CarTalk.com Blogs Car Info Our Show Deals Mechanics Files Vehicle Donation

Accuracy of speed and mileage on vehicles.(show #1232)

Hearing the call from Mike from Montana driving the BMW about the accuracy of the speed of the car. I would think that the speed of vehicle and or the mileage is determined by the revolution of the wheels (“risky business”). We may have occasionally seen a person pushing one of those little wheel measuring devices. As the wheel goes around it indicates the distance on a gauge.Think what would happen if that wheel was taken off and a small skate board wheel was put on. The revolution distance is shorter and so rotation is faster than the larger wheel. Different wheel and rim sizes would effect the speed and mileage of vehicle. I have known for a long time that my vehicle’s speed is “faster” than indicated on the speedometer from the radar speed indicators that are often in school and slow speed areas. What say ye MIT grads?

I am not an MIT graduate,but changing the wheel size most definitely affects the speedometer and odometer reading. I was hurting for money and found some oversize tires that would fit the 1968 AMC Javelin that I owned. These tires were two sizes larger than the original equipment. When the speedometer indicated 65, the car was going about 78.

Wow! AMC Javelin…that’s worth a ticket. If that car could talk…

@Seafood snacks–That AMC Javelin was a great car. MIne had the 232 cubic inch 6 cylinder engine. I bought the car in May of 1971 for $1200. It had studded snow tires on the back which were illegal for summer driving and needed new front tires. I lived in married student housing at the time and found a pair of barely used tires that my service station mechanic said would fit even though they were two sizes larger than the original equimpment. I then bought 2 new tires from the service station for the other end of the car. At any rate, we were on a road trip from Indiana to Nebraska. On the Interstate in Iowa, my wife noted that the engine seemed noisy. She also noted that the Greyhound buses seemed to be driving awfully slow in the state as was everybody else. When we stopped for gas, the mileage was about 16 mpg where normally the car ran about 22 mpg. It also was down a quart of oil and I had never had to add oil between changes before. I also had the accelerator almost on the floor to run between 70 and 75 (the speed limit was 70) by the speedometer. Suddenly, everything clicked–When we calculated our speed for a measured mile with the odometer reading 72, we were actually going about 90 mph. When I figured this out and ran 60 by the speedometer, the engine became quieter, the gas mileage came up to about 22 mpg and I was running along with the Greyhound buses at their speed;

My son had a '68 Javelin that had a 290 V8

Tire wear and tire pressure also impact the rolling radius of the wheel, so until cars’ speedometers use GPS instead of gears, we will have the small inconvenience of our speed reading (usually) fast.

I believe Mike said he thought the odometer was off by approx. 5.4% which would correspond to about a 1 inch difference in the wheel size. I was in a Toyota dealership the other day and they had a car on the floor with dealer-added chrome wheels an inch larger than the usual. I asked the salesman, “Won’t that throw off the speedometer?” He just shrugged. My guess is that when manufacturers offer multiple wheel sizes they don’t recalibrate the odometer/speedometer. And since they’re engineers, they take the safer choice and calibrate it for the larger wheel size, so drivers with the smaller wheel size think they’re going faster. Does anyone here know whether the speedometer in new cars is adjusted based on the wheel size they chose?

Getting back to the subject at hand, I can assure you that on interstate highways mile markers are EXACTLY 5280’ apart. The only exception might be if the lollipop needed to be in the middle of a bridge where there was nowhere to stick a short pole. The federal government has to know exactly what they are paying for. Even the number of fence posts per mile has a federal specification.

Click and Clack’s laying the blame for the caller’s car not doing a measured mile in 60 exactly seconds is BOGUS. I worked for my state’s department of transportation for two summers during college. I was lucky enough to get onto the civil engineering side of the team. That way I didn’t have to cut weeds or hold a slow/stop sign to direct traffic. You know CIVIL engineers, those guys who use doilies on their saucers and sip their tea with their pinkies held out while wearing their striped bib overalls and engineer caps. I worked on a survey crew most of that time. I can’t tell you how many times I yelled “CHAIN” every 100 feet those summers, but we measured up and down that stretch of interstate MANY times. There is a number pressed into the wet concrete of the shoulders too. IIRC, those are exactly 500 feet apart.

On a slightly different note, if you want to check you speedo, check the time between mile markers. Then divide that time into 3600 to get your exact MPH. That’s what I did before GPS.

And now for some complete digression. I had a '69 Javelin with the 290 V-8 that my oldest sister bought brand new. She drove it for 10 years. I bought it from her for $500 when she moved to a Toyota. I drove it two more years, and sold it for $1000. It’s one of the many cars I wish I had kept.

try using a gps that has an instant readout of your speed. my garmin nuvi has both a speed and odometer which seem to be accurate in the SW US. good luck

MG McANick and Chuckie are Exactly right, those mileage markers are exactly 5280’. Also, since GPS is accurate to 10 to 30 feet when it has 12 satellites in view, you can use it for distance. Since its an absolute error that doesn’t depends on distance, the longer the total distance measured, the more accurate the reading.

I’ve always calibrated every car I’ve owned and seen errors from 0 to about 7%, always on the high side. So, considering all the lawyers out there, I think its to make sure if you’re in an accident or get a ticket, you can’t blame the car maker that the speedometer was reading too slow!

My GPS shows the same difference in speed vs what I see on the speedometer as your caller mentioned and I have always wondered which one was correct. Since people tend to pass me when I am doing exactly the speed limit according to the speedo I think the GPS is more likely correct. It certainly seems to correctly reflect the distances I drive long distance. I drive a BMW 325ci, use a Tomtom One GPS and my tires are the correct ones for my vehicle. Perhaps there is some conspiracy among manufacturers.

MG McAnick I agree with you. Also they might not place a marker in the middle of an off ranp.

I think Joel’s tire size hypothesis is dead on. I did some back-of-the-envelope math regarding the mile marker possibility - 5000’ +5.4% = 5270’, or pretty damned near a mile. What if the caller’s 5.4% measurement was slightly off, and the miles were marked at 5000 feet incorrectly due to someone’s formula error?

Neat, but I think I like the tire size idea better.

I have to wonder how the mile markers are set. If they are set as the crow flies, vs actual mileage as you are going up and down hills that could be a factor., though it would seem under that scenario you would have more distance between markers.

I had a 2001 BMW Z3 that was also about 6% slower than registering on the speedometer. Instead of using mile markers, I used a gps and compared the speed on the gps vs the speedometer.

I think the mile markers are over the road distance markers, not ‘as the crow flies’.

I am a mechanical engineer who has done a lot of vehicle and tire tests. The problem of having the odometer disagree with the highway mile markers can easily be resolved.

The speedometer cable in all cars should make 1000 revolutions per mile. The size and amount of wear on your tires can contribute a small error. It is easy to check. Measure 528 feet along the street, disconnect the cable from the speedometer head in the dash, put a piece of tape (flag) on the end of the cable, and count the (100) revolutions as your helper slowly drives the 528 feet. If you are obsessive, increase the length of the test course. You can buy a small gear box with interchangeable gears at your local auto parts store to change the number of revolutions to exactly 1000 per mile.

The distance between highway mile markers is accurate, but beware of discontinuities. I know of a mile marker near King City, CA, that shows two figures that are significantly different. The highway might have been rerouted or some other change in the survey was made.

Go to http://www.net-comber.com/tirecalc.html for a tire size vs. indicated speed calculator. I think it will solve the question without access to an engineering degree.

In one word, GPS. A simple GPS nav device will be WAY more accurate than any string!
However, you would need a pretty sophisticated statistical analysis to justify the grant application!

In california (land of fruits, nuts and flakes) it is illegal for a manufacturer to sell a car that displays a speed slower than the actual. This is to prevent speeders from claiming their speedometer is off and it is the manufacturer’s fault.