Reeko

A quick check show that 245/45/R18 should fit. They have almost exactly a 3% greater diameter, and .6 in wider footprint.

DisneyDestroyer

:grin: thanks for the info, but it doesn't solve the larger problem of inaccuracy.

Gord96BRG

There is no problem of inaccuracy. The odo is well within specifications, both design and legal. It isn't meant to be perfectly exact, and couldn't be even if they wanted it to be (unless you want your odometer to cost more than the entire rest of the car...).

Regards,
Gordon

RobDickinson

Oh. My. God.

Listen to the facts.

Manufacturers cant make 100% accurate speedo's.

Manufacturers cant have the speedo show a speed of less than what the car is mooving at. (i.e. cant show 60 and you realy be doing 65).

Manufacturers set it so it always over reads.

In Europe it can be as much as 5% out and still be ok, and it gets worse the faster you go (in pure MPH terms).

Your speedo may not even corralate to your odometer, who knows, but things are made to tolerances not exactitudes.

Get over it, or go design a cheap, perfect speedo.

Reeko

There is NO problem of inaccuracy. It is well within specs.

In my opinion, since the stock tires wear very quickly, I plan to get better tires when needed. I might go slightly larger (just because I want wider tires, not to trick the ODO. )

DisneyDestroyer

"Within specs" and "accurate" are two different animals. Please don't try to tell me they are the same. Accurate is black or right, right or wrong. It either is or it isn't. 2% out of 100,000 miles is significant enough (2000 miles) that you can't say it's accurate. 10 feet out of a mile is too much to say it's accurate. It's true that different people have different person opinions of when something is "too off to be accurate". That's why the range provided by regulations is useful here, to avoid legal entanglement for Mazda.

Within specifications and government regulations, sure. Accurate, no.

Reeko

Sorry,
There are no absolutes.
Accuracy is never absolute, it is defined within some margin of error.

So, should I assume that you have verified all of your previous vehicles odometers to a level of accuracy that meets your requirement? I assume you did this using some highly accurate device (I would trust a GPS on level ground, but not milage markers).

Gord96BRG

Actually, you're very wrong about what defines the term "accurate". If you had studied science or engineering, you would know that it is physically impossible to get 'accurate' as you think - what you CAN define is degree of accuracy, ie acceptable margins of error.

EVERY "accurate" scientific measurement includes a +/- margin of error. You want to measure temperature? It's 75F, +/- 3F on your average household thermometer. If you use a calibrated scientific instrument, you might get 75.395F, +/- .005F.

Even on a racetrack where they use electronic transponders and very fast computers to calculate lap times to .001 seconds (which actually means their lap times are accurate to +/- .0005 seconds), they can't tell you what speed the car is actually travelling, because the exact distance travelled over a lap varies depending on the line the driver takes.

So, your odometer is accurate to +/- 1.5%. If you spent thousands of dollars, you might get speed measuring equipment that's accurate to +/- .1%. You will NOT EVER get speed measuring equipment that has NO margin of error, as you seem to imply. It's impossible.

Moral - give it up, your odometer is perfectly accurate to acceptable legal and practical requirements. Nobody will do anything to change it, and if you make a big fuss they will start laughing at you for your ignorance.

Regards,
Gordon

PS - Why do you presume that it's only Mazda that doesn't have odometers accurate to .000000001 feet per mile? The accuracy requirements apply to ALL MANUFACTURERS, not just Mazda. As has been said to you over and over, it's not a problem, and it applies to ALL cars. Why not try to actually learn something instead of being so pig-headed that you make yourself look silly?

DisneyDestroyer

I make no presumptions, I have had the same issue with other cars I've had.

I didn't give my car back, etc. because it's not that much of a problem.

Neither changes the fact that it bugs me.

druck

You are never going to get much better accuracy - remember the diameter of the wheel changes considerably with tread wear and tyre pressure. The cars odometer will be calibrated for the average condition of the tyres. So it may get better towards the mid life of the tyre then go back the other way.

Cheers
---Dave

Haris

Dude, you gotta be kidding me if you were planning on getting 100% accuracy in speedo and odo? lol. There is not mazda, honda, toyota, nissan, chevy, ford, mercedes, bmw, audi, vw, dodge, or any other car that's affordable that is actually 100% accurate in odo and speedo. Just deal with it. Wow, 2000 miles out of 100,000 miles is really big deal. If your selling a car, and u say it has 100K or 102K, it wont make a big deal. It's basically a same thing. Unless at 100K miles, your car will actually only have 90K miles, then there is no problem you should worry about.

MuzzleFlash

If you're getting accuracy under 2% stop complaining. Sheesh!

Construction radars are typically off-axis due to their short range. That means they suffer from something called cosine error causing them to read a tad low.

Small percentage changes in tire diameter can come from pressure and load. That means things like temperature and the number of passengers will affect the number or revolutions per mile and affect accuracy.

Your mileage is excellent - especially cosidering EPA highway ratings are done at 55 MPH IIRC.

If you want someone to blame for spedos being biased slightly to the high side, blame class action lawyers and the government. If Mazda calibrated them dead nuts, there would still be a few that read slightly low. Enter some class action scheister to sue Mazda for millions claiming some percentage of spedos induced their drivers to speed by reading low. After he skimmed 99% of the settlement for his fees, we would get to split the remaining 1% 40,000 ways. Of course, Mazda would factor the legal costs into its product and we'd all indirectly pay.

RXTACY

I work for the WASHINGTON State Department of Transportation. Transportation departments consider mile post markers addresses, like a house number.

They are not accurately placed.

Many times the path of the roadway will change, changing the distance, but they do not relocate the mile post markers. Think about it. Personally, I would not even trust the "test mile post markers" you sometimes see around.

Your speedometer is likely accurate within reasonable tolerance.

TybeeRX-8

And if no one here knows it, even police radar is not 100% accurate. Each officer is required to be trained and the gun calibrated for each daily use. And, according to some officers i know, except maybe in a school zone, you will get a 10 mph leaway, e.g., they won't write a ticket unless you are 11 mph+ over the posted limit. I'm sure there are numerous tales about getting stopped at lower speeds, but that's a product of the local revenue collectors. My complaint is that they don't write tickets for people going more than 11mph LESS than the posted limit!

Nubo

Yes, I'm thinking an inertial navigation system would work nicely

As other posters have mentioned, if it really bugs you; switch to larger tires to compensate. As the OEMs will certainly be done for by 20,000 miles you'll have plenty of chance to recoup before the warranty expires.

Nubo

Your idea that accuracy is a binary condition isn't true. It's not a matter of opinion - there is no perfect accuracy even with the finest instruments. I imagine you could measure the distance accurate to within angstroms - but that scientific instrument would come with specifications saying something like "accurate to within +- 4 angstroms per mile" or some such, and there would be many other considerations as well. how will you measure the distance - what part of the road will you use? How many times will you change lanes? Which tire's rotation is "accurate" - they will all be different! What about the elastic deformation of the tires? How many times will you run over road stripes? All of these variables add up. In the world of scientific measurements this type of thing results in measurements that are given in terms of "significant digits". For example, if the measuring instrument is accurate only to 1/100 of a mile per 600 miles, your fancy odometer reading of a long trip might say 600.0247 miles. However that number is not reliable and you would have to limit the measurement to 5 significant digits - 600.02. Similarly, the other error factors all impact the quality of measurement and could further reduce the number of significant digits you could claim as "accurate".

Don't even get me started on Fractal Geometry -- but consider "what is the length of a coastline"?

pp13bnos

I can't beleive that there are people like this......its truely amazing....

allstate

My point all along. Those who are upset about the inacuracy of their spedo's, how are you using something that is inacurate (the mile markers) to measure accuracy Even GPS is not 100% accurate. I know because I use them all the time....I'm a pilot.

Reeko

I do however remeber seeing Calibrated mile markers on some hiways. They had signs that said they were calibrated for verifying your speedo/odo.

Don't know how accurate they were, but they were probably better than the typical mile marker.

fluid dynamics

You guys are measuring spees and MPG to the .1 unit? Do you realize that the accuracy the measuring instruments you base your conclusions on are not within the precision of your data analysis? In other words, car speedos do not measure within +- .1 mph. Just getting within 5% would be reasonable.

Concern over missing another 800 miles of your warranty over this? Get real! Nothing is guaranteed in life. And as far as MPG concerns, welcome to the rotary world! I'm gettin 18 on my less powerful and lighter 87 RX7 during normal daily driving.

You either love the rotary or buy a 350z, or some other clone.