Speedo Accuracy

Discussion in 'Electrical' started by AussieSteve, Nov 6, 2009.

  1. AussieSteve

    AussieSteve Active Member

    I've fitted a (Cateye) bike speedo/computer to my bike and originally set the wheel circumference to the figure in their chart for my tyre: 26" x 1.95" MTB - 205cm.

    Today, I got around to checking the actual circumference, under load - 202cm.
    (I set the bike with the valve at the bottom and marked the spot, then sat on the bike with my normal weight distribution, rolled forward one turn and measured the distance.)

    I also have a cheap Chinese speedo on my electric bike and it was even worse - 208cm recommended, 200cm actual.

    I only weigh 50kg, so most people would have a much greater error. Each brand of tyre would also be different. Pressure makes a fair difference too.

    I've noticed some of the extraordinary speed and calculated RPM claims for these bikes and wonder if this has something to do with it?

    ... Steve

  2. AussieJester

    AussieJester Member

    I tend too agree, unless the speed is GPS certified i treat all claims with a certain degree of scepticism. Cateye are ne of the better quality brands too i hear...There SHOULD be an option to enter in your exact tire size IMO.

  3. AussieSteve

    AussieSteve Active Member

    Sorry, I didn't make myself totally clear. There is an option on even the cheap speedos to enter the exact wheel circumference, in increments of 1 cm, (which I've now done on both bikes).
    The problem only occurs if the values in the supplied charts are used.
    As I discovered, it also pays to measure it out by rolling the wheel, fully loaded. A simple measurement of the diameter multiplied by pi will be a fair way off. (Several cm)
    Yesterday, my top speed was 36kph. Tomorrow on my next ride, with the accurate circumference?? (I could calculate it, I guess, but I'll wait until tomorrow)

    ... Steve
  4. AussieJester

    AussieJester Member

    I still tend to think they aren't 100% accurate, a GPS is.

  5. AussieSteve

    AussieSteve Active Member

    If you have a GPS. Still, for those like myself with only a speedo, nothing wrong with having it as accurate as possible. (Can't justify the price of a GPS purely for speed measurement)
    Incidentally, many of the cheap Chinese speedos don't work on these bikes due to RF interference with the micro-controllers. (See my thread in 'Introduce Yourself') :


    The Cateye was the first one that worked reliably for me.
    ... Steve
    Last edited: Nov 7, 2009
  6. AussieJester

    AussieJester Member

    totally agree was just mentioning it buddy, i have been using a cheapo brand myself and question its accuracy, i now have a Drain Brain for my new build that measures all the battery info it also has a speedo incorporated hopefully this will be more accurate than my cheapo speedo...The one i have is also wireless & drops out periodically i think due to the length of the fork i use, on old bike with shorter fork it was flawless...

  7. AussieSteve

    AussieSteve Active Member

    I'd say that even the cheapies would usually be fairly accurate if set up correctly, since the micro program only needs to count turns and add the circumference each time, unlike analogue (mechanical and electric) speedos.
    Having said that, the cheap one on my electric bike recorded a max speed of 85.7 kph one day, on a road with heaps of bad potholes. That could only have been caused by chatter of the reed switch sensor. If a power supply was available, a hall effect sensor would be more reliable, like those used for CDI triggering on most small engines.
    ... Steve
  8. happycheapskate

    happycheapskate Active Member

    I thought it was best to use an actual tape measure on an inflated tire&wheel. I tape the 1" mark to the tire, all the way around with cheap electrical tape, then turn the wheel to measure all the way around the center tread, pulling it taut. Works for me. All the speedos I've owned let you enter in mm for example 2033.

    If your speedo bugs out, check the battery with a test meter and replace if necessary, then check your magnet and sensor set up. (or do that first if its obviously moved.)
  9. AussieSteve

    AussieSteve Active Member

    Hello happycheapskate,
    It might work for you, but it's not accurate. The rolling diameter under load is noticeably smaller than unladen. As I mentioned earlier, pressure makes a major difference, too. (Do some tests - I did.)

    Both of mine work in cm, a Cateye and a cheap Chinese one. That's irrelevant, though.

    ... Steve
  10. happycheapskate

    happycheapskate Active Member

    If the circumference of the tire is 2033, and you enter 2033 on a cycle computer asking for the circumference, how is that inaccurate? How would it be less accurate than measuring wheel roll-out unless you have some really knobby tires, perhaps?
  11. happycheapskate

    happycheapskate Active Member

    I understand the diameter will change significantly under load, but if you are measuring the actual circumference of the inflated wheel&tire, esp on a slick, that is your circumference input, and you don't need to measure diameter and calculate circumference.
  12. AussieSteve

    AussieSteve Active Member

    You're almost there.
    When you sit on the bike, you would agree that the tyre compresses?
    Therefore, under that compression, the diameter of the wheel is smaller - yeah?
    Right - the (true) circumference is the diameter multiplied by pi - correct?
    Measure the radius under load, (axle centre to ground) and multiply by 2pi.
    Or the exact same thing is to sit on the bike and roll forward one turn with your full weight on the bike.
    If you still can't get your head around it, do the tests that I suggested, with a couple of different tyre pressures and you'll see what I mean.
    I could spend another hour typing trying to explain this or you could spend 10 minutes testing.
    ... Steve
  13. happycheapskate

    happycheapskate Active Member

    I've got you beat by a mile!

    The tape measure sits ON the tread of the tire, all the way around. I don't care $ to doughnuts what the diameter is. I measure from point A all the way around the center tread of the tire to point A. This is the same as the cumbersome method of "rolling out" the tire distance, but faster and more reliable. The bike is upside down. The tape measure measures the tire, like you'd measure your belt. Hint: you don't need the chart that came with the kit.

    If you know these measurements you can write them on your computer or a card in your saddle bag, and switch the computer to either measure when you trade tires for winter, unless you are using the meter for total miles ridden ever on an engine. I'd rather use an hour meter if I can find one that runs on 6v.

    I hope you don't have to figure out your diameter to get your pants belt size. :)

    Last edited: Nov 29, 2009
  14. AussieJester

    AussieJester Member

    Some people just don't want to listen/learn im afraid AussieSteve ...the whole tire compression and the math has seemed to baffle the poor chap :-S

  15. AussieSteve

    AussieSteve Active Member

    Hello Kim - yeah. It only would have taken a couple of minutes to verify what I said. Enough for me. (After all, he's got me beat by a mile.)

    ... Steve
  16. loquin

    loquin Active Member

    Correct. Directly measuring the circumference of the tire is definitely inaccurate. With any tire. Even if your tire pressure is extremely high, the error will be measurable - the wheel will move less than your direct measurement of circumference would suggest, and, the lower the tire pressure, the greater the error.

    The technique is dead simple. Mark the floor, place the tire atop the mark, with the valve stem exactly on it. With your weight on the bike, roll the bike forward for one tire complete rotation, and mark the floor again. Then, directly measure the distance between the two marks. Dead simple. Not cumbersome, and the most accuracy. No calculation required.

    happycheapskate - The calculation was just to show you how much the tire deflection under load actually impacts the accuracy of trying to measure circumference without being under load. Just a quarter-inch tire deflection under load represents nearly 2 percent error using your method. (the bike will be going about 2% slower than your speedo indicates)
    Last edited: Nov 30, 2009
  17. AussieSteve

    AussieSteve Active Member

    Yep - during my tests I measured 5cm, (2"), difference in circumference between an unladen wheel and loaded wheel with low pressure. (Wheel circumference 205cm - 2.4% error)

    By the way, I only weigh 50 kg, (110lbs). A higher rider weight will result in a higher error.

    Pretty good estimate, loquin.

    Last edited: Dec 1, 2009
  18. AussieJester

    AussieJester Member

    happycheapskate might also need help working out HIS belt size, my guess... he walks around with the crotch of his pants around his kness while displaying 23in of underwear .... :smilielol5:

  19. AussieSteve

    AussieSteve Active Member

    Yeah, I already know mine.
    I put a line on the ground, then lay down, rolled once .....

    ... Steve
  20. AussieJester

    AussieJester Member

    I had the sexy wiminez at the store take mine! she put the start of the tape on my belly button and wrapped it around my waist all the way back to my belly button
    Unfortunately she didnt take into consideration my need for breathing,
    when i exhale my pants fall down :biggrin::biggrin::biggrin: