Happy Feet

I'm not a math whiz so I though I'd ask the question.

I have a bike with two wheelsets and currently only one computer. Just basic functions, distance, speed, time.
I am trying to decide whether I should buy a different/second computer for each wheelset or whether the difference in sizing will be negligible enough not to bother. I don't need super great accuracy, within a Km over 100km or so would be acceptable. I am not a data nerd.

The wheels (according to the charts):
1. 700cx32mm for road - 2155mm circumference.
2. 650bx42 for gravel - 2100mm circumference.

The computer is currently set for the road set, 2155mm. How much will it differ if used for the gravel set?

Thanks in advance.

Sy Reene

This is an interesting question, but waging not an easy or quick answer.

RChung

2.5%

If you don't want to change the wheel circumference setting every time you switch, set it for 2127mm. That will make each off by about 1%.

Happy Feet

I didn't think of that.
And, just to be sure, that would mean about 1 km/100km deviation, yes? And less than 1/2kph if doing 50kph.
I meant it when I said I wasn't a math mensa.

Sy Reene

Are the tables assuming a specific inflation/tire drop percentage?

Happy Feet

I don't think so. They are just the charts the computer companies list to set the computer.

Cyclist0108

2.6%

79pmooney

There's a non-math way to compare any two wheels. (Old as the hills. It's called a roll-out.) Get your bike. Put it on any flat surface with the tire valve worn down . Put a piece of tape on the surface at the valve. Now roll the bike until the valve comes down again. Measure valve to valve. Repeat for your other wheels. The percent difference is your percent difference in recorded speed.

(Edit: this takes into account that tires are often not actually "42c" as labeled on the sidewall.)

Bmach

What do you have for a computer? I have two and both allow you to setup multiple bike profiles. So if yours can have different profiles just do that.

Happy Feet

That's the decision. My computer is a one bike model. If I bought a new one it would be a two bike model.

Happy Feet

Those distances are already known. 2155mm for the 700's, 2100 for the 650b's. The question is will the 55mm difference make that much difference data wise, in real life use.
I like the idea of splitting the difference between the two sizes. That means there is a 27.5mm difference between what the computer reads and what is actually being covered.

----------------------------------------------------------------------------------------------------------------------------------------------------

I did some math using those numbers converted to cm's.
I think...

One revolution is 2100mm = 210cm's.
1km = 100000cm's.
100000/210 = 476 revolutions.
27.5mm = 2.7cm's deviation per revolution.
476 revolutions x 2.75cm's = 1285cm's
1285cm's = 12.85m

So... if my thinking is correct, I would have a discrepancy of 12.85 metres for every 1 km travelled.
or.. 12.85 meters x 100km's = 1.285 km's per 100 km's traveled.

If that is correct, would it also stand to reason that if I were travelling at 25km's per hour I would have a speed discrepancy of .32 km's per hour?
1.285 meters per 100km's / 4 = .32.

alo

If you are really concerned about being accurate, this is the way to do it. Wheels may not always be the same as the theoretical. Then as the tire wears, it will change a little. Not enough to be concerned about.

alo

If the numbers you gave in the first post are correct, one is approximately 2.6% different to the other. So at 38 kph it is out by 1 kph.

I would just leave it as it is, and allow for the difference with the other wheel in my head.

WhyFi

According to your first post, it's assumed, not known.

Happy Feet

Well yes, technically I see the point. It's a basic bicycle computer though so I am going by the charts. When you are talking millimeters one can also be off in the the roll out measurement.
I think setting the computer half way between the two measurements will do for now, for the way I use such data, until I get a different computer with two settings. Then I'll delegate this one to a different bike.

I'm not really into stats and only use it as a rough guide when riding. On tours in my region turn offs or crossroads are pretty far apart so I only need to know approximate distances, km's not meters.

Litespud

2155/2100 = 1.026. That’s the conversion factor. If you leave the computer set for the 700c wheelset, your actual speed when using the 650b wheelset will be the computer readout divided by 1.026 (ballparking, computer reads 40 kph, actual speed will be 39 kph). Similarly for distance travelled - when the computer tells you you’ve travelled 100 km, you will have travelled an actual 97.5 km (ie. 100/1.026). As you can see, the differences are inconsequential, and both possibly within the margin of error when using what are likely estimates of tire circumference to calibrate the computer. The numbers you cite (2155 mm and 2100 mm) likely don’t account for variances in tire manufacture, inflation pressure or rider/bike weight, all of which will affect the distance the wheel covers on one full revolution, the distance that the 2155 mm and 2100 mm are supposed to represent. Which is why some folks are recommending that you take an actual measurement.

Happy Feet

Point understood