Tire width impact on fuel economy / MPG

[Ecky]

Originally Posted by CapriRacer

Ah ..... not exactly. All standard load passenger car tires (the most common type) have a rated pressure of 35 psi, except for those tires who use the metric system, then it's 2.5 bar (250 kPa, ~36 psi).

Please note: This is different than the max usage pressure which is sometimes written on the sidewall (and the other times what is written there is the rated pressure.)

This is different than bicycle tires because bicycle tires assume a single rider and a relatively narrow range of weight, while a car might have many passengers and a wide range of weight.

The tires aren't rated at 35 PSI, to my knowledge. The placard in the door says ~35 (give or take) because that's generally the correct pressure for the stock tire size that comes with the car. If you upsize or downsize, this pressure is no longer correct.

[CapriRacer]

Originally Posted by Ecky

The tires aren't rated at 35 PSI, to my knowledge. .......

Look at my webpage on the subject: Barry's Tire Tech: Load Tables

Notice that the load table indicates that the max load occurs at 35 psi. We tire engineers called that the rated load - and the corresponding pressure is called the rated pressure.

While I only published one page from an older TRA yearbook, the other pages are similar, regardless of year. In other words, these things don't change over time. If you go to other tire yearbooks - like ETRTO or JATMA, you will see something similar.

And, yes, if you change tire size on your car, that might change what pressure might be appropriate. However, it doesn't change the tire's rated pressure.

[freebeard]

The settled best practice among the hot VWs since the 1960s is to downsize the front tires. Cal Look drag racers use 135/15s and 185/15s in back. I used 145s/165s and found that it was good for 2-3mph top speed. Steering (tracking) improved, ride was unaffected. The Achilles heel is braking. The speedometer is affected but not the overall drive ratio.



Front/rear tire pressures were unbalanced to accommodate the rearward weight bias. On the panel van I had 185/14s on front and 215/15s on the back.

[California98Civic]

Originally Posted by CapriRacer

Look at my webpage on the subject: ....

I did and so did some others, and thanks. This chart is interesting. It shows in general that wider tires are better but there are two cases on it that leave me wondering.



When I examine the individual data points, I see a couple things that interest me beyond the clear trend of wider tire = lower Crr.

1) The chart suggests that between two 75 profile 14" tires the 195mm width has lower Crr than the 215mm, yielding a Crr improvement from maybe 11.3 to maybe 10.4. Why is that? I'm not trying debunk anything. I am confident you know the explanation.

2) The chart also suggests that one might improve rolling resistance by reducing width from a 185/65-R14 tire to a 175/70-R13 tire--maybe reducing Crr from 12.0 to 11.5. Why?

Again, this is not an effort at debunking... just trying to understand why this happens so I can maybe make better guesses at good FE tire/rim combos in the future.

[UFO]

Forgive my lack of knowledge on the subject, but there seem to be two terms being used, and I'm not sure if they are interchangeable.

Can the "coefficient of rolling resistance" be equated to rolling resistance? If not, in the case of that chart posted above, what is the variable that is not stated that would convert a coefficient to the actual value that contributes to efficiency loss in driving?

I ask because in the case of aerodynamic drag, there is the Cd (coefficient of drag) that has to be multiplied by the Cda, or effective frontal area to get actual drag. Is there a similar equation that relates to rolling resistance?

[oldtamiyaphile]

Originally Posted by California98Civic

Again, this is not an effort at debunking... just trying to understand why this happens so I can maybe make better guesses at good FE tire/rim combos in the future.

It's unlikely that given the range of tyres tested, that they would have all been the same make and model. Differences in tread pattern, compound, construction, load rating etc will be enough to cause the odd blip if you look hard enough.

That's why the bike graph I posted is so neat. That test was run with a single tyre model (GP4000 IRC) - width being the only variable.

[roosterk0031]

Tires were all goodyear something, but between sizes I doubt construction is the same at some size changes have to be made.

[California98Civic]

Originally Posted by roosterk0031

Tires were all goodyear something, but between sizes I doubt construction is the same at some size changes have to be made.

Yes, he says "these were Goodyear Integrity's"... Which is why I ask the question.

Originally Posted by oldtamiyaphile

... that's why the bike graph I posted is so neat. That test was run with a single tyre model (GP4000 IRC) - width being the only variable.

But CapriRacer says all the tires in his data were the same brand/model, too.

[CapriRacer]

Originally Posted by UFO

Forgive my lack of knowledge on the subject, but there seem to be two terms being used, and I'm not sure if they are interchangeable.

Can the "coefficient of rolling resistance" be equated to rolling resistance? If not, in the case of that chart posted above, what is the variable that is not stated that would convert a coefficient to the actual value that contributes to efficiency loss in driving?

I ask because in the case of aerodynamic drag, there is the Cd (coefficient of drag) that has to be multiplied by the Cda, or effective frontal area to get actual drag. Is there a similar equation that relates to rolling resistance?

To get RRC (Rolling Resistance Coefficient), you take RRF (Rolling Resistance Force) and divide it by the load. So to get the drag on a car, you need to know how much load is on each tire and the RRC of the tires

- EXCEPT -

Both RRF and RRC assume the inflation pressure is the same as the test value - which is typically the rated pressure. If someone wants the drag on a car due to the tires and the usage pressure is not the rated pressure, they would also need to know what affect inflation pressure has on RRC and/or RRF. There are ways to estimate this.

- AND ANOTHER EXCEPTION -

The author of those charts - Bruce Lambilotte from Smithers - had shared some additional data at a tire conference that indicated that RRC wasn't always constant. I wasn't quick enough to fully understand what he was saying - and he never actually published the data - but it stuck with me, because I had had other conversations where some folks assumed the RRC was a constant at a given pressure. I would really like to clear this up, but so far I haven't been able to.

[CapriRacer]

So let me talk about the charts on my web site.

First, the charts are from a presentation made in Dec 2007 by Bruce Lambilotte of Smithers to the California Energy Commission as a result of the CEC commissioning the study. They were trying to write a regulation and needed the data to do so. I just happen to stumble on the presentation.

Unfortunately the data said the CEC wasn't going to be able to write a regulation based on that data.

(And just an FYI, shortly after that time frame, NHTSA came forward and said they would write a regulation for the entire country, so the CEC decided to wait and see. There's a long story here, but the short version is that NHTSA is due to publish one in Aug 2018.)

OK, back to the charts.

If you'll read the whole web page carefully, you will notice that I thought there might be OE tires in the mix - and for those who don't know, OE tires are designed to the specs published by the vehicle manufacturer, and typically, those tires are designed for better RR than replacement market tires. I couldn't detect any obvious ones, but that doesn't mean they aren't there.

So if you see a tire that seems to have a lower RRC value, it just might be an OE tire.

The other source of anomalous data is that not every tire is in exact proportion to one in a different size. It wasn't stated in the presentation that any effort was made to pick a tire line that was as free as possible from production compromises. If I were doing the study, I would have worked very closely with a tire manufacturer to select a tire line that didn't have unusual design and production compromises that would affect RR. Let me give you an example of such:

Tramlining - aka groove wander - is where a tire tries to follow the grooves cut into pavement to drain water off the road surface and prevent hydroplaning. So the grooves in a tire must NOT line up with grooves in the pavement.

That's a problem because there is no standard for the spacing of the pavement grooves - so by trial and error, every tire manufacturer has a list of what spacing does not work, and then designs the grooves in their tires to NOT be those.

That means that what would work for - say - a 175mm cross section, might not work for a 185mm cross section, and the overall amount of tread rubber might not be proportional and, therefore, the RRC would not line up where it is supposed to.

There are other things that could cause the data not to exactly match up where it should, but I think you get the idea. Just for reference, when I did the regression I got an R squared value of 0.66 - which is not very good. There seems to be a lot going on in this study and it would be great if we could find more data from another source with a better R squared value.