It does but by a tiny amount (not by any amount that would change the rollout much).

I can't believe someone has tested this by now. Njkayaker will be proven correct when one of you tests this.

Bottom line is that the radius is the distance from surface to center. The tire will deform around the contact point to provide for this.

That is my assumption. The casing is meant to be flexible, but not elastic - although certainly not perfect (either by intended design or by limitations in design, i don't know)

Intended design, or we wouldn't bother with pneumatic tires because they wouldn't have any advantages.

Start here: Assume the tire itself doesn't stretch or slip. It deforms, but that's bending, not stretching.

Now, put the bike on the ground and weight it down. The nominal radius below the hub changes.

Fine.

But now assume there's an ant. A magic ant that can walk all the way around the tire, including under the contact patch. This distance around the tire -- because there is no stretching -- is the same, regardless of what shape the deformed tire takes.

As the tire rolls around, constantly deforming, this distance still remains the same, thus the roll out distance doesn't change as a function of weighting.

Does that help?

You need to think outside the circle, my friend...

Those show exactly what I'm saying too.

Here, they use the "tank" analogy too.

http://en.allexperts.com/q/Tires-235...ce-vs-Tire.htm

Here are some real engineers who say the same thing:

http://eng-tips.com/viewthread.cfm?qid=94153


The term for the number is "rolling circumference".

http://www.michelinman.com/glossary/

Hopefully, Michelin is a authoritative enough source!

I have tested it (as I noted in one of my posts). In my tests, using 10 samples of each, the noise from measurement error was greater than the difference between the averages of weighted down and not weighted down. and both were very small, less than 2mm. To test I put a very small dot of whiteout on the tire, and slowly rolled the bike out. Then I measured between the centers of the two dots on the ground. Not super precise, but precise enough for these purposes.

And the radius doesn't matter any more - you don't have a circle anymore!

By the way, the tank example indirectly proves my point. The tread does have a rollout. The roller and tread (assuming no slippage - funny how that keeps coming up) have a fixed relationship - since the roller will rotate x number of times for each time the tread rotates, you can calculate velocity and distance by counting the revolutions of either and multiplying by the appropriate circumference. However, the radius of the tread does not matter - and in fact is a nonsensical concept, since the tread isn't in a circular pattern (technically it will have multiple radii, depending on the curve each part of the tread is describing). The tread is a better analogy for the bicycle tire since 1. it contacts the ground and 2. is not a circle.


JB

But if the flat spot followed the ant, the distance around the tire would be smaller. This is what happens in real life. The circumference of the tire may not change, but the effective circumference does.

Guys, when the tire is weighted, the circumference is decreased. Thickness of the tire in that area is traded for width of the tire in that area. If the tire had a fixed width then you'd be right the circumference wouldn't change since the thickness of the rest of the tire would increase to compensate, but since the tire CAN trade thickness for width, the circumference decreases under load. It's as simple as that.

The ant would measure a smaller circumference.

You've proved that your measurement tools aren't accurate, not that there's no difference. You need a better DOE.

You measured one rotation of the tire. I'm willing to bet if you measured 100 rotations of the tire 10 times you would find a consistent, measurable, repeatable difference.


Edit: Tribble explained it more effectively. Listen to what he said.

It's still incorrect and there are ample sources that indicate that it is incorrect. And he's shown nothing that supports him.

I've done this as well. Also with white out and other marking substances like paint and nail polish. But...

No marking substance left dots, but a Rorschach like smudge, fortunately it repeated the same pattern for each printing on the ground, just a bit fainter each time.

I did not measure between 'dot' centers, but instead from common edge to edge of the smudge on the ground - easy to do as the smudge pattern was consistent.

I also did not measure between two points, but the total distance of about 5 points. This is similar to averaging, but introduces less human measurement error.

Then I repeated.

This all said I didn't write down my results so I can't report. ;)

Go take a mtb tire that is barely inflated and roll it out. Now roll it out with pressure on it. You will see the difference in one rev.

Well, doh, of course. The elasticity is contained/limited by the casing design, but the has to be some!

You have two different circumferences in contact with the ground. One or both of them have to slip.

All I can tell you is that when building machines that pulled a plastic film down a line, how much we squeezed the rollers together that the film went between changed the feed rate of the film. Film going through rollers is no different than the ground going by.

Not true. If there is point of no slippage (say, directly beneath the hub), than the change in distance along the circumferential direction of the tire has to equal the change in linear distance on the ground.

Take a hula hoop. Squash it until it's a suitable ellipse. The distance around the outside edge is still the same, and will roll out exactly the same as unladen (feeling like a Monty Python reference, there).

Again, assuming no compression in the circumferential direction (which njkayak claims is an incorrect assumption, but I'm not sure about).

I'll give you a better example. Pretend your tire is wood. Pretend when you get on, the flat that is created on a rubber tire would be like cutting a flat on the wood tire. Now, as you roll out that wood tire, you have to cut that flat all the way around. If you did this, your wood tire just got smaller. Now, imagine if you could glue those flats back on after leaving contact. Rubber tire.

No, the deformation of the tire handles that. A point on the tire doesn't take a circular path: it rides a "squarish" path, from front to rear, as the tire rolls. (We don't care what the distance of the square path is.)

It sort of looks like this:

\0/

The rolling circumference is the circle in the cup and the tire circumference is larger.

The tracks on a tank illustrate this.

(If there was appreciable slippage, the labels on the tire would move from their position over the valve stems.)

This is a horrible example. Cutting removes material, and is more akin to compression.

The tire bulges (in circumference) elsewhere.

What the hell is the circumferential direction? New standard for notation for this thread. Either radial, tangential, or longitudinal direction. Paint sketch indicating these to follow.

Attachment 166592

Crude notation sketch.

Yes, the tire bulges elsewhere, but that's irrelevant for this discussion. The tire also bulges in the longitudinal direction.

The contact patch gets both longer and wider. I doesn't necessarily follow that the cirdumference just decreased, since more of it is in contact with the road.

JB