point X on the globe." But two positional measurements taken within a relatively short period of time will be VERY accurate with respect to distance and bearing from one point to the other. It's saying, "I'm VERY certain that point X is 88 feet from point Y, I'm just not exactly sure where on the globe either one of those points is."

 

Is that a reasonable explanation? Because I get asked this question by someone whenever we have the GPS in the car.

Edited May 19, 2006 by JSWilson64

[+JohnInDC]

That sounds about right to me. Another way to say it may be that the many sources of GPS error remain relatively constant from moment to moment. Or! If they don't, then the GPS software is programmed to ignore such errors below a certain level. In fact one or the other of those pretty much has to be true, otherwise whenever you were standing still with a margin of error of 30 feet, the GPS would have you hopping around inside that circle like a bug in a skillet.

[WR8Y]

OK, dumb question time. If my GPSr is only accurate to 30 feet, for example, how can it peg my speed within 1 mph at 60 mph?

 

I think the answer is that the "Accurate to: 30 feet" display is relative to the displayed coordinates. The GPSr says, "You are within a 30 foot radius of point X on the globe." But two positional measurements taken within a relatively short period of time will be VERY accurate with respect to distance and bearing from one point to the other. It's saying, "I'm VERY certain that point X is 88 feet from point Y, I'm just not exactly sure where on the globe either one of those points is."

 

Is that a reasonable explanation? Because I get asked this question by someone whenever we have the GPS in the car.

 

I think ya got it. For example, I walked a 3 mile path a few weekends ago and set a waypoint on the side of a hill at a trail mile-marker. My EPE displayed was in the 30-50 foot range. On the way back, as I walked by the mile marker post, I went to .02 mile display. (That's only 100 feet from one side of the screen to the other.) The post appeared to be RIGHT where I 'set' it! (Certainly within a foot or two.)

 

But, anyway, you explained it well.

[+Renegade Knight]

OK, dumb question time. If my GPSr is only accurate to 30 feet, for example, how can it peg my speed within 1 mph at 60 mph?

 

I think the answer is that the "Accurate to: 30 feet" display is relative to the displayed coordinates. The GPSr says, "You are within a 30 foot radius of point X on the globe." But two positional measurements taken within a relatively short period of time will be VERY accurate with respect to distance and bearing from one point to the other. It's saying, "I'm VERY certain that point X is 88 feet from point Y, I'm just not exactly sure where on the globe either one of those points is."

 

Is that a reasonable explanation? Because I get asked this question by someone whenever we have the GPS in the car.

 

Thats pretty much what's going on. The atmospheric conditions (and other things) that cause the Error in a GPS can vary a lot over time. But during the 1 second or less for a GPS update cycle, not so much. So the error is exactly as you have reasoned out. Plus your MPH is an average over time (or a certain number of points) and the averageing will tend to reduce any random error.

 

WAAS and other means of differentially correcting a GPS work on that first part. The ground station knows exactly where on the globe it is, and it knows exactly where it's calculating it is. It compairs the two and computes the differential correction to use with the info your GPS is recieving. Then your GPS knows better where on the globe it is.

[peter]

I think the answer is that the "Accurate to: 30 feet" display is relative to the displayed coordinates. The GPSr says, "You are within a 30 foot radius of point X on the globe." But two positional measurements taken within a relatively short period of time will be VERY accurate with respect to distance and bearing from one point to the other. It's saying, "I'm VERY certain that point X is 88 feet from point Y, I'm just not exactly sure where on the globe either one of those points is."

 

Is that a reasonable explanation?

Partly, but two additional factors should be noted. First is that GPS receivers directly measure velocities relative to the satellites by observing the Doppler shifts of the radio signals. So in addition to the 'pseudo-range' information based on travel time differences and used to calculate the position, the receiver uses 'pseudo-range-rate' data based on the Doppler shifts to determine speed and track bearing. The second consideration is that there's some filtering of the speed data. It is very accurate when you're moving at a constant speed, but you can usually notice a lag of a couple seconds if you make rapid speed changes. Slam on the brakes at a red light and you may notice that the GPS speed takes a few seconds to drop to 0. This filtering trades off quick responsiveness to get better accuracy.

[+SwampYankee]

We run drag boats.....Our speed is done with GPS......