Bounce Test for Rolling Resistance?

[Ernie Rogers]

Hello, folks,

Let's talk about an idea that came up once before but didn't get any discussion. The idea of measuring the rolling resistance coefficient by just bouncing the tire.

The rolling resistance coefficient (call it Crr) is closely related to the coefficient of restitution (call it CR). You get the coefficient of restitution by simply bouncing a tire--

CR = (bounce height) / (drop distance)

The quantities Crr and CR should be related by something close to this--

Crr = K (1-CR)

K is a "constant" that depends on the conditions--

How high you drop the tire
How much weight is dropped
Tire pressure and tire load

With some study, we should be able to determine the right test conditions and the value of K.

A bounce test might be something that can be done more easily than buying a rolling resistance test machine, and easier than doing a coast-down test, or using a ramp.

Additional point: A bounce test might also work for testing the efficiency of road surfaces.

I think I will pass this on to some teachers for possible thesis topics or a science fair project. You might pass the idea along if you think it has merit. There is probably an ASTM committee for Crr, and they might take an interest.

What do you think?

Ernie Rogers

[NeilBlanchard]

Hi Ernie,

I seem to remember a reference by someone at Aptera (Steve Fambro?) who spoke about a homemade jig they made with a wheelbarrow-like cart and a ramp, and they measured the coasting distance.

[CapriRacer]

I'm wondering:

1) Since inflation pressure is a major factor in the spring rate of a tire, and since tires grow (and shrink) based on that pressure, that setting the pressure accurately enough might be a problem!

2) Since most of the rolling resistance inherent in a tire itself is a function of the tread compound, would differences in the weight of the other components adversely affect the accuracy of the result? Put another way: Would the presence (or absence) of rim flange protectors have an effect on the results?

[Ernie Rogers]

Originally Posted by CapriRacer

I'm wondering:

1) Since inflation pressure is a major factor in the spring rate of a tire, and since tires grow (and shrink) based on that pressure, that setting the pressure accurately enough might be a problem!

2) Since most of the rolling resistance inherent in a tire itself is a function of the tread compound, would differences in the weight of the other components adversely affect the accuracy of the result? Put another way: Would the presence (or absence) of rim flange protectors have an effect on the results?

Hello, Racer, you might have to explain your questions--

I don't get the first one. The tire would be mounted on a wheel and set to the desired pressure. Maybe you are suggesting there is a change in pressure between before and after mounting on a car? Such a change could be corrected for.

Second question, first part--Is it true that tread area dominates rolling resistance? That suggests that tread design is also an important factor, if true. Second part--what is a rim flange protector?

Ernie

[Piwoslaw]

Hmm, so if Crr depends (among other things) on the weight being dropped, then should we drop-test just tires with rims, or tires plus 1/4 of the weight of a car?

[dcb]

It seems you can push the car on level ground with a bathroom scale and the force to maintain a speed is the one you want:
Puma Race Engines Technical Guide - How engine power and top speed are related

That will encompass part of the driveline, any brake drag bearing drag, etc. You might even want to put someone in the drivers seat and hold the clutch in for a more realistic number.

Sorry if this is obvious info.

[CapriRacer]

Originally Posted by Ernie Rogers

Hello, Racer, you might have to explain your questions--

I don't get the first one. The tire would be mounted on a wheel and set to the desired pressure. Maybe you are suggesting there is a change in pressure between before and after mounting on a car? Such a change could be corrected for.

If you were to measure a tire immediately after inflating it - and then measure it 24 hours later, you'd find the tire would be physically bigger. I think there is even an ASTM test procedure that covers this - and we refer to the dimensions as "24 hour dimensions" to indicate these are "grown" dimensions.

I don't know why I just thought of this, but it is common in most (if not all)tire testing to wait 24 hours after the tire is mounted to conduct the test to account for this growth.

But I guess where I was just brainstorming on this was: It is common for those of us who have done tire testing, to add a few extra psi so that the pressure after growth is correct without adjustment.

Plus I would think that since a half a psi is considered by many to affect the spring rate enough to be felt in racecars, then the pressure for any test where the spring rate becomes important, would also need to be precisely set. For example, tire force and moment testing was always done with a pressure gauge that measured to the nearest 0.1 psi.

So I was wondering - out loud - if you take a tire and repeatedly test it, would small increments in pressure result in large enough differences to make discerning differences in tires difficult.

Originally Posted by Ernie Rogers

.......

Second question, first part--Is it true that tread area dominates rolling resistance?

..........

Yes.

Originally Posted by Ernie Rogers

.........
That suggests that tread design is also an important factor, if true.

Only in the sense that the volume of rubber is part of the "design". I'm sure that tread patterns with many small elements (like a winter tire) would generate more movement - and therefore more RR - but my sense is this movement is only of minor significance compared to what effect the tread compound has overall.

Originally Posted by Ernie Rogers

........

Second part--what is a rim flange protector?

Ernie

This:



It is an projection out from the sidewall that protects the rim flange from hitting the curb (or whatever). It adds quite a bit of weight to the tire, but hardly affects the RR at all! But I wonder if dropping a tire from a height might engage enough of the sidewall to affect the results.

[bgd73]

tires are quite tricky. the 24 hour dimensions, great thought. It is nice to see a reality written. temp changes of five degrees, angle of weight on tire and suspension etc, it makes it more difficult to nail down a real number for any except ones own car and result. there is also directional tread versus non. I have no excuses for non-directional treead, but ya know a majority of drivetrains are mounted sideways pretending itself balanced. alot of factors....

[dcb]

I'm a little lost at the application here.

Are we just looking at the crr of the tire?

Are we hoping to be able to go into a tire store and locate the best tires by bouncing them?

Are we looking to measure the crr of tires on a vehicle? (and what about the rest of the drivetrain?)

It is an interesting exercise to bounce a tire, have to get it square on every bounce and growth, temperature, etc. But it does seem more practical to just see how much force it takes to move a vehicle that has just been driven around to warm everything up.

Is there an intended audience of the tire bounce test? Sorry if I'm not getting it, just not sure what problem it solves.

[Ernie Rogers]

Hello, DC,

Yes, my idea was to find a quick and dirty way to get a number for Crr. I know it's not a great idea, just thought I might entice someone to work through the details on it.

So, the tire would have to be mounted and pressurized by a presently undefined procedure, and bounced by itself, not on a car.

It wouldn't be hard to simulate the same deflection as seen when rolling on a car, but I haven't bothered to do the math.

I saw a rule of thumb here somewhere, and failed to write it down (my memory says this, I could be hallucinating). The rule-- "a one psi increase in tire pressure reduces rolling resistance by ___" Did somebody say something like that?

Otherwise, I would propose this approximate rule--

"Tire rolling resistance varies inversely with tire pressure." It should be more valid toward higher pressures.

Now, the question was asked, could we detect the change in rolling resistance, say from a 1 psi, or maybe 0.1 psi, pressure change, in terms of the effect on fuel economy?

I'll plug that into a fuel economy program, say for a Prius, and see--

Tires at 40 psi, Crr = 0.009...........49.15 mpg at 60 mph
Tires at 41 psi, Crr = 0.00878........49.59 mpg at 60 mph.

The difference is 0.44 mpg.

In my experience, this 0.9% change is virtually undetectable in the real world where wind cannot be eliminated from tests. (But I could detect a 2 psi change.) A 0.1 psi pressure difference would be truely insignificant.

Could the difference of-- .009 /.00878 = 1.025-- 2.5% in Crr
be detected with a wheel tire tester? Maybe just barely. Any thoughts from the real world experts here?

Ernie Rogers

Originally Posted by dcb

I'm a little lost at the application here.

Are we just looking at the crr of the tire?

Are we hoping to be able to go into a tire store and locate the best tires by bouncing them?

Are we looking to measure the crr of tires on a vehicle? (and what about the rest of the drivetrain?)

It is an interesting exercise to bounce a tire, have to get it square on every bounce and growth, temperature, etc. But it does seem more practical to just see how much force it takes to move a vehicle that has just been driven around to warm everything up.

Is there an intended audience of the tire bounce test? Sorry if I'm not getting it, just not sure what problem it solves.