How accurate is GPS Speed on iQue?

[si4xp]

I find the speed displayed on iQue with WAAS enabled is about 5 km/h more than that shown on the car speedometer.

(1) What is experience of other members?
(2) Does anyone know what is the accuracy of GPS speed on iQue with and without WAAS enabled?

[Holgado]

Maybe your car speedometer is not accurate. I have tried both speeds on my car and are quite the same, no significant difference.

[alexcue]

My money would be on the iQue (or pretty much any modern GPS). I think, if i can remember correctly the accuracy is up to .1 mph. Most speedometer are off, some like motorcycle ones can be really off. Tires (size etc) and wear all have an effect on the car speedometer.

Now this is from the Garmin site:

Velocity: 0.05 meter/sec steady state

I'll let the mathemeticians in the bunch analyze what exactly that means.

[si4xp]

Originally posted by alexcue
I think, if i can remember correctly the accuracy is up to .1 mph.
Now this is from the Garmin site: Velocity: 0.05 meter/sec steady state

0.1 mph accuracy translates to 0.05 metres/sec (0.0447 m/sec to be exact).

May be I should look at my car, especially since I may driving faster than what the speedometer says, although it is not serious enough for me get speeding ticket.

[Ken in Regina]

A friend and I have seen this discrepancy in testing six vehicles in our respective families. It has been pretty consistent. What we have seen is that the odometers/tripmeters are generally in agreement with our Garmins (and paper maps) but the speedometers are consistently about 5% optimistic, eg. the speedometers read about 5% faster than the Garmins report the vehicles to be going.

This is verified on a couple of the vehicles by using cruise control and a stopwatch in one of those speedometer test zones they have on some stretches of highway with the signs up every kilometer for a stretch of at least five kilometers.

The manufacturers of the vehicles are Honda (3), Nissan (1), Ford (1) and GM (1). The Garmins are my iQue and my friend's GPS V.

When I put new tires on my 2000 CR/V last summer, the new ones were just slightly less than 7% larger diameter, as calculated from the specs published on the tire manufacturers' web sites. Both of our Garmins now report the speedometer to be just about dead on and the odometer and tripmeter are just over 5% pessimistic (says I didn't go as far as the Garmins and paper maps say I've gone).

We did not have WAAS enabled for any of this. And none of this was terribly scientific .. simply curiousity and some simple testing, more as an excuse to play with the toys. :) That's why I say "about" 5% difference.

Point is, there appears to be a consistent optimism built into these different makes and models. My friend and I are inveterate cynics and we figure the manufacturers are intentionally building in a bit of a margin to avoid liability suits from folks caught speeding.

[rjack22]

My speedometer is exactly the same as my iQue.

I would believe the speed on my iQue over that on the car.

[donc13]

Originally posted by si4xp
I find the speed displayed on iQue with WAAS enabled is about 5 km/h more than that shown on the car speedometer.

(1) What is experience of other members?
(2) Does anyone know what is the accuracy of GPS speed on iQue with and without WAAS enabled?

I've checked my speedometer against radar (I'm a retired cop, piece of cake to get the locals to clock me) and checked the iQue against my speedometer. The iQue and my speedometer in my truck are always in agreement, and correct as tested on radar. In my RV, the speedometer reads high [I'm not going as fast as it says I am], the iQue (as expected) is correct.

In other words, the iQue is accurate, speedometers vary.

[amw157]

Most people do not know this, but car manufacturers who sell cars in the US are required by law to "fudge" the speedometers by some fixed percentage. I do not know the exact figure, but I know it is a mandate (just like airbags, ABS, and OBDII).

I will try to find a link that proves this.

Andy

[amw157]

This doesn't exactly prove my point but it does produce some interesting fodder.

Car and Driver (http://www.caranddriver.com/article.asp?section_id=4&article_id=1906&page_number=1)

Andy

[doclaw64]

Originally posted by amw157
[B]Most people do not know this, but car manufacturers who sell cars in the US are required by law to "fudge" the speedometers by some fixed percentage. I do not know the exact figure, but I know it is a mandate (just like airbags, ABS, and OBDII).



I did not know that the US does that but I do know that german cars are set that way from the factory. It is about a 4-5% difference. The speedometers are optomistic. ;)

[rwsmith123]

99.999% of the time the iQue speed will be more accurate than the speedo, one exception for me was the other day when the iQue said I was going 1554 miles per hour! What's that just under mach 3?

I have a '96 Saturn and the speedo isn't fudged, it matches the gps.

[kefer]

My BMW speedometer is off by almost exactly 5 mph - which as others have said is by design (really annoying, too). If I check the average speed computer in the car by resetting it (which will give an "exact" speed readout momentarily) it is in exact agreement with the iQue, and about 5 mph less than the speedometer reading.

So yes, I believe the iQue is very accurate in most cases, whereas many cars are not (however, if your speedometer is reading slower than the iQue, something probably is wrong).

[Robertp366]

In the auto world it's fairly widely known that manufacturers set their speedometers to read a bit faster than true speed. But according to a very senior engineering executive I know at GM, its not a federal mandate (like airbags, etc) but a legal issue steming from a few lawsuits back in the 60's.

For most of today's cars, the speedometer is run by the vehicle computer, and can be re-programmed. This is essentially standard on US trucks because of the wide variety of tire sized in the 'standard options' lists (has to do with GVWR). As a result, there are many aftermarket chip programmers that can modify the speedometer accuracy. Also, any good speedometer shop can adjust yours - though finding a speedo shop is a lot harder than it was in the 60's to 80's. I just asked the maintainence manager at the local Highway Patrol Office where they have their speedometers callibarated.

And I agree with most everybody that the GPS speed is right on.

Robert

[jonasolof]

There was a UN economical commission directive #39 that has been ratified by many countries on this, including the EU. The directive regulates a lot of parameters around the measurement of speed in vehicles.


The indicated speed should never be lower than the actual speed.
The following relationship is to be upheld between the speed indicated by the speedometer (v1) abd the real speed:

0<_(v1-v2)<_v2 + 4 km/h

Tests should be done at 40, 80 and 120 km/h


(My translation from the swedish text)

Thus, never under but at the most ten percent + 4 km/h over.

This ruling applies to tests for the the general mutually accepted "type certificate" for standard cars equipped with the tyres recommended by the manufacturer. Winter tyres do not constitute normal equipment.

[GPSfan]

Car speedometers are required by law to never show less than real speed. Since the wear of tyres affects the speedometer, car manufacturers intentionally makes the speedometers to show 5-10% more than real speed to be on the sure side.

On my car, I can by entering a key-combination, access a diagnostic mode. In this mode the speed measured by the cars motion sensors under the car (this speed is used by Stability Control) is displayed on the car computer and this is 100% acurate. It is also exactly the same as the Ique shows.

[donc13]

Originally posted by amw157
Most people do not know this, but car manufacturers who sell cars in the US are required by law to "fudge" the speedometers by some fixed percentage. I do not know the exact figure, but I know it is a mandate (just like airbags, ABS, and OBDII).

I will try to find a link that proves this.

Andy

Actually, that's not true. They are required to be within plus/minus 5% on speed and odometer.

[Q-Eye]

Precision is relative. Even the best GPS can be slightly off if it is calculating speed in 2D mode when the terrain is not level. For instance, if you are traveling down a 6% grade, your actual speed will be 0.179% faster than the calculation that doesn't take the elevation change into account. So if Betty thinks you're going 100kph down that hill, you're more likely going 100.179kph. I doubt that the police will notice ;)

[blasterblaster5]

My tires are showed 30% wear. My 2002 TL is always shows 3Km less than the Ique. but the off the topic question is what do the manufactures gain from this?

[rocket666]

My tires are farily new, and my speedometer consistently shows 5 mph less on level ground freeway speed. All this means is I use qBuddy all the time and travel 5 mph faster now:D

edit: I meant shows 5 mph more on my car's speedometer.

[LeRoux Strydom]

I used Heidi to calibrate the speedo on my old Porsche via a simple potentiometer adjustment. I don't care about the odo accuracy on this car. At 240000 km it no longer matters. :rolleyes:

The speedo on my BMW is consistently 7km/h fast over the entire speed range compared with Heidi. Yes, I tested up to 225km/h, but shhhh...... I use this fact to adjust my driving speeds accordingly. The on-board computer's speed readout is only 1km/h faster than Heidi, so I concur with the opinion that the speedo discrepancy is a programmatical "error" intentionally introduced, but that modern cars' electronics are capable of very accurate speed measurements provided that you do not change the tyres from the specified size.

I guess you could say that I believe Heidi's speed measurements under optimal conditions over any car's speedo.

LeRoux

[apersson850]

When I tow a trailer after the car, the difference between the iQue speed and the speedometer increases somewhat, as can be expected.
If there is a difference, it increases more on the front wheel drive than on the RWD, which also could be expected.

Both cars use pulses from the driven axis for the speedometer. The FWD calculates an average of the two ABS sensors in the front. The RWD has a single ABS sensor, before the differential gear, in the rear axis. So that's an average computed mechanically, so to speak.

In Jonas' reply above, he got the formula somewhat distorted. It should look like
0<=(v1-v2)<=0.1*v2+4 km/h

v1 is indicated speed, v2 the real speed.

When testing produced vehicles, to determine compliance, the constant offset is 4, 6 or 8 km/h, depending upon the type of vehicle. But when getting an approval, it's 4 km/h for all types.
So in some vehicles, coming off the assembly line, a deviation of 8 km/h + 10% is considered OK.

[Wiggers]

I'm not sure how true this is, but a Belgian friend told me it is quite acceptable to drive at an indicated 10% above the speed limit there. Maybe this is to allow for the calibration of speedometers to read fast? I also recall hearing that British speed cameras are often set to 10% + 2mph above the signed limit, but I wouldn't like to test it! Probably varies from county to county and month to month anyway, depending on how the budget is going... ;)

[amw157]

Originally posted by donc13
Actually, that's not true. They are required to be within plus/minus 5% on speed and odometer.

I stand corrected. You are right. My sources have evaporated! Sorry for any confusion.

Andy

[apersson850]

To avoid any confusion, at least to new readers of this thread, it has been proven by links provided above, that this is not the case in Europe. Here they have to indicate at least the actual speed, never below.

[djeady]

My Garmin indicates that both of my Volvos (2002 XC70, 86 740) are doing 112/113 when the speedo reads 120.

Oddly enough my 99 Honda Goldwing is bang on - they have a reputation for inaccurate speedos.

[apersson850]

If I drive my Volvo 850 Turbo, 1994, at an indicated speed of 100 km/h, the iQue says it's close to 97 km/h.
We tried with a colleauge's Volvo V70 2.4T, 2000, today. When driving 100 km/h, the iQue said 89 km/h.

So newer isn't necessarily better in this case.
Besides, his was slower. :D We were trying it out with TimeTrial.

[albertan]

Our 2 yr old Kia van with tire pressures at spec reads 4-5 kph greater than Ella (short for Computerella, as christened by my daughter) across the speedometer. My 6 yr-old Ford Contour (Mondeo) speedo reads within 0.2 kph of Ella on stock tires. This is odd, as I would have expected the car speedo to report higher velocities as the tires wear out and become smaller. Presumably on new spec tires the Ford would under-read Ella.

[0gopogo]

When I started driving mechanical speedometers where the norm. As said earlier here, those could not and were not expected to be accurate. By law and experience we expected them to lie at a rate of about 10% at 100km/h; probably less if slower, maybe more if faster. We also expected them to show a higher than the actual speed, but never less, for obvious reasons.

So we used stopwatches to determine the true speed and the "lie" factor. That was in the 80s. A 5% error @ 100 km/h was pretty cool, the lower, the better, and surely, the speedometers were never slow.

After buying the iQue a year ago and using it (in the US, only) on two rental cars and my own car, too, I am amazed to observe that either the modern speedometers and the iQue measure speed quite accurately, or with the same trend of error. Anyway, they agree within 1-2 miles/h.

Back in the 80s and 90s I considered driving at 10% over the speed limit (as measured by speedometer) as legally safe. 20% over speedometer up to 130 km/h were believed to be still OK to get away with.

Too bad I am still applying that 20% rule today. Boy, have I been lucky...

20% in a 35 miles/h speed limit means you I could go 42, and would be reasonably safe. (There's a magic number, courtesy of Douglas Adams. Now add to that 10% because your speedometer still lies. The speed must must be right, after all many Houstonians do it.)

;)

[JohnVa]

One thing I think we need to consider is "steady state" vs changing state. Their is a time lag on gps displays. I think the biggest place I have noticed this is in altitude display where the display is considerable behind while driving in the car. This was really apparent when cresting mountains where we were trying to determine the highest point while driving. We would be past the crest by several hundred yard before the dispaly would catch up. If it's true there why not also in the speed indication?

JohnVa

[rkevwill]

Since everyone is piping in about this topic, why not me. (especially since it has always bugged me on my cars) I have noticed in my old bmw, that I have had since 1984, that it has always been optimistic. Interestingly, as one poster noted, my computer when reset on cruise, gives me quite an accurate speed. Normally the speedo is anywhere from 5-10% optimistic. Worse, my 2001 Porsche, (which I recently sold) was easily 10% optimistic at around 60. My wife's Expedition, is nearly dead on, and her old 77 Mercedes 450sl is nearly dead on. All these are compared to the Ique, which I assume to be quite accurate.

The explanations about the European laws make this much clearer for me. As I said, this has ALWAYS bugged the heck out of me.

And here I thought that BMW and Porsche just wanted you to think you were going faster, so you wouldn't speed so much!

[CougerofEQ]

If you are going the speed limit per you speedo and get a ticket you could sue could you not?

Tire and auto warrenties are also mileage based so again, they would get sued if they did anything to shorten the warrenty time - best to show a slower speed(less miles) than to get bad PR or a lawsuit.

Its just the world we live in - cant blame the car manufactuers for protecting themselves.

[apersson850]

That's not the world we live in. That's just the small part of the world a few trigger-happy want-to-sue-everyone-just-because-we-can Americans live in.
Here we buy ourselves an iQue (see, I'm on topic!), so that we can find just any obscure town. We don't sue the community for being placed in such an obscure, almost impossible to find, location.

Now that was an idea for you Americans, right? :D

[paul6347]

Since it's up to 3 pages, I'll post. In my 1997 Cadillac Deville, with my old warn stock tires, the spedo was dead on. Now since I changed the tires with new thread, the spedo is UNDER 3 mph at 65. On my 2002 Dodge Stratus ES with 20,000 miles, the spedo is dead on.

My experience with older cars with mechanical spedo's, I found that with higher milage cars (over 120k) the spedo would wear (the friction coupling inside which keeps the needle up to indicate speed), and actually show lower speeds. I got a ticket for that one time. And the judge doesn't buy the wearing speedometer story. So if you have an older car, it's a very good idea to compare the speedometer reading with your iQue (back on topic!)

Paul

[Monkey Hanger]

My Mercedes speedo consistently reads about 5% faster than my iQUE. I believe that car manufacturers will err on the side of safety and calibrate a speed to read high and not low!
Otherwise manufacturers would be held liable for speeding tickets! :D

Another point is "Does the GPS system take into account accurately the distance increase caused by going up/down hills?" This would also have the effect of showing a higher speed!
In this country there isn't a level road anywhere!

[si4xp]

Originally posted by paul6347
My experience with older cars with mechanical spedo's, I found that with higher milage cars (over 120k) the spedo would wear (the friction coupling inside which keeps the needle up to indicate speed), and actually show lower speeds. I got a ticket for that one time. And the judge doesn't buy the wearing speedometer story. So if you have an older car, it's a very good idea to compare the speedometer reading with your iQue (back on topic!)

Paul

My car, the Ford Taurus, has more than 240k Kms on it, and yesterday when checking the speed again I found it to be showing the speed lower by 7 kph (5 mph) compared to the GPS. Therefore, at 70 mph on speedo, the car would actually be doing 75 mph, which could be a borderline for getting a ticket at some stretches in US.

[donc13]

Originally posted by Monkey Hanger
My Mercedes speedo consistently reads about 5% faster than my iQUE. I believe that car manufacturers will err on the side of safety and calibrate a speed to read high and not low!
Otherwise manufacturers would be held liable for speeding tickets! :D

Another point is "Does the GPS system take into account accurately the distance increase caused by going up/down hills?" This would also have the effect of showing a higher speed!
In this country there isn't a level road anywhere!

Unless you've got some really steep hills you drive on, the difference in speed (when assuming you're on level ground) is insignificant.

[donc13]

Originally posted by CougerofEQ
If you are going the speed limit per you speedo and get a ticket you could sue could you not?

Tire and auto warrenties are also mileage based so again, they would get sued if they did anything to shorten the warrenty time - best to show a slower speed(less miles) than to get bad PR or a lawsuit.

Its just the world we live in - cant blame the car manufactuers for protecting themselves.

No, you couldn't sue (well..you could sue, but you'd lose). Speedometers are guides, not calibrated devices. It is up to the driver to determine the variation from normal and adjust accordingly. Besides which, only a very tiny percentage of speeding tickets are written for less than 10mph over the limit.

And if you speedometer show you higher than your actual speed, the same would happen to your mileage shown...it too would be high. So the idea of being able to have mileage show low to insure you got your full warranty vs. showing speed high to insure you didn't get tickets are in opposition to each other.

And yes...way to many people here (America) are willing to sue for anything, and way too many lawyers are willing to take those people as clients.

'Tis a sad state of affairs for the civil legal system.

[Monkey Hanger]

Originally posted by donc13
Unless you've got some really steep hills you drive on, the difference in speed (when assuming you're on level ground) is insignificant.

The point is that all roads have an incline, no matter how small (and even a 5% incline is small) and that whatever the incline is it will add to the apparent discrepancy between the car and the iQUE!

[Monkey Hanger]

Originally posted by donc13
No, you couldn't sue (well..you could sue, but you'd lose). Speedometers are guides, not calibrated devices. It is up to the driver to determine the variation from normal and adjust accordingly. Besides which, only a very tiny percentage of speeding tickets are written for less than 10mph over the limit.


Not quite true. Manufacturers have to be able to prove that their devices conform to a given standard for the country in which they operate. Well, they do here!
I believe our calibration is within minus 5%. No plusses are allowed. i.e the speedo must operate within no more than 5% fast.

[apersson850]

Monkey, if you go back a bit in this thread, you'll see references to the applicable directive, valid within the EU (of which UK is a part, like it or not). Then you can see that the 5% you refer to aren't dead on target.

Regarding the difference when driving on hilly roads, it's neglectible on most surfaces. At a 10% incline, the horizontal speed is still 99.5% of the speed your speedometer will show.
10% is, in most areas, a significant slope.

For diehards, a 20% slope yields 98.1%, a 30% slope 95.8%.
If it's worse that that, you'll hardly get any speeding ticket at all, since you are probably pressing the brake pedal as hard as you can... ;)
I've driven on 30% slopes, in Cornwall. If feels like going over the edge, literally, when you can't see the road over the hood of the car. So I know they exist, but they are rare.

[donc13]

Originally posted by Monkey Hanger
Not quite true. Manufacturers have to be able to prove that their devices conform to a given standard for the country in which they operate. Well, they do here!
I believe our calibration is within minus 5%. No plusses are allowed. i.e the speedo must operate within no more than 5% fast.

Manufacturer's certify their vehicles meet applicable standards, they do that by testing represenative samples and applying the results to the entire line. They do not test each and every vehicle. They don't have to prove each device conforms, only that they have designed and tested to insure conformity.

That means that any individual speedo can be anywhere from way off, to dead on. They are after all, mechanical and electronic devices subject to manufacturing variations and faults. It is up to the driver and/or owner to determine how accurate the gagues are and to take whatever corrective action is necessary. If that means having the manufacturer bring the device back into complicance under a warranty provision...fine, they'll do it. But it also means that the driver is the one who is ultimately responsible for operating the vehicle within applicable limits.

Bottom line...suing a manufacturer because you got a speeding ticket based on a defective speedometer will get you exactly nowhere...other than a bit less money in your pocket after you pay the lawyers fees.

[doclaw64]

Originally posted by donc13


And if you speedometer show you higher than your actual speed, the same would happen to your mileage shown...it too would be high. So the idea of being able to have mileage show low to insure you got your full warranty vs. showing speed high to insure you didn't get tickets are in opposition to each other.


Actually, it depends on the manufacturer. My porsche is dead on for the odometer but the speedometer is 4% too high. I thought the odometer would be high as well but it is not.

[Ken in Regina]

In the six vehicles that my friend and I have checked out, the speedometer was always optimistic (speedometer reporting faster than the actual speed as reported by our GPS units) and the odometers were pretty much dead on.

In the case where I installed new tires on my 2000 Honda CR/V which are about 7% larger diameter, according to the respective tire manufacturers' web sites, the speedometer changed to pretty much dead on and the odometer went to understating the distance travelled by about 6%.

...ken...

[apersson850]

In the directive, valid within the EU, is clearly stated the tolerances allowed on vehicles used for certification, as well as the tolerances allowed on any vehicle out of the assembly line. The former is more stringent, of course, but both are defined.

Also, there is no need for the odometer to be optimistic just because the speedometer is. These are two, admittedly related, but separate components.