Why are bike speedos not accurate.

[Rocketjock]

So I'm cruisin by one of those big courtesy roadside speedo signs that lets you know how fast you are going when I notice that my digital readout says at least 10k faster. The bike's an 06 Bandit and there is nothing wrong with it. It gets me wondering. Then I read about some guy using a hand held GPS to check his speed. I've got one of those so I give it a try. Sure enough, pretty well any speed from 30 on up shows me at at least 10k faster. I ask them at the shop and they say that all bikes are deliberately mis calibrated. Is this true? And, if so, why? You get accurate readings from cars, don't you? So, why not bikes. This bugs me. How do I fix it. You gotta be able to fix it, right?
And, oh ya. I own a 17' boat with a Yamaha 115hp and Yamaha gauges. I also have GPS on board. The gauges and GPS are almost identical speed wise.

[Red01]

It's true, the speedos on bikes tend to be optimistic. OTOH, if you check the odometer, don't be surprised to find that it's pretty accurate.

There are products out there, like the Yellow Box that will correct the error, but most will then make the odo inaccurate. Pick your poison.

Oh, and it's pretty typical for cars to be off, too... just not as badly as bikes tend to be.

[orionburn]

I went through the same things when I first got my FZR. Figured the speedo was bad given the bike was almost 20 years old. Then started hearing that as well that all speedos are pretty much off. Honestly I've been hoping they'd set some of those up around here again so I could do a few runs by it and find out for sure what speed I'm actually doing.

One thing I have wondered about. Does the difference grow exponentially the faster you go? Or is it generally the 10mph "rule" no matter what speed? The speedlimit on the highway I take to work is 65mph. Usually I do 80mph, which I assume to be around 70mph. I've passed a lot of cops when I've been in the 80-85mph range and haven't had any problems.

Normally I just throw in a little common sense - if I'm passing EVERYBODY (and rather quickly) then I'm going to fast. If I'm a little above average speed of those around me then I don't even pay attention to what the needle shows.

[JReviere]

This subject has been beaten to death on many bike forums and mail lists.  The simple facts are:  US DOT allows a 10% error to allow for differences in air pressures and non-OEM tires. Manufacturers choose generally to put this error on the "SAFE" side. It protects them from litigation when someone is hurt, they can't blame the manufacturer saying they didn't know they were going as fast as they actually were going. It also plays well the with governmental Safety NAZIs who like to see people going slower than they think they are going.
In most cases, the error is a PERCENTAGE of the speed indicated. I've found the error to be between 5% and 7% most of the time.
A lot of people on the ST1100/1300 lists and forums have installed either GPS or a common BICYCLE speedometer which can be calibrated to the actual wheel diameter making it very accurate.  
I just do the math in my head. I've found my B-12 is 7% optimistic. So, I do 10% mentally, divide that by half and half again, putting the figures together, I come up with 7.5% of indicated which is close enough for me...  Then I simply add (mentally) the number I've calculated to the speed indicated in order to know what speed I need to indicate to go the speed I want to travel.  Not difficult at all.
JR
Too soon old and far too late smart
Lake Livingston, TX

[Red01]

In most cases, the error is a PERCENTAGE of the speed indicated. I've found the error to be between 5% and 7% most of the time.

That's pretty much guaranteed on cable driven speedos. Electronic speedos may be a steady percentage, or an error that gets worse the faster you go. The latter is how my B12's speedo works - at least compared to those portable radar trailers.

[Rocketjock]

So, they could make em accurate but they don't. Go figure.

[Red01]

Just read a test on the new 1250 in MCN and they said their when their test bike's speedo read 65 it was doing 60.something... and said that was pretty typical for todays machines.

I agree, it's BS they won't make 'em more accurate.