Jump to content

Your GPS versus your Speedometer - need your input.


GOT GPS?

Recommended Posts

My 96' Escort Wagon, according to my GPS goes up to 4 to 5 miles per hour faster than what the speedometer says. I was traveling exactly 70mph(Michigan speed limit), and a motorcycle cop put on his siren, and telling me to knock down my speed, so it makes me wonder if the gps should be used while driving??

 

Question:

- What does your GPS say, when you are driving 40mph or 60mph. What make or model vehicle are you driving and how does the GPS compare with your Speedometer??[/b]

 

Convert kilometers/hr to mph by multiplying by 0.62137

1 mile = 1.609344 kilometers

http://www.google.com/help/features.html#calculator

 

----------------------------------------------------------

Pictures of the insides of my Old GPS V

My Home Page about what is GPS

Strange

Link to comment

quote:
Originally posted by GOT GPS?:

My 96' Escort Wagon, according to my GPS goes up to 4 to 5 miles per hour faster than what the speedometer says. I was traveling exactly 70mph(Michigan speed limit), and a motorcycle cop put on his siren, and telling me to knock down my speed, so it makes me wonder if the gps should be used while driving??

 

_Question:_

- What does your GPS say, when you are driving 40mph or 60mph. What make or model vehicle are you driving and how does the GPS compare with your Speedometer??[/b]


 

One of the local police departments puts out one of those roadside Radar units quite often (the ones which tell you to "Watch Your Speed" and flash the speed of approaching vehicles.

 

I've found the radar speed and the speed indicated by my GPSV rarely match -- usually differing + or - 5 mph.

So personally I would never rely on my GPSr as a speedometer.

 

I wouldn't expect it to be totally accurate anyway. If I'm standing still and average a waypoint the difference in accuracy can range anywhere from 12 to 28 feet in just a few seconds. I can't imagine what the variance in accuracy would be moment to moment while traveling at 60 mph. I would think that would affect the GPSr's speedometer.

 

Jolly R. Blackburn

http://kenzerco.com

"I'd like to buy the world a coke, but I only have fifty cents"

Link to comment

GPS ground speed is accurate to within a tenth, we use it in the aircraft all the time and it jives with the radar azimuth speed check as well as DME (crude) checks within a couple MPH, but that is a couple mph at over 400 knots..lol, but the reason you may see innacuracy is tree cover can delay the speed update and it is poor for measureing acceleration, only a perfect constant speed, which it does very well.

Link to comment

Automobile speedometers tend to be off a bit, especially in older vehicles. It's been my personal observation that my old car was 10 km/h fast and my newer car is about 8 km/h fast. This means that when I think I'm doing 100 km/h by the speedometer, I'm really doing 92. I have used my GPS and observation of other traffic to confirm this. I would have to say that I trust my GPS as a speedometer more then the mechanical, speedometer that can also be off if your tires are not exactly the right size too.

 

Rob

Mobile Cache Command

Link to comment

Range rate error (that will affect speed type calcs) even in the days of Selective Availability was really very stable at approx 1km per hour. Basically a "non moving" GPS would on average move 24km in 24 hours.

 

Without SA range rate errors are really very stable and big jumps in subsequent positions doesn't really occur unless there's specific interference or obstruction issues.

 

Without SA range rate error is in the centimetres per second bracket and of little impact on GPS speed.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif

Link to comment

quote:
Originally posted by Jolly B Good:

I wouldn't expect it to be totally accurate anyway.


 

In this case, your expectations would be incorrect. The speed indicated by the GPS should be much more accurate than the speed indicated by your car's speedometer.

 

Your comments about the position moving around don't apply, since the GPS unit does not determine speed from successive positions.

Link to comment

I find the GPS speed to be very accurate. I would absolutely believe it to be accurate within 0.1 mph presuming no signifigant interference (overpasses, trees, skyscrapers) I find that my Dodge truck speedo indicates about 1 mph faster than GPS at any speed above 20, and my BMW motorcycle reads 60 when the GPS shows 55. The BMW bikes are known for having "optimistic" speedometers, and in fact the specification from the factory allow up to a 10% error on the "optimistic" side, plus a static error of a few mph!

Link to comment

I checked the speedo on my wifes 1.8 litre 'Family Saloon' against the reading on my Legend. At a speedo reading of 30mph the Legend was showing 25mph. The difference gradually decreased as the speed increased until they were reading exactly the same at 90mph. The trend continued and at a speedo reading of 110mph, the Legend was showing 114mph.

 

John

And should you ask, am I at ease there? I'd answer "Yes. Oh yes indeed".

For my heart it dwells in lonely places, where springs leap down, where ravens feed.

Link to comment

GOT GPS?,

Based on your story, your GPS and the cop both agreed and that your speedometer is the odd man out.

 

Trust the GPS, but only after you get a good lock.

 

________________________________________________

 

Garmin eTrex Vista, Legend, and GPSmap 162 with Bluecharts/Fishing Hotspots/POI/Road & Rec

 

Ducks - flying geocaches of meat

Link to comment

There's a little experiment you can do to verify the best measure: drive along on the interstate at 60 mph, first by car speedometer and then by GPSr. Time the interval by the mile markers that are typically posted along the way. Whichever one sets your speed such that you are closest to 60 seconds to cover the mile is the best measure.

 

My 94 Summit is really bad on this, overreporting speed and distance covered by about 5%. I have found the GPSr to be pretty good, accurately predicting the next milemarker even after 10-15 miles of pacing.

 

Max

Often wrong but seldom in doubt

Link to comment

My GPSV is always on the dash, and when I'm not headed for a cache, it stays in the "big 2 fields" mode with speed displayed. The GPS is far, far more accurate than most car speedometers, even under poor signal conditions, and I have found my car speedometer reports 65 MPH when I'm really doing 62 or so.

 

--

Scott Johnson (ScottJ)

Link to comment

I googled this...

 

<<When we design vehicle instrumentation, there is a deliberate intent to ensure that the spedometer reads more than it should. EEC and Federal legislation forces this because of an accuracy requirement (as distinct from tolerance) of V(act) +0/-10%. We therefore ensure that we meet this by aiming for 5-7.5% over actual speed with a tolerance of +/-3% as the fines / litigation costs for going outside homolated standards in a production vehicle are too great. [TwoSheds] Speedometers must by law indicate the max velocity that the vehicle is capable of achieving.>>

 

Jolly R. Blackburn

http://kenzerco.com

"I'd like to buy the world a coke, but I only have fifty cents"

Link to comment

This is a fairly good explanation of why a GPSr's speed reading may not always be accurate...

 

<<The later GPS receivers have algorithms that "cheat" to compensate for

losses of signal. The "cheating" usually means that the GPS will assume

that during the period when the signal disappears, the receiver is

travelling at the last known speed and direction. The more sophisticated

assume the receiver is travelling at the last known *accelleration*. Some

receivers average the speed over a period of several seconds (maybe as much

as 30 seconds), so will not respond instantly to changes in speed. This is

not a system limitation, but a deliberate feature in the software of the

receiver. Thus different people may have completely different experiences

based upon the particular model of GPS they have used.

 

A GPS receiver in a car will typically lose satellite lock fairly

frequently, as you travel under bridges, trees, power lines etc. The

accuracy of the GPS also depends upon the geometary of the satellites. If

the satellites are all clumped in a small segment of sky, the accuracy

decreases. A hand-held in a car typically can "see" only a small section of

sky so this condition is quite usual. As the car negotiates bends in the

road, different sets of satellites are "seen", which can cause a sudden jump

in the GPS reading. Even a permanent in-car GPS with rooftop aerial can

easily fail to track enough satellites when operating within streets

surrounded by high-rise buildings. The so-called "city canyon" effect.

 

So whilst the speed readout on a GPS is accurate to within 0.1MPH or so, it

may frequently display an erroneous figure for any of the reasons above,

plus a few others. To use it to check a speedo, drive at a constant speed,

preferably in a constant direction so the satellite constellation stays the

same, and wait till the readout has stabalised to a constant reading.>>

 

Jolly R. Blackburn

http://kenzerco.com

"I'd like to buy the world a coke, but I only have fifty cents"

Link to comment

This is a very interesting subject at a very unique time. Just last weekend I was traveling a streach of highway that was straight, flat and no overhead obstructions. My cruse control was set at 69 mph, when a County Shereff Deputy stopped me. My first reaction to myself and to him, was where did the speed limit change to 55 mph? He said "it didnt the speed limit is 65 mph" I said then whats the problem, his clame is that I was doing 80 mph. At the time I had been traveling for 2.5 hours, the first leg of my jurny was on 70 mph interstate and then the road I was on that was a 65 mph highway. I had been watching my GPS for some time and it consistantly was reading 67.4 mph, and showed a Max speed for the duration of my trip at 74.6 mph. He of course did not want to hear that and wrote me a ticket any way. My delema is wheather to challange him and his radar unit in court or not. I understand all of the legal implications of going to court, I am just looking for some good documintation on which to base my defence, in referance to GPS techonolegy and it's accuracy.

Link to comment

Here's a bunch of useless ramblings on the subject:

 

A well-written article in either Car and Driver or Autoweek within the last five years explained why speedometers are almost always calibrated to read slightly over the true speed, with european-produced cars the worst offenders. I don't remember it all that well, but I do recall that in Europe, or at least parts of Europe, manufacturers pay a fine for producing a car whose speedometer reads too low under any circumstance.

 

Manufacturers also had concerns about liability and are concerned that if all of the tolerances and conditions end up the wrong way (oversized or over-inflated tire, under-sensitive servo on the needle in analog speedos or long lag time on early digital speedos, tire speed differential on long curves, etc) that they will turn a car out the door with a sometimes under-reporting speedo, causing the customer to speed unwittingly.

 

Thus, they design in a "cushion", most aggressive (5MPH over at 60MPH in some cases studied) in many euro-cars because of the laws and fines, to make sure the "worst" case is an accurate speedometer.

 

I remember Japan and America were different, but I can't remember which produced the 2nd-biggest cushion on average. I think it was Japan, but I am not sure. My American-built Japanese sedan seems to underreport pretty substantially, 4-5MPH at 60MPH, based on my GPS reading on long, open, straight roads.

 

Now, for some really useless ramblings,

remember the Heisenberg uncertainty principal; you cannot determine both the precise location and precise velocity of any object. The more precisely you measure the location, the less you know about the velocity, and visa-versa.

 

A GPS measures your location at a certain instance in time. A single GPS location fix knows nothing of your velocity.

 

Multiple fixes over time are averaged for a velocity calculation, but of course that is merely an average speed between two or more points. But which points? When and where were they taken? Where, in fact, were you really going exactly that speed? See? If you know the velocity, you don't know the location! :^}

 

Seriously, a GPS speed with a strong fix series should be very accurate with decent satellite geometry (which should yield a high accuracy reading on your receiver), no blockage, and a straight level road at a constant speed.

 

Speed variation turns a GPS into the world's slowest-refreshing digital speedometer, yielding random intermediate results. Turns are just as bad, especially when you are turning more than 15 degrees between average-basis fixes, as they introduce geometry changes between your point fix lines which are likely not smoothed well by the speed averaging algorithm, so toss out those speed readings too. Since roads are generally not steep enough to make a significant mathematical difference in the speed calc most of the time, small grade changes are a lesser factor.

 

So you are left deciding between a cable counting the rotations of an uncertain-diameter wheel over time, or averaging location fixes over time. Good fix GPS in straight, unblocked territory with constant speeds is your best bet (even intermittant blockage is okay if your speed is constant and reflected signals are not a problem) . Choose the speedo for other moments.

 

Challenging a radar gun in court is not impossible, but a lawyer does it best. Unfortunately, the court is unlikely to put more faith in your gadget than the officer's. Most radar tickets are won on the picking apart of technicalities of the procedure used to determine the speed with the radar gun (Were all procedures in the radar gun manual followed by the officer? Imagine answering questions in court about the operating methods from one of your electronic gadgets. Again, lawyers experienced with this can do this well, according to what I've read, but I know I couldn't pull it off.)

 

It is probably much cheaper to go to court and hope the officer is a no-show. If you don't want to try that, you might get a big reduction in fines simply for showing at the hearing date on your ticket and offering a "my bad! I'm sorry, I didn't speed intentionally; I was just keeping my eyes on the road and forgot to check the speedo. When I first saw the officer flashing to stop me, I glanced down and saw only 69 on the speedo." sort of speech. Good luck.

Link to comment

I am pretty sure GPS measure speed using doppler shift, not (just?) the distance between two readings.

 

The speed of your GPS is more accurate than your speedometer, or a radar gun that is out of tune.

 

Wyatt W.

 

The probability of someone watching you is directly proportional to the stupidity of your actions.

Link to comment

Doppler shift would make sense for a speed reading, and I would be impressed if these little handheld units did it. It is too late in my head to ponder the algorithms, but if you consider the rotation of the Earth, the movement of the satellites and the curvature of their orbits, etc, I know I wouldn't want to do it with my slide rule.

 

I guess it would be possible. If it is true I will give regard my GPS V with even more respect than I do already.

Link to comment

Embra is right on. Do NOTE: If you change the profile of your car tires ie. from "78" series to "70" series, this changes the calibration of gearing, ex. measure the circumfrence of two different series tires - different circumfrence length, from OEM. The speedometer gear in the transmission needs to be changed. Go to a speed shop and have them recalibrate your gearing.

Link to comment

More useless information...

 

You can debate whether or not the GPS is more accurate at highway speeds, but I can tell you that at extremely low speeds the GPS is definitely more accurate.

 

It has been my experience - to the degree that I now rely on GPS in my work at times - that at speeds below 5mph, the speed reading of the GPS is far more accurate and useable than that of any cable driven speedometer that I am likely to have.

How this applies to anyone else is a matter of speculation, I guess. But it has been very useful for me.

 

"...clear as mud?"

Link to comment

I've found my GPS to be right on target with my Honda CRV. Sometimes it's off between 1/2 to 1 mph, and that's usually when I don't have a good amount of satellites connected.

 

Older cars, I believe your speedometer can lose it's calibration. Another factor could be if you ever changed the size of your tires, this could alter the accuracy of your speedometer.

 

Link to comment

One could probably imagine there will be some comparison differences comparing cable drive speedo to GPS speeds.

 

However with GPS speeds the error is relatively constant (for the same point in time) at most speeds but the % error can be considerably different.

 

At 5mph the the % error could be as much as 20%, where as if doing 60mph the % error could be 1 or 2% (generally less), which isn't a big drama if all things are considered.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif

Link to comment

Glad to see that someone posted about tire size.

 

Not counting the manufacturer induced speedometer error, tire size is the likely culprit. Also, I have heard that tire sizes are not always what are posted. A 235-85R16 made by one tire company may actually measure different from another manufacturer. So if you have ever changed tire designs, brands or sizes, this could throw your reading off. Tire inflation pressure can throw it off also.

I have the same size, make and tread pattern on my truck as it did when I first purchased it. I have found from every speed display I have ever passed the speedo on my Ford pick up is right on or give or take 1 mph. My wifes Honda is off by as much as 5 MPH at highway speeds.

 

Most police/highway patrol cars have calibrated speedometers, or at least a specific conversion chart somewhere in the car for that vehicle that shows actual vs. indicated speed at several different points.

 

Patti and Matthew McKenna, Watsonville California

Garmin 72

Link to comment

Here's my experiences thus far:

 

93 Pontiac Sunbird - bang on.

02 Chevy Venture - bang on.

98 Honda Accord - goes slower than spedometer by two or three kilometres per hour.

 

Hardly scientific, but maybe Garmin has shares in GM?

 

Even though you enter the wrong waypoint....you're still in the right spot!

Link to comment

Ok folks, I was reading this thread and thought I’d check things out. I went to work and picked up a radar set that wasn’t in use. The radar set I used was a Genesis GHS model radar which is a approved by the DOT of Pennsylvania as a approved speed timing device. This radar was within the proper dates for calibration and was calibrated by an approved speed timing device testing station. I also performed the tuning fork and internal tests for the radar to ensure its accuracy. The GPS I used was my Magellan Platinum with v5.12 software. The short answer is, the GPS was right on with the radar. The only difference was the radar was faster in detecting acceleration and deceleration. When speed remained constant the radar and GPS agreed.

 

Something that I found that was interesting was when I first started testing the EPE with my Meriplat was fluctuating between 90 and 30 feet. Of course after more locks with satellites the EPE was down to its usual 7-12 feet. Even during the larger EPE the speed was right on. One thing that should be noted is the radar rounds all speeds down. IE if my actual speed is 55.9 the radar will show 55.

 

Hope this helps!

Link to comment

Jolly B. Good pointed out in his post that the GPS unit tends to have a lag time with speed readings, which I have found to be true with my GPSr. It can even take a moment to hit zero after coming to a full stop, especially stopping from higher speeds.

 

That being said, the actual speed reading at the moment that the GPS unit gives is very accurate, more accurate than a mechanical speedometer(which is what most vehicles have). The trouble is when the speed is changing, the GPSr lags in reporting the changes compared to the speedometer.

 

I do use the speed indicator on the GPSr when driving on freeways. Keeps my eyes on the road better because it is closer to a "heads up" display with it on top of the dash.

 

That Quack Cacher:

Lone Duck

 

When you don't know where you're going, every road will take you there.

Link to comment

quote:
Originally posted by smithcon:

A GPS measures your location at a certain instance in time. A single GPS location fix knows nothing of your velocity.

 

Multiple fixes over time are averaged for a velocity calculation.


Completely and absolutely incorrect. As I wrote above, the velocity is not calculated from multiple fixes.

Link to comment

In a car with both of my receivers going, they both show the same speed. So does my friends' receiver of a different brand. Car companies build in a certain error in their speedometers. Tires size affect speedometer accuracy also.

when it come to speed- "In GPS I trust".

As far as radar goes, Those folks are in the business of making money. Power corrupts. 'nuff said.

 

Bender

 

Searching, for the lost Xanadu

Link to comment

I've had my Mer. Plat. in three different cars plus my 13 ton service truck and the Plat. always read the same or 1mph less on the open road. As for tire size or speedo gears changing the reading, I got a ticket for doing 66 in a 55 but I knew I was doing 60. Took car to speedo shop and sure enough, it was off 10%. The judge let me off with court costs.

Link to comment

quote:
Originally posted by fizzymagic:

Completely and absolutely incorrect. As I wrote above, the velocity is _not_ calculated from multiple fixes.


 

no I wouldn't call it completely and absolutely incorrect as this method is exactly what some (many) recreational units use.

 

There's certainly other things using GPS and Doppler shift in the carrier signals but I wouldn't think that this principle is used in ALL recreational units for velocity calcs.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif

Link to comment

OK, I couldn't help but throw my two cents into this ring...

 

Here's my setup:

98 Subaru 2.5 RS

TerraTrip 303Plus Rally Odometer

Etrex Vista

 

The rally computer takes the VSS (Vehicle Speed Signal) from the transmission. (The same source as the factory Speedometer) The Teratrip uses a factor I supply/calculate to to calc the distance the vehicle has traveled. A factor of 4100 for example means that that 4100 pulses the TT gets, it racks up one mile.

 

Subuaru quotes that there should be 4104 pulses per mile. (one pulse for every 1.2865 feet traveled.)

 

After some experimentation and frustration, I determined that the odometer on my car is very accurate, however, the Speedo is a consistent 5mph slow.

 

That is when I set the cruise control for "60" the GPS and the rally computer agree that I am in fact covering 65 miles per hour.

 

* The factor was determined after multiple runs along a 37 mile baseline of nearly straight road in Northern Minnesota in varying weather conditions. According to the GPS receiver's odometer, the mile markers at each end of the baseline are accurate to within .02 miles of each other.

 

JB

Link to comment

quote:
Originally posted by Kerry:

quote:
Originally posted by fizzymagic:

Completely and absolutely incorrect. As I wrote above, the velocity is _not_ calculated from multiple fixes.


 

no I wouldn't call it completely and absolutely incorrect as this method is exactly what some (many) recreational units use.


Could you give me an example of s recreational unit that uses successive position fixes to calculate velocity? I know that none of the current crop of Garmin or Magellan units do so.

Link to comment

I have extensive experience here. I drive a 4x4 with larger tires and know for a fact the speedometer is not calibrated. Assuming that you have good satelite reception, your gps is mounted properly, the gpsr is not in a battery saver mode or similar, I have found the Gps to be much more accurate than the vehicle. I have checked this constantly over the last 3 years of driving several vehicles, including boats, and have never found the GPSr to be wrong. I have even done the math on routes I knew my exact distance in mileage and compared that to the GPSr odometer and then to the vehicle odometer (which will also be wrong if the mph gauge is wrong). It even works well on commercial airlines (satelite reception is tricky here) though you have to take into account head winds, tail winds, etc.

Link to comment

GPS is going to be more accurate.

 

Tires wear, air pressure changes sizes, temperature effects them too (GPS compensates fro temp). Not to mention speedometer gear ratios can be only adjusted so much. Not every ratio is possible without having multiple gear pitch (tech term for numer of teeth per inch).

 

Then you get into parts wear.. Speedo's drift over time, gears wear out, tires wear out...

 

Basically all this has to mesh for a speedo to be accurate. Frankly, when its dead on, consider yourself lucky. Thas a lot to adjust for, and there is nothing being adjusted.

 

I am pretty sure American speedo's are required to be within 5% accuracy, when they leave the factory. Key part being when they leave.

 

Police speedo's are to be within 3%.

 

With GPS, a good lock, is really all you need for it to triangulate your possition, and determine your speed.

 

 

Now a question on odometer... Since it was brought up...

 

If you travel straight over a mountain, your actual position may only change by 1 mile. Howeve, due to going up and over, you may have actually had to travel 2 miles.

 

What distance will a gps read? I would imagine better models will register 2 miles, but what about cheaper models?

 

Has anyone tested this?

 

It would certainly have an effect on the odometer over time.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...