Experiment: coast down distances (rolling resistance) @ various tire pressures

[MetroMPG]

The purpose of the test was to get an idea of the difference in coasting distances in my '98 Metro/Firefly over a range of tire pressures, 20-60 psi.

EDIT Jan 23/09: Posted a summary of this thread at: http://metrompg.com/posts/tire-press...resistance.htm

Tire tested: 155/80/R13 Goodyear Invictas, rated 44 PSI max sidewall.

EDIT - Also tested Oct 13, 2011... Bridgestone RE92 tires, multiple runs at pressures from 20-70 PSI. See this post: http://ecomodder.com/forum/showthrea...tml#post265379

Placard info: vehicle placard recommends 32 PSI front/rear @ max load.

Pressure Gauge: readings were taken with an Accutire digital gauge, 5-99 PSI rated, with a manufacturer accuracy claim of +/- 1% + 0.5

Weather: 19C / 66F, 10 km/h SSW wind (roadway ran SW/NE)

Methodology:

• tires were pumped up to 60 PSI, drove to test route (< 5km), pressure adjusted
• car was driven up a small hill (approx. 6 ft. elevation, 8:1 slope), turned around and stopped at a marked point
• engine off, transmission in neutral, brakes were released
• car rolled down short hill onto a flat run-out road
• where the car stopped, the road was marked
• pressure was adjusted (dropped 5 PSI)
• rinse & repeat
• NOTE: only one run per PSI

I used a bicycle wheel to measure the rolling distances - counted revolutions from the starting point, then converted circumference to total feet.

Ideally, I would have simply measured the coastdown distance from a constant speed at the same point on a level road, but my cruise control isn't working, and I didn't want to deal with the possibility of driver error (varying speeds). Even more ideally, I should have done multiple runs per pressure level.

Since the car accelerates from rest and coasts to a stop, the differences in pressure are amplified compared to a simple coastdown test.

Raw results for PSI/feet travelled

20 / 479.3
25 / 524.8
30 / 621.0
35 / 621.0
40 / 639.6
45 / 687.5
50 / 702.0
55 / 699.3
60 / 702.0

[Daox]

Very good info!

[NeilBlanchard]

Hello,

It is interesting to note that the plateau seems to happen roughly at the 44psi sidewall max pressure.

[trikkonceptz]

Nice to see that 50psi seems to be the magic number ...

[MetroMPG]

Originally Posted by Daox

Very good info!

Not really. Very good info would have been multiple runs at each pressure level to ensure I was getting accurate readings.

(You can tell the accuracy is questionable by the pair of identical distances at different pressures.)

It's useful info as a "snapshot", but I wouldn't draw any hard & fast conclusions from it. Maybe it'll encourage other people to try it for their particular vehicles / tires.

[dcb]

Cool Test How fast were you getting up to? I'm wondering if the air drag at the higher speeds is what is leveling off the results.

[MetroMPG]

Not fast - 18.5 km/h @ 60 psi (the only time I checked).

But potentially approaching 28.5 km/h with the cross breeze (anyone want to do the vector calcs? ), but the road was also partially sheltered by trees & buildings, so I doubt I was seeing the full wind reported at the weather station.

I figure the varying breeze may have contributed to the pairs of matching results.

[MetroMPG]

Based on my aero+rr spreadsheet, at 10 mph (16 km/h), the proportion of power required to overcome rolling/mechanical vs. aero losses is: 0.44 to 0.05 hp.

At 15 mph (24 km/h) it's 0.74 hp vs. 0.16 hp for rolling/aero.

[SVOboy]

Mehbe some more people will get out there and look for some data, but that's encouraging stuff, I think.

[ttoyoda]

"Not really. Very good info would have been multiple runs at each pressure level to ensure I was getting accurate readings. "
You did do well, and deserve the credit. Very-very good info would be to either measure the temperature of the tire before and after each run, OR take an air tank with you and do the test again, but INCREASE the air pressure after each run.
You might consider checking the tire pressure before you begin the run, and right after you finish, to see if there is a measureable difference.