Cyclometer Tire Size

[GEOlson]

I did some reading/searching of tire size for computer input. I use a Sigma Sport 1606L on my bike and have always used the chart in the manual. For my tires, Continental Ultra Race, 700C x 25 I used 2146. Yesterday I did my commute from work to home on Ultra Sport 700C x 28 and I changed the wheel size setting to 2149 as the manual says. When I got home I saw that my distance, which was 12.21 miles on 25mm tires was about 1/4 mi. short. So obviously my 28mm tires are not 3 mm bigger in circumference. I attempted to roll out the wheel and measure but I need a better methodology. However, what would be the best way to convert the difference in measurement? The reason I particularly care is because I had a heck of a tail wind yesterday and was very close to beating my commute record.

[Derailed]

The distance your computer measures is just the product of the number of wheel revolutions it counts and the circumference you input (also multiplied by a conversion constant to get from mm to miles because you're using different units):

d = n x C x const. (where d is measure distance, n is number of revolutions, C is circumference, and const. is the constant to convert the units of your circumference measurement to the units you want for your distance measurement)

If you trust the distance of 12.21 miles measured with your 700x25 Ultra Race tires, and you want to make your measurement with the 700 x 28 Ultra Sport tires get the same reading, you just need to adjust the circumference measurement up by the same percentage that you're short in distance:

Again, assuming you trust the 12.21 mi measurement, and your Ultra Sport measurement was exactly 1/4 mi short, it is short by 0.25 / 12.21 = 2.048%.

So, you just need to increase the circumference value for your 700 x 28 Ultra Sports by this same percentage: 2149 x 1.02048 = 2193.

I do note, however, that that's pretty big for a 700 x 28 tire!

Personally, I prefer to do my own roll-out measurement rather than trust the included tables (although, honestly, the tables usually work pretty well). I use a 75 foot tape measure and try to measure seven or eight revolutions with me sitting on the bike to spread the inevitable measurement uncertainties out as much as possible. And, to make the calculation super easy, I use a Google spreadsheet set up so I just enter the total distance and number of revolutions, in feet and inces, and the spreadsheet displays the circumference in millimeters.

[ItsJustMe]

I put my hiking GPS on the bars and measure it that way, then do the correction with the obvious simple math. I only bothered with this once because I really don't care that much about accuracy, it's pretty close with the default settings.

If you just want the new wheel measurement to match the old one, just do the math. If with a setting of 2149 it's reading 11.96 and you expect it to be 12.21 (1/4 mile short) then you should change it to 2149 * 12.21/11.96 = 2194 - that should make the same route with those tires read 12.46 miles.

(Edited because I misunderstood the description the first time)

[GEOlson]

Thanks guys, I will roll out the wheel today and see how it compares. I thought I had about 84 3/16" yesterday, but it's a litter tricky telling when the stem is perfectly vertical.

[Litespeedlouie]

FWIW, I found my Sigma 906 table to be pretty different from my measured rollout.

[Derailed]

Thanks guys, I will roll out the wheel today and see how it compares. I thought I had about 84 3/16" yesterday, but it's a litter tricky telling when the stem is perfectly vertical.

That's why, ideally, you'd use a long tape measure and observe the distance for several revolutions (and then divide by the number of revolutions).

Have fun!

[Werkin]

First off Continental tires are not true to nominal size. Ultra Sports always measured >2mm narrower on my 20mm rims, some of their models are further off, GP 4 Season for instance. Which brings up another possible cause for discrepancies, narrow rims increase diameter, wide rims decrease diameter, for the same nominal tire width and air pressure. Also, a parabolic shaped tire like the Maxxis Re-Fuse, and tires that have a thick puncture resistant strip down the center, will have wonky diameters compared to nominal sizing. Using the roll & average method, and manually inputing the circumference increases accuracy.

[pdlamb]

I'll typically set my tire size to the manual, then climb to the top of a nearby ridge and coast down. This road has (pretty good) mile markers, so I'll calculate the difference for a mile and correct the calibration. After that I'm usually within 1% until the tire wears and gets replaced.

Why coast? I ride much straighter coasting than pedaling!

[seeker333]

That's why, ideally, you'd use a long tape measure and observe the distance for several revolutions...

There's two very faded 100 feet marks painted on the pavement in front of my house, measured with a 100' tape.

I put masking tape on the front rim for a visual index at the 6:00 position, aligned with first paint mark, roll out the 100 feet distance counting revolutions, stop on second mark, dismount and calculate last fractional rotation via spokes, i.e. 17/32 = 0.53.

Make sure tires are inflated to pressure you intend to always use, and it's helpful if you measure on a slight incline so that you can coast in a perfectly straight line. Pedaling can make you swerve a little bit. Following a 100' painted line in an empty parking lot might help, but I've gotten good results without it.

This method works very well. I consistently get agreement with my GPS-equipped buddies over a 35-45 mile ride within +/-<1%.

I could perhaps improve my accuracy by rolling out 200 or 300 feet, but i think 99.X% accuracy is good enough for this purpose.

[GEOlson]

It turns out the 28mm tires weren't that far off, 0.07 mi, from the 25mm, but I'm still going to try rolling out and when I ride tomorrow I will bring my GPS. In the past my 25s seemed pretty good with the GPS, but I didn't expect a difference in changing the tires.

[GEOlson]

I rolled out 10 tire lengths and got about 69.5 feet (I think - my tape is double sided, it may have been 69' 5") I will do a short ride to the end of town and back with a GPS, too and see how it compares, it's a little over 3.5 miles. I will report back. :-)

[GEOlson]

I got 3.61 on the bike, I rolled in into the garage so it would turn over that last hundredth, and the GPS I'm estimating was at 3.555, it seemed to turn over to 3.56 a little after I removed the bike's computer. I will calculate it shortly, I just needed somewhere to write it down since I can't find a pen. I also noticed my cadence sensor wasn't working, so I'll fix that, too.

[yosarian9]

Could you not use a fabric tape measure like for cutting fabric / sewing clothes and wrap it around the tire?

tape_measure.jpg

[GEOlson]

I used a long tape like that, 100 ft. The only other fabric tape I had is only 5 or 6 ft long, I think. But here's what I did with the GPS log.

The cyclometer showed 3.61 miles which is 19,060.8 feet and the current wheel size was set at 2149 millimeters which is 7.050525 feet.
The GPS showed a distance of 3.555 miles which is 18,770.4 feet. In order to get how many times when wheel turned I divided the cyclometer distance in feet by the circumference setting: 19,060.8 ft. / 7.05025 ft. = 2,703.45 wheel revolutions.

Since the GPS said it was actually 3.555 miles, of 18,770.4 ft., I divided that distance by the number of times the wheel turned: 18,770.4 ft. / 2,703.45 revolutions = 6.098 ft. wheel circumference.

6.098 ft. x 25.4 mm. = 2127.579 mm.

So that should be a nearly true length for the circumference. And now my speeds and distances should be nearly spot on. Of course GPS can only estimate, but it's the best measurement tool I own that can measure a long distance (anything over 100 feet).

[ZippyThePinhead]

I'm not an expert but a couple of observations:

1) Tire size varies by manufacturer. You can take a couple of 700x25c tires-- say, Gatorskins and Marathon Pluses-- mount them, and measure the circumference. It won't be the same. But the instructions that come with your cyclocomputer will give a single value for the circumference of all tires in a particular size. For example, my Cateye instructions say to use 2105 mm for a 700x25c tire.

2) Cyclocomputers basically count the number of circumferences you cover as you ride. As the circumference of your tire changes, the distance recorded by your cyclocomputer will also change slightly.

Let me give you an example. I mounted a new set of 700x25c Gatorskins last year. I measured the circumference, and it was quite different than the number my Cateye computer instructions said to use. This made me curious. Why would the number supplied by Cateye (2105 mm) vary so considerably from what I measured (2155 mm, more than 2% difference)? To make a long story short, I calculated an effective circumference based on the radius of the wheel when loaded and unloaded (i.e., with me off the bike and on the bike). I found the difference between the unloaded radius and the loaded radius was about 4-5 mm, which meant the unloaded circumference overestimated the loaded, effective circumference by roughly 1.5%. After 3000 miles, the rear tire (where my magnet for the cyclocomputer sensor is) is worn and pretty well squared off. The circumference has diminished slightly as the tire has worn. The cyclocomputer is still set using the effective circumference of the unworn tire, however, which means my computer is not on the money.

What I found:

New Gatorskin, unloaded circumference: 2155 mm; loaded effective circumference: 2124 mm

Gatorskin, with 3200 miles of wear, unloaded circumference: 2140 mm; loaded effective circumference: 2109 mm

In both cases, the unloaded circumference is around 1.45% larger than the loaded circumference.

Interpolating, I figure I should be setting my cyclocomputer to use an effective circumference of (2124+2109)/2=2116 mm, if I plan to set it and forget it. This will cause my speed and mileage to be overestimated when the tire is new and underestimated when the tire is worn, but it should average out by the time the tire is replaced, assuming the tire achieves an average lifespan.

[Giro]

Derailed's post above succinctly describes the late, great Sheldon Brown's method for precise calibration. If you ride a measured course, all you do is go for a bike ride and no fiddling around with measuring tape, riding in a straight line, etc.

Sheldon Brown's Cyclocomputers in general

Measured course method.

His site is a good one for most bicycle-related information.

Both Google Maps and Bing Maps (and other free online maps) will give you a measured course. For a bit more accuracy, have the map give you the measurement in kilometers rather than miles so there is less rounding off of the distance to the nearest tenth of a unit (since kilometer is about 0.6 miles, rounding to the tenth gives the measurement to about 0.06 miles or +/- 0.03 miles. 0.03 * 5280 feet = 158 feet. One other advantage of this method is if you use the online map to plan bike trips, the measurements to turns etc. should match-up well provided tire pressure and bike load is about the same.

[GEOlson]

Maps are a specialty of mine. :-) My favorite method for measuring distance would be to use a polar planimeter to measure out the scaled length, but the problem with most maps is that they are 2 dimensional and relief can easily create distance inconsistencies because vertical change isn't considered. Of course it wouldn't matter for intervals like miles but for hundredths I think hills could influence a difference between cyclometer and online maps. But I don't know if G! Or Bing measures 3D, I've heard map my run/ride does.