How much effect of Tire Pressure vs. Rev./Mile??

[ueberooo]

I'm running on an AWD, and I'm wondering how much of an effect there is on tire pressure and revolutions per mile in a modern steel belted tire.

Suppose I'm running the left rear at 27 PSI and the right rear at 33 PSI, and suppose that at 30 PSI I would get 800 revs/mi.

What would your guess be of the new revs/mi running at 27/33 pressures? (If it matters for the guess, assume it is an unloaded average compact car, around 3000 lbs inclusive of passengers).

 

[Coprolite]

I bet that you wouldn't be able to tell reliably. So, 800+/-.

 

[JTK]

I'm in the midst of a 'discussion' in regards to this exact topic with a fella over on a Subie board.

I know AWDs like these can and will bind, given somewhat large differences in tire circumference, but I don't think they are as freakishly sensitive as some worry.

You know there's countless owners running around with the fronts well worn compared to the rear tires, will no ruined drivetrain components.

Same goes with inflation. You can guarantee those can be all over the place depending on the owner.

Joel

 

[sunfire]

Originally Posted By: ueberooo
What would your guess be of the new revs/mi running at 27/33 pressures? (If it matters for the guess, assume it is an unloaded average compact car, around 3000 lbs inclusive of passengers).


My guess is... the diameter of the tire doesn't change enough to really matter at 27, 30, or 33 psi. But the rolling resistance of the tire does. Rolling resistance decreases as tire pressure increases.

 

[ueberooo]

Thanks Joel, yes, I'm trying to compensate for uneven treadwear with higher pressure. I just don't quite know if that will make enough difference; I got like 3 or 4 / 32nds on one tire and 6 or 7 on on the other.

Ballpark estimates for % diff:

circumference ~ mile/in/800 ~ 79 in

now 3/32 in * 2 pi ~ 0.6 in, or 0.76% difference

I'll try to do a chalk test to see that I get significantly less than this difference. If I run the car straight for 20 feet I should get almost two inches of difference in the relative positions of the chalk lines with the above difference: .6/79*240in = 1.8 in

So, if I were to run it with pressure difference to compensate, for 40 ft and get less than 2 inches difference in chalk lines I'd have significantly reduced the stress on the AWD by a binary order of magnitude. I may give this a try soon.

 

[mechtech2]

The diameter has nothing to do with it. Tires are not freight train steel wheels.

It is the radius from the axle center the the ground that counts.

And it does change quite a bit . Take a tire with 10 lbs in it, go to 20, then 30, then 40. There is a big change at first, but the disparity diminishes as the pressures get higher.
But yes, there is a difference.

And another thing - heavy loads will change the radius also [for any given pressure].

Someone will say it doesn't matter, etc. But it sure can, and it depends.

 

[jjjxlr8]

..

 

[ueberooo]

okay, nice, thanks, I'll keep you all posted as soon as I get results.

 

[fsskier]

I have been waiting for someone that has real measurements but....

A little history: In the old days of time and distance rallyes we installed odometers that read extra digits, and sports car magazines were full of tech articles on rolling diameter. The conclusion at that time.... about the time that premium cars.... mostly foreign cars.... were coming with radial tires.

Bias tires were hugely effected by pressure. Multiplying the radius from the hub center to ground time 2 Pi R gave a reasonable rolling circumference. The bias sidewalls.... basically triangles when viewed from the side... kept the easily deformed tread moving at the calculated speed. So a tire 80 inches around might only have a 78 inch rolling diameter.... and "glass highway" technical testing, complete with pictures showed that the tread pattern was greatly squished together, to make 80 inches of tread go by in only 78 inches.

A radial tire that measured 80 inches around went very nearly 80 inches per revolution regardless of pressure because of the effect of the steel belt, and also the radial sidewalls no longer trying to control the speed of the contact patch.

Imagine a caterpillar..... if the track is 2 feet tall but 12 feet long the 2 Pi R formula is obviously wrong, but measuring the total track length would be quite accurate.

Those who liked to win T&D rallyes (Heck that's everybody) made the switch to radials and we all immediately noticed that tire pressure had negligible effect on odometer readings compared to the bias tires that were common at that time.

That said, there is still a very small effect from pressure. For several years GM used the difference in rotational speeds to detect low tire pressure - using the wheel speed sensors from the ABS - but the difference is so small that that system was problematical. Many found that even a short drive on gravel set off the sensor..... the right side tires are often on looser gravel and very slightly "spinning" on every upgrade and setting off the alarm.

Wow, that got long...... I will wait for somebody that has newer actual data..... but suspect that the "rolling diameter" pressure effect is very small.

 

[leeharvey418]

I think fsskier hit the nail on the head.

With the ABS-based TPM systems, it would have taken enough deflation to get to the point where you were seriously deforming the tire (like his bias-ply squished tread story) before the system would detect low pressure.

...and as for the tracked vehicle - remember that if you're calculating ground speed, it's 2πrω, where r is the pitch radius of the drive sprocket and ω is the angular speed of the drive sprocket.

Ugh... stupid fonts not recognizing lower-case pi ( π ) and omega ( ω )

 

[mechtech2]

Also consider that cab drivers commonly used to run low tire pressures to increase the perceived miles, and thus charge higher rates for the same distance. Oldest cab driver's trick in the book!
And with radial tires.

 

[ueberooo]

Well, I tested for excessive difference in rotation today at the ikea parking lot, and over 80 ft found no difference, and over 280 ft (full length of lot) found possibly 1 hour to 2h 30 m of difference (30 to 75 deg) http://binged.it/AqIeY3 ; the passenger side rotated a little more, however, most of this is likely due to unevenness in the road surface (see tread wear numbers).

Tread:
LF 5 RF 6
LR 6.5 RR 3.5

The right rear was significantly overinflated to compensate for the worn tread, almost 35 psi, while the LR was ~26 or 27 psi; default inflation is 32 front / 29 rear.

Pressure:
LF 32 RF 32
LR 26.5 RR 34.5

Since right front should not be turning faster than left front unless terrain causes difference in distance, most of the difference of right rear is likely due to terrain. I would estimate worst case 45 degrees over 280 ft if terrain was perfectly even, or one revolution every 2240 ft, or 2.3 revolutions per mile (tires are 205/60r15 and rotate at 817 per mile) which is less than 0.3% of the tire's rev/mi.

Maybe some day when the weather is nicer (not raining) I'll go back and do this test with all tires at equal pressure for comparison.

 

[Vikas]

Before the TPMS were mandated, the car computer had ability to compare revolutions of all wheels to determine if one of the tire was running low. Is it possible for the computer to know the 10% difference in the air pressure? I do not think so.

 

[CapriRacer]

Originally Posted By: ueberooo
Well, I tested for excessive difference in rotation today at the ikea parking lot, and over 80 ft found no difference, and over 280 ft (full length of lot) found possibly 1 hour to 2h 30 m of difference (30 to 75 deg) http://binged.it/AqIeY3 ; the passenger side rotated a little more, however, most of this is likely due to unevenness in the road surface (see tread wear numbers).

Tread:
LF 5 RF 6
LR 6.5 RR 3.5

The right rear was significantly overinflated to compensate for the worn tread, almost 35 psi, while the LR was ~26 or 27 psi; default inflation is 32 front / 29 rear.

Pressure:
LF 32 RF 32
LR 26.5 RR 34.5

Since right front should not be turning faster than left front unless terrain causes difference in distance, most of the difference of right rear is likely due to terrain. I would estimate worst case 45 degrees over 280 ft if terrain was perfectly even, or one revolution every 2240 ft, or 2.3 revolutions per mile (tires are 205/60r15 and rotate at 817 per mile) which is less than 0.3% of the tire's rev/mi.

Maybe some day when the weather is nicer (not raining) I'll go back and do this test with all tires at equal pressure for comparison.


If I do the math, a 1 hour (by the clock) difference would equal 0.2% and a 2 1/2 hour difference would equal 0.5%.

I think that says definitively that there is a difference.

If I remember correctly, Subaru says that for some of their vehicles the difference in circumference can only be 3/16" = 2.3%.

 

[ueberooo]

Originally Posted By: CapriRacer
Originally Posted By: ueberooo
Well, I tested for excessive difference in rotation today at the ikea parking lot, and over 80 ft found no difference, and over 280 ft (full length of lot) found possibly 1 hour to 2h 30 m of difference (30 to 75 deg) http://binged.it/AqIeY3 ; the passenger side rotated a little more, however, most of this is likely due to unevenness in the road surface (see tread wear numbers).

Tread:
LF 5 RF 6
LR 6.5 RR 3.5

The right rear was significantly overinflated to compensate for the worn tread, almost 35 psi, while the LR was ~26 or 27 psi; default inflation is 32 front / 29 rear.

Pressure:
LF 32 RF 32
LR 26.5 RR 34.5

Since right front should not be turning faster than left front unless terrain causes difference in distance, most of the difference of right rear is likely due to terrain. I would estimate worst case 45 degrees over 280 ft if terrain was perfectly even, or one revolution every 2240 ft, or 2.3 revolutions per mile (tires are 205/60r15 and rotate at 817 per mile) which is less than 0.3% of the tire's rev/mi.

Maybe some day when the weather is nicer (not raining) I'll go back and do this test with all tires at equal pressure for comparison.


If I do the math, a 1 hour (by the clock) difference would equal 0.2% and a 2 1/2 hour difference would equal 0.5%.

I think that says definitively that there is a difference.

If I remember correctly, Subaru says that for some of their vehicles the difference in circumference can only be 3/16" = 2.3%.



However, the difference between right front and right rear is only an hour. I forgot to post the raw readings above. It was:

Chalk reading at ~280':

LF 9:30 RF 11:00
LR 9:30 RR 12:10

Also, 3/16" would translate to a 1.52% difference for my tires, which are around 80" in circumf.... 3/16/77.6*44/7 = .01518 .

Such diameter difference would translate to about 51" (compare to tire circumf of 77.6"), or a 237 degree difference. This is quite a bit bigger than observed.

So, I'm concluding that the tire pressure compensation works to some degree. Another contributing factor for the lack of observing this difference might be that the tread was not as unevenly worn in the center.