How "optimistic" is your speedo?

[dowmace]

I finally got around to attaching my Dream to my bike today so I could use it's GPS and found that my speedo is almost exactly 10% off what the GPS says, I'm more inclined to believe that the GPS is more accurate because following it keeps me moving well with the flow of traffic but if I use the stock speedo I end up getting past like grandma on sunday.

[Alex]

Sounds about right.

http://www.ninjette.org/wiki/Why_is_...eally_going%3F

[sombo]

I have recently discovered that mine seems to be closer to 5%. One of those signs that shows your speed shows me at around 56-57 when the speedo shows 60. It's a permanent one maintained by the DOT,so the radar seems pretty accurate; at least to my car anyway. It has shown dead on to my car's speedo every time I go past and always about 5% lower then my bikes speedo. Now I'm going through and redoing my math when riding to make sure I'm not going too fast lol.

[backinthesaddleagain]

My jeep is spot on. Bike is optermiztick by 10%.
Posted via Mobile Device

[Keenan84]

Mine is off by a full 10 percent. its ridiculous. any way of fixing this??

[Momaru]

I'm off by 10% as well, whether I pass the cop going 20 or going 90

[sombo]

Originally Posted by Keenan84

Mine is off by a full 10 percent. its ridiculous. any way of fixing this??

Nope, there's actually nothing wrong with the bike. I know of a few people that used to be mechanics for dealerships that told me the factories put the standard at +/- 10% on the speedos. It's also not just the ninjette either.

[dowmace]

Originally Posted by Keenan84

Mine is off by a full 10 percent. its ridiculous. any way of fixing this??

I'm using my Dream for a speedo now, that way I don't have to do any math, the last 10 miles of my ride are saved also so I can prove I wasn't speeding. And it's dead on accurate

[Keenan84]

this is so stupid. I could understand if it was like 2 or 3 km off, but 10?!? what a piece of ****. I shouldnt have to compensate in my head while im trying to ride. Unacceptable.

[kastas155]

put a 110/90 front tire and it fixes almost dead on. mines only off by 1-2 percent. when im going 56 on gps it says 55. so there you go.

[backinthesaddleagain]

If the speedo is optimistic, why isn't the odometer, or is it?

[Keenan84]

Originally Posted by backinthesaddleagain

If the speedo is optimistic, why isn't the odometer, or is it?

****, i never even thought of that!

[Alex]

The odometer error is not necessarily the same as the speedo error. It's pretty common for the speedo calibration to read quite high, while the odometer is much closer to accurate.

[tapdiggy]

The discrepancy in speed seems to come from the speed gear in the wheel hub instead of the gauge. I tested it. I put my after-market speedometer, using the stock hub gear to count rotations, and my bicycle computer, with a magnet on the wheel and sensor on the fork to give rotation readings. The calculations for the two gauges being the same: wheel circumfrence multiplied by wheel rotations in given span of time. When all other factors are equal, the deviation between the bicycle speedo and motorcycle speedo reading was 1 to 1.2. That translates to an actual speed of 25 reading 35, or 50 reading about 60

[sombo]

Originally Posted by tapdiggy

The discrepancy in speed seems to come from the speed gear in the wheel hub instead of the gauge. I tested it. I put my after-market speedometer, using the stock hub gear to count rotations, and my bicycle computer, with a magnet on the wheel and sensor on the fork to give rotation readings. The calculations for the two gauges being the same: wheel circumfrence multiplied by wheel rotations in given span of time. When all other factors are equal, the deviation between the bicycle speedo and motorcycle speedo reading was 1 to 1.2. That translates to an actual speed of 25 reading 35, or 50 reading about 60

I think you math is a bit off there. It's by 10% not 10mph. Meaning that at 35mph on the gauge you would be doing 31.5mph(35-3.5). Some believe it is by design to make us think we are going faster then we really are. All I know is that more then one mechanic has told me the factory tolerance on virtually all bikes for the speedo is +/- 10% accuracy. Sucks but not much you can do other then figure on being at least 5-10% slower then what your gauge shows.

[tapdiggy]

Originally Posted by sombo

I think you math is a bit off there. It's by 10% not 10mph. Meaning that at 35mph on the gauge you would be doing 31.5mph(35-3.5). Some believe it is by design to make us think we are going faster then we really are. All I know is that more then one mechanic has told me the factory tolerance on virtually all bikes for the speedo is +/- 10% accuracy. Sucks but not much you can do other then figure on being at least 5-10% slower then what your gauge shows.

My math is sound. We are saying the same thing: the ratio 1:1.2 is the same as saying the indicated speed is 20% faster than the actual speed. The exact percentage likely varies depending on how fast you go. I used 20% in my application so that, at highway levels of speed, my actual won't exceed my indicated, even with the increased distortion at higher speed. My numbers are obtained with an aftermarket speedometer that uses the formula from my previous post(wheel circumference X [rotation/time]). With actual numbers installed in the equation, it works out as follows: 25 MPH=40.2336 kM/h=670,560 mm/min. My front wheel with a 100/90/16 tire has a circumference of 1844 millimeters. 670,560 divided by 1844 equals 363 rpm. So, *deep breath* 1844 multiplied by 363 rpm equals 25 MPH. As I said, these numbers are used with a magnet and sensor. If the same numbers are used with the stock hub speed gear, they produce an indicated speed 10-20% faster than actual. My original position that it may be the hub gear that contributes to speed optimism was really the point; the experimentation I described was just evidence. If anyone cares to improve the math, I would be interested.

[Alex]

Truong - in your first post you say actual speed of 25 mph but reading 35 mph. That's an increase of 40%, and unlikely to happen. 1:1.2 for 25 mph would be 30 mph, not 35 mph. An actual speed of 50 mph but reading 60 mph is an increase of 20%, and while possible, is still on the high end of what other folks have been reporting.

Aside from the math, the best way that folks dial in those bike speedos is by using a GPS and calibrating them until they match GPS speed. There's no way to accurately measure the circumference of the outside of the tire with enough accuracy to get the speed within a mph or two.

[backinthesaddleagain]

Originally Posted by Alex

The odometer error is not necessarily the same as the speedo error. It's pretty common for the speedo calibration to read quite high, while the odometer is much closer to accurate.

Thanks alot Alex, so I really do have 10,700 miles on my '09. Next thing you are going to tell me is that there is no 7th gear that unlocks at 11,000 miles.

[FrugalNinja250]

Last year I decided to look into the speedometer/odometer accuracy issue out of curiosity. I took apart and analyzed a spare speedometer that I had laying around, and what I determined is that there are two different systems at play, one for the odometers and one for the speedometer. Both are driven by the same cable but their operating principles are fundamentally different. The odometer system has fixed gear ratios; in other words, the number of turns of the odometer wheels is rigidly linked to the number of turns of the front tire, an exact ratio of 876.5:1; in other words, it takes 876.5 turns of the wheel to advance the odometer exactly one mile. To get that many wheel rotations in one mile you need a tire circumference of 72.29 inches, which corresponds to a calculated tire size of 110/80-16 on the Ninjas through '07 at least..

Unlike the odometers, the speedometer is not rigidly connected to the cable. Instead, a magnet on the end of the cable spins inside a steel bell, using magnetic drag to rotate the bell and the speedometer needle that's attached to it. There is a coil spring attached to the needle shaft which counteracts the magnetic drag and also returns the needle to zero when the bike is stopped.

Because it is impossible to manufacture a mechanical speedometer with perfect accuracy on an assembly line, something has to be made adjustable so that the unit can be calibrated after assembly. From what I can tell so far that calibration is done in part using the spring that brings the needle back to zero. One end of that spring, a type of spring called a "clock" spring, is affixed to the needle shaft. The other end is attached to a metal tab that's part of the speedometer housing. Bending that tab alters the amount of spring force that counteracts the magnetic drag force. Making the spring tighter lowers the reading for a given speed, looser makes it read higher.

There are two ways that a speedometer can be out of calibration. The first way is for it to read high or low by an amount that is the same at all speeds, this is called offset. The other way is for the scaling to be wrong; in other words, the percentage of error varies with speed. Adjusting the clock spring only addresses offset error, not scaling error. I haven't determined if there's a way to adjust scaling error as of yet. I'm building a calibrator to work with my speedometer using a tachometer and a variable speed motor to drive the speedo cable, I should have that in service in the next couple of months.

Here's a sketch of the speedo gear ratios:

speedoratios.jpg

[adouglas]

Originally Posted by Keenan84

Mine is off by a full 10 percent. its ridiculous. any way of fixing this??

Black tape over the speedo so you don't have to look at it.

You ought to be keeping your eyes on the road anyway.

Most of the time I just go with the flow of traffic, so what the speedo says does not matter.

I actually don't mind the mental exercise of conversion at all. Doing math in your head is good for your brain and I try to do so whenever I have the opportunity.

[adouglas]

The fact that the odo and speedo are independent is good to know. A wrong speedo I can live with, but a wrong odo gives us inaccurate fuel mileage calculations and throws off the maintenance schedule.

Now I know I don't need to correct for inaccurate odo readings.

Thanks!

[jfauerba]

Mines off exactly 10% according to my GPS at all speeds. This is a lot better than being off 8% or 12% since it is easier to calculate on the fly. I wish somebody can figure out a way to fix the % error with the same size tires. Bigger tires would fix the speedometer error but will mess up the odometer.

[FrugalNinja250]

Originally Posted by jfauerba

Mines off exactly 10% according to my GPS at all speeds. This is a lot better than being off 8% or 12% since it is easier to calculate on the fly. I wish somebody can figure out a way to fix the % error with the same size tires. Bigger tires would fix the speedometer error but will mess up the odometer.

I'm working on that problem and will post results in a couple of months, maybe sooner. If you want to do it DIY it will involve a variable speed drill, optical tachometer, some wood for jigging, and some adjustments inside the speedo head. I'm pretty sure I can get offset error zero'd out now, just need to figure out how they adjust scaling error. My goal is to have my speedo exactly dead accurate at 60mph and within 1-2mph at other speeds below 100mph.

[250rStar]

sounds about right to me hahaha i used the gps on my phone and about the same as everyone else 10% off pretty lame and stupid if you ask me

[Verus Cidere]

Last time I checked, I was off by more than 10 percent (showed 70 while following behind my parent's car that was going 60). Haven't checked again since I put the sport demon on the front though.