Annoying… speedo errors

[adouglas]

My car has a speedo that is dead nuts accurate.

My bike's speedo is hopelessly optimistic.

WHY in this day and age are manufacturers INCAPABLE of giving us accurate instrumentation.

I could live with one or two percent error, but mine's somewhere in the vicinity of 8 or 9 percent. That's just silly.

It's so far off the speedo is useless so I don't bother looking at it.

Rhetorical question, really… just venting. I just dropped $90 on a SpeedoDRD calibrator to fix the problem… something THAT SHOULD BE BUILT INTO THE BIKE TO BEGIN WITH.

[Reswob]

I think bike manufacturers make the speedo 10% fast on purpose. The 250 did it, the Bonnie does it, and my cousins R6 does it. That's too many to be a coincidence. All you really need to do is figure out what indicated speed is the max speed you want to be going on the highway (i.e. keep it under 90 indicated / 80 actual). Off the highway I rarely look at the speedo anyway.

[Bill N]

Originally Posted by Reswob

I think bike manufacturers make the speedo 10% fast on purpose. The 250 did it, the Bonnie does it, and my cousins R6 does it. That's too many to be a coincidence. All you really need to do is figure out what indicated speed is the max speed you want to be going on the highway (i.e. keep it under 90 indicated / 80 actual). Off the highway I rarely look at the speedo anyway.

I agree, I'm quite sure its a liability issue, in other words a lawyer solution for manufacturers. Most errors seem to be about 10% which makes the math easy. My FJR is about 5% but my 250 and other bikes I've owned are right at 10%. Easy to get your brain around that. 70 indicated = 63 actual. I've confirmed this error many times on roadside radars that display actual speed as you pass by. Bill

[Alex]

I've heard that it is because it is so easy and common to change gearing on motorcycles (with different size sprockets), the speedo is calibrated so it's highly unlikely it will ever read low, even if the owner changes the final drive ratio a bit. Don't know if there's anything to it, but perhaps it's a partial explanation.

[KawiKid860]

I just do 3-7mph indicated over the limit depending on whether I'm in the city or highway. Doesn't bother me a bit, it's just the way bikes are.

I've also heard that it's because the tire wears down with mileage causing inaccuracy.. dunno how true that is but it makes sense. Not for a huge difference though, and car tires wear down too.

As far as gearing goes, most people I know tend to go down in gearing for more grunt so that'd make the speedo read even higher.

even at 10% inaccuracy I wouldn't call the speedometer useless, just throw a few mph onto what it reads and you're golden.

[dfox]

Originally Posted by Alex

I've heard that it is because it is so easy and common to change gearing on motorcycles (with different size sprockets), the speedo is calibrated so it's highly unlikely it will ever read low, even if the owner changes the final drive ratio a bit. Don't know if there's anything to it, but perhaps it's a partial explanation.

speedo sensor is on the front wheel though... gearing changes are a non-issue.

might have more to do with tire size and life than anything else. on an ever changing diameter, you can't be 100% accurate, ever. throw on a bigger tire, it screws it up even worse.

it reads the number of revs and translates that to mph or kph. it gets to miles or kilometers by estimating the circumference of the tires on the road. take a turn, your diameter changes. new tires? different circumference than worn tires. bigger tire than stock, circumference changes.

[KawiKid860]

Originally Posted by dfox

speedo sensor is on the front wheel though... gearing changes are a non-issue.

MOST bikes read the speed off the countershaft. I believe the 300 does this as well as far as ninjettes go.

[Alex]

Originally Posted by dfox

speedo sensor is on the front wheel though... gearing changes are a non-issue.

On the 250, sure, but that's rare for most bikes at this point. The 300 moved to a pickup on the transmission. Speedo errors do seem pretty common across many makes. Our Buell Blast from a million years ago had a pretty accurate speedometer, for what it's worth.

[dfox]

Originally Posted by KawiKid860

MOST bikes read the speed off the countershaft. I believe the 300 does this as well as far as ninjettes go.

never owned another bike, but I find that very interesting. edit---- sensor I wast thinking of on the car was for stability control, not actual speed sensor... --- move along.

put a larger tire on though, and your error will go up. larger tire diameter, the slower your speedo will read.

[cbinker]

All japanese bikes read 7-10 percent fast, been like that for 10-20 years. Rumor I heard is they do it to make you think you are going faster then you really are to keep YOU out of trouble.

[MR2Aaron]

Originally Posted by KawiKid860

I just do 3-7mph indicated over the limit depending on whether I'm in the city or highway. Doesn't bother me a bit, it's just the way bikes are.

I've also heard that it's because the tire wears down with mileage causing inaccuracy.. dunno how true that is but it makes sense. Not for a huge difference though, and car tires wear down too.

As far as gearing goes, most people I know tend to go down in gearing for more grunt so that'd make the speedo read even higher.

even at 10% inaccuracy I wouldn't call the speedometer useless, just throw a few mph onto what it reads and you're golden.

The speedometer is driven off of the front wheel. If you change gearing, the front wheel in unaffected.

Or do the newer bikes have some sort of electronic deal that reads off of the transmission like a car?

[Alex]

Originally Posted by MR2Aaron

Or do the newer bikes have some sort of electronic deal that reads off of the transmission like a car?

Yes. Very few bikes still read speed off a front-wheel sensor like the older ninjettes.

[MR2Aaron]

Originally Posted by Alex

Yes. Very few bikes still read speed off a front-wheel sensor like the older ninjettes.

That's kind of a bummer. Reading the front wheel seems like a pretty elegant solution to the gearing problem, to me.

[adouglas]

None of the proposed rationales make any sense to me. Tire wear? Non-stock gearing? Seriously?

Let me add one more to the mix: Sensing the front wheel will introduce inaccuracies in both speed and odometer readings if you get the front wheel off the ground.

I mean, come on. Really, now.

This is a street-legal vehicle. The stock instrumentation should read accurately using stock tires and stock gearing. If you mod the vehicle, you're on your own.

Error due to non-standard gearing isn't a "problem." It's a side-effect of modification. Mod your gearing, you should expect to have to modify other parts of the bike too.

[MrMax]

The new Honda's are supposed to be spot on now according to what I hear. The car division got sued a little while back due to some of their cars reading fast also. I guess it was proven in court that some read up to 7% too fast. the lawsuit was concerning that the car would run out of miles on the warranty that much faster.

But yes...almost every bike Ive owned jap/euro/USA had read >7% too fast as well.
Regards,
F

[rasta]

Didn't realize that happens. Oh well, it's like purposely setting your clock a little fast, just to be safe from those nasty speedtraps

[Alex]

Originally Posted by adouglas

I mean, come on. Really, now.

I'm not sure what you want to hear, Andrew. It's being done intentionally. The likely reasons have been laid out. Why do you think the speedometers are set to read high, while the odometers are usually much more accurate?

Google finds me this doc related to EU requirements for speedometers, with the very last statement in the doc making it clear that in no circumstances can the speedometer read lower than actual speed. Seems like it's one very good incentive to make sure there's quite a bit of wiggle room on the other side of that line.

[headshrink]

Unfortunately the CHP didn't by the moto speedo-error excuse. To be fair though, I did earn that one regardless. Amazing how effective these meetings can be for behavioral modification. LOL!

[csmith12]

You can't trust the wheels for speed, what if only one of them is on pavement? Even on a what feels to be a "smooth" road, rear wheel speed can differ from front wheel speed. Bumps and a heavy handed throttle makes it even worse. You can't trust the trans because of gearing.

You want a more accurate speed? Get a GPS w/ mount and also take note, that those are not 100% accurate either based on the update frequency. However, they are much more accurate than the stock speedo.

[BRad704]

I've tracked my speed versus my Garmin 405 and I'm about 9-10% slower than what the speedo says. I just keep that in the back of my mind and haven't really had an issue worth complaining about yet.

But to AD's point, you'd think things would be engineered true to life, instead of building in some idiot-space to "keep us out of trouble".

[Bill N]

Originally Posted by BRad704

I've tracked my speed versus my Garmin 405 and I'm about 9-10% slower than what the speedo says. I just keep that in the back of my mind and haven't really had an issue worth complaining about yet.

But to AD's point, you'd think things would be engineered true to life, instead of building in some idiot-space to "keep us out of trouble".

You can thank the lawyers. In the gun world its known that factory triggers are known as lawyer triggers. Obviously for liability issues. I bought a beautifully designed target rifle. The trigger was extremely heavy. Easy/cheap mods fixed it. No need to fix our speedo's just realize they are about 10% optimistic and move on. Bill

[KawiKid860]

Alex has a good point. Throw on non standard gears, speedometer mph now reads below, opportunity for a stupid lawsuit.

[adouglas]

Installed the SpeedoDRD this morning (REALLY easy on the GSXR) and I'm now within 1 or at most 2 mph of reality.

I used 8.5% of correction.

Hard to get it precise with GPS lag being what it is. There aren't many opportunities to go steady-state for a whole mile around here. At some point I'll go find a measured mile and do some timed runs.

Great product. Wish it wasn't necessary.

[Panda]

It's not covered under warranty if you prove to then the speedo is off?

[adouglas]

They're ALL off. So was my 250.

Dumb.

Nothing wrong with the bike. It just comes that way. It's common knowledge... besides, not worth the hassle to go after Suzuki about.

[Panda]

Yeah both my bikes and my car is off too. I agree it's annoying. I keep forgetting to mention it at the dealer though.

[adouglas]

If it bugs you the SpeedoDRD is not all that expensive and is REALLY easy to install. All you need to do is find the connector for your speed sensor, unplug, plug this thing inline and program it. The latter is not intuitive in any way but it's not hard. Instructions are clear... just don't lose 'em!!!

My speed sensor is clearly visible on the left side of the engine so all I had to do was lift the tank and find the plug.

[Mocha Man]

I have to go about 15 mph over on the highway to keep up with traffic. I'm not sure if it's the crappy drivers in my area or my speedometer though... Whenever I drive by one of those in the city that tracks your speed it checks out just fine.

[FrugalNinja250]

Some thoughts on the issue...

The pregen speedo is mechanical, it uses a spinning magnet driven by the speedo cable to induce magnetic eddies in a steel bell that is connected to the speedo needle. As a result, the spinning magnet drags the bell around with it. There is a clock spring (spiral spring) that resists the movement of the speedo needle, so it takes more RPMs of the speedo cable to drag the needle further around. Tiny variations in magnet strength and clock spring manufacturing will create error.

Calibrating one of these involves spinning the magnet at a known speed and then bending metal parts to change the spring tension and the angle of the magnet inside the steel bell. It's finicky, precise work that requires skill. On an assembly line the goal is to get it close enough, within specified tolerances. Since the goal is to not have the speedometer read low, the tolerances are likely set up such that resulting error range will make the speedo read higher than actual speed so that tolerance error can't make the speedo read low.

I did some digging in the FMVSS for regulations on accuracy, but the only thing I found was that commercial speedos had to read within 5mph at 50mph. The general trend seems to be that speedos should be within 10% of actual, with 6.5% found in an EU document.

Electronic speedometers can be very accurate, but it's likely that some positive error is built in so that tire size changes (and gearing changes for speedometers driven off the transmission) are less likely to drive the speedo into negative territory.

I found no evidence that there is some greater conspiracy to make all speedos read high by a particular number. If there were, then the speedo errors being reported would be very consistent, but mine is off by 5 at 70, about 3 at 45. Others report errors from 2 to 10 mph at various speeds.

Since speedometer error is NOT an affirmative defense against a speeding ticket, it's in the best interest of everyone on the manufacturing side of things to ensure that the speedometer never reads lower than the actual speed.

[Whiskey]

Pretty much every speedo runs 8-12% high, I've heard a few times that it's a case of companies avoiding any costs if you're done for speeding & claim that it's a vehicle fault. If it's calibrated high they're able to say that without having someone take a look at the bike.

Originally Posted by MR2Aaron

That's kind of a bummer. Reading the front wheel seems like a pretty elegant solution to the gearing problem, to me.

But causes a problem with dual front discs

[rojoracing53]

It bothered me so much I spent $400 on a Koso instrument cluster and got so much more then just corrected speed. My most expensive mod but probably my most appreciated.

[MR2Aaron]

Originally Posted by Whiskey

But causes a problem with dual front discs

Does any Ninjette actually have dual front discs, though? I didn't think so.

Regardless, that's hardly an intractable problem. ABS systems have sensors that measure wheel speed (trigger wheels, usually), and those are available on bikes with two discs.

[Whiskey]

Originally Posted by MR2Aaron

Does any Ninjette actually have dual front discs, though? I didn't think so.

Regardless, that's hardly an intractable problem. ABS systems have sensors that measure wheel speed (trigger wheels, usually), and those are available on bikes with two discs.

The world doesn't rotate around a single small capacity sports/sport-tourer model.

And how many modern 600+cc machines stop on a single front rotor?

It's cheaper & easier for companies to go with a single type of system, having a specific system for small bikes and another for twin disc bikes is expensive.

Having a fairly rigid cable guide coming off between the wheel & disc of the majority of their line-up wouldn't be 'elegant'

[JA-Moo]

Throw on a 20.00 sigma sport bike speedo if you are OCD about a speedometer. Stock speedos have been off for the last 5 or 6 decades, if they ever were accurate.

[fast1075]

Both my Buell Lightning, and my Honda Elite 110 scooter are dead nuts on. My "new" 2006 ninja is however a tad optomistic.