simonplatt

whilst watching the distance the last figure is unreadable. i want to set it to 1 decimal point.

motorthings

you might want to search online for the manual pdf for the computer you have, to see if it tells how to change the decimal places. if it's not in the manual, contact the manufacturer.

acidfast7

That's some Teutonic precision right there

What brand is it?

Moe Zhoost

Precision does not mean accuracy. You can't really calibrate a cyclometer to be that accurate.

acidfast7

Why not?

Anyways ... I was just tooling home with the TAGES KM function on and with only two digits after the decimal point (km) the final digit was moving quite rapidly. So, I must assume that the OP is correct and that the final digit would be totally worthless.

wphamilton

use black vinyl tape over the unnecessary 2 digits. Or pre-set the distance to something over 1,000 miles and the extra digits go away.

locolobo13 

Ignore the last digit while riding. If you record the mileages round them.

Alternatively there are plenty of bicycle computers that show only one decimal place. Buy one of those.

ItsJustMe

Unless they have some unusual sensor on the wheel, it only gets one pulse per revolution. One revolution on a 700c wheel is about 2100mm or about 7 feet. 3 digits means that last digit is measuring a distance of 5280/1000 = 5.3 feet.

You're measuring 5 foot distances with a 7 foot stick. It's going to be jittery at best. on subsequent revolutions it'll go 7,14,21,28,35, 42,49,56 feet, which will translate to:

0.001, 0.002, 0.003,0.005,0.006, 0.007,0.009, etc

You have a cumulative error that's a significant percentage of the indicated value, so either that last digit is unneeded (which is probably the case, in which case why show it) or it's inaccurate (usually off by a large part of its indicated value) in which case why show it)

acidfast7

My bike computer requires the total mm of the circumference the wheel + tyre. As ISO chart is included is the user is lazy but one can also use a piece of string, which I did and I think it came out to 2096mm. When I change tyres, I recalibrate. I also compared the tyre that I got about 4000km out of an measured a 2mm difference in circumference (2 / 2096 = 0.95%), which I'm OK with.

https://202.215.251.86/data/resources...e_chart_v2.pdf

So I don't really understand this whole measuring a 7-foot length with a 5-foot stick, as the computer will also keep track of fractional units as well.

ItsJustMe

It's not that you can't calibrate the computer to be accurate overall. it's that the display is showing a precision that far outstrips the granularity of the measurement.

To take it to an extreme - Let's say you have the same display, showing 1/1000th of a mile, and it only clicks every 800 feet. Clearly the last few digits are completely useless, because they're basically never right.

Having a display to the 1/1000th mile is IMPLYING a precision that simply is not there. In the actual example, you see that it misses the 0.004 mile marker. Because it is trying to display a precision that is not supported by its measurement method, it displays 0.003 past the point where it SHOULD be saying 0.004 and never shows it, skipping straight to 0.005

It's displaying pointless precision in other words. There's simply no point to having it there. You can't ever really know if that last digit is correct or not, so there's no point in showing it.

GamblerGORD53

You can have my ill bought never used stupid one decimal Bontrager speedo. OK for cars, but for bikes?? Completely useless and impossible to calibrate. You can barely see 1/10 mile down the road. 2 decimals is perfect.

acidfast7

Thanks- Now I see what you are saying.

locolobo13 

Another question about that. How many times does the magnet have to pass the sensor before the computer registers 1 revolution? It seems like mine is 2 or maybe 3 times. That is going to effect the accuracy as well. I suppose the 2nd time would be the most accurate.

GAJett

Wanting high accuracy and precision? String won't cut it 😀. Doesn't account for weight on the tire which reduces the radius, hence the diameter. Best to do a "roll-out test", where you weight the tire with the expected mass, then measure the distance of one or more revolutions.
But you still haven't accounted for varying weights or tire pressures! Those will also impact your accuracy and precision to some (probably trivial) degree.
Specific discussions of precision were addressed above. AT BEST, with a single sensor (magnet), your precision is +/- 1/2 the diameter and is theoretically CUMULATIVE (+ one time, - the next and random, so it can't be calculated)!
If you wish to increase your precision you will need to add more sensors. Two sensors, you're at 1/4 the diameter. What the heck! Put a sensor on each spoke, you're now at 1/64th the diameter for a 36-spoke wheel😱. (But at what increase in rotating mass?)
Yes, I know, reductio ad absurdum. Two digits is more than enough for me. But that's just me. YMMV. Just ride. /<snark>
Cheers!