I'm gonna dump on "poor" jonathanb715 a bit more to make some points which I think are important to make these kinds of discussions successful.
Originally Posted by
jonathanb715
Ok, I made a mistake in my assumption, and then compounded it by trying to extend a 2 dimensional model to a complex multidimensional reality. For auto tires, it turns out that there is a derivation and experimental data to back it up (quoted from an engineering forum:
http://eng-tips.com/viewthread.cfm?qid=94153):
That is a reference
I provided. If you didn't realize that, it means you ignored people's counter argument.
Note that for this argument, it would have been very strange if this issue had not come up before. The fact that you did not provide any references to support your position is telling.
If you are going to have these kinds of arguments, providing references is important to do.
Originally Posted by
jonathanb715
I still think it isn't large enough to get into saying you have to do a rollout weighted down or else it won't be accurate - the error in the measurement technique most of us will be using will wash it out, so neither method should improve accuracy necessarily.
This is basically "waving your hands and saying it doesn't matter". I said earlier that the difference was going to be on the order of 1%.
If you are going to have an argument about a theoretical issue, it's important to keep issues of measurement errors out of the discussion.
Originally Posted by
jonathanb715
I'd just like to point out that my difference in averages was very small - well under 1% (2mm), and the difference in samples was somewhat larger (+-3mm). Even adding the difference in between the largest outliers gives you a change in rolliout of less than 0.5% - still not real reliable, but probably accurate and definitely not worth worrying about when setting your computer. Yes, more samples would make this a more valid data set. FWIW, I'm the only one who bothered to post any data at all.
The "theory"
predicts a difference on the order of 1%. If your measurements can't reliably show that difference, then your measurement technique is not sensitive enough!
Originally Posted by
jonathanb715
Originally Posted by njkayaker
I don't think the tire stretches much at all. The point of measuring a deflated tire is to illustrate the effect.
They both result in defomation, but that doesn't mean the effect is identical.
JB
No, it seems
more reasonable to expect that a deflated tire defines the lower bound for the effect of reduced pressure on the "rolling circumference". The effect doesn't have to be "identical": it just has to be "close enough".
Your mistake here was not indicating what you thought the effect would be. You rejected it out-of-hand with "that doesn't mean the effect is identical", which might be true but might not be! It's possible that the effect is "identical" enough and you provided nothing to support your rejection of it's relevance.
You rejected it not because you had a reason (you didn't provide any) but because you could not conceive of being wrong!
Originally Posted by
jonathanb715
Sigh. When measuring circumference or rollout, you are measuring from the same point on the outside of the tire to the exact same point. That point does have to cover that flat part as the tire rolls through, then it lifts off the ground and will follow the circle until it comes in contact again at the beginning of the flat section - but you don't measure rollout when it gets there there, you measure it at the center point , where the hub is closest to that point (assuming that's where you started). If you insist on using circular measurements (radius) for non-circular shapes, you will get results like this.
The arrogance here (and in other posts) is noticeable. If you want to be arrogant, don't ever be wrong!
The only place where what the "point" does is smack dab in the middle of the contact patch directly under the hub. What the point does in any other place has no effect on the "rolling circumference" measurement. (The point could take a trip to France and it would not matter).
It's important to understand that the point in the center of the contact patch directly under the hub is a point on a circle of a rigid (nondeformable) wheel, the circumference is less than the tire circumference. The circumference of that virtual wheel is the circle whose radius is the height of the center of the hub from the ground.
Originally Posted by
jonathanb715
Think back to the tank tread. Does it have a circumference? Yup. Can you make it into a circle? Pretty close, as long as the plates are small and the hinges between them are flexible. Does that change it's circumference? Nope, unless the tread shrinks or stretches in the process. In that respect the tread acts a lot like a bicycle chain.
You have to be very careful with the tank tread analogy. You did not read my post, where I brought it up, very carefully.
The problem with the tank tread analogy is that you don't have a hub to use that matches the "rotation" of the tread. All the available hubs are too small (this is why I used the analogy). The
reason you would not use the tread is because you need a hub to put the magnet on! If you pick a roller on the ground (with the tread in between), you could put the magnet there and input pi*2*r into the computer (r is the height of the hub axis). The length of the tread doesn't matter (and is an
unusable measurement).
===============
Note that I was wrong in saying that "circumference" only applies to circles (but that's a minor thing). It still seems weird to me to apply it to non-circles. Note that I never disagreed that the circumference of the tire is constant (that seems obvious to me).