Tire Wear

When shopping for a new set of tires I began to wonder what tire wear rate looks like. Is it linear or do tires wear faster when new or toward the end of their lifetimes? How does it vary from front to back and side to side? Is the difference between city and highway driving large enough to measure?

Once I had the new tires I set out to explore some of these questions. I started measuring the tread depth weekly in the center groove of each tire. Initially I used a cheap digital caliper but the repeatability was poor. The caliper I have has a probe tip that is fairly wide such that the edges would hit the edges of the narrow center groove. Since the edges of the probe are sharp, slight differences in pressure would result in values that differed by up to 0.03 inches or so. I thus purchased a Fowler X-Tread tire tread depth gauge, model 74-225-500, that is designed for the job. This is essentially a digital caliper but the probe tip is rounded such that it fits the bottom of the tread with much better repeatability. Also the gauge has a wide base to make it easier to hold perpendicular to the tread. Finally, the probe is spring loaded so the same force is applied each time. Every 1-3 weeks all four tires are measured in 6-9 places and these results for each tire are averaged to get the tread depths.

The plots below shows the results so far. Currently I'm only checking the effect of mileage which is plotted along the X axis. The Y axis shows the measured tread depth. The values each week are shown by the symbols and the lines through the data points represents piecewise linear curve fits of the data. The two horizontal lines in the lower half of the plot represent two criteria for specifying the lifetime of a tire. The lower curve represents 2/32", the legal limit in most states for minimum acceptable tread depth and the height of tread wear indicators built into the tire. This depth corresponds to the common "Lincoln head penny test". The second line represents 4/32", a depth that may represent a safer minimum.

The first plot shows the actual measurements of the tread thickness. I've had the tires rotated twice and there's an obvious change in the wear rate at each rotation. As seen before the first rotation, the rear tires wear faster than the front tires and both left tires wear faster than the corresponding right side tires. After rotation the rear tires have moved to the front and vice versa so the tires originally on the front show an increased wear rate and the tires formerly on the rear show a decreased rate. The trend switches again at the second rotation where the tires end up back in their original locations. Between the rotations the data is quite linear. That will be easier to see in the second plot described later.

41,244 bytes

The difference between front and rear was not unexpected as the load is different on the two sets of tires, they are affected differently by corners and the suspension settings are different. The difference from right to left wasn't expected. This is probably an indication that my alignment isn't perfect although no such effect can be noticed while driving.

The estimated tire lifetimes shown on the plot were obtained by linear extrapolation of the four curves to find where the specified tread depths would be reached. This results in four predictions of lifetime. Since I'll change all four tires at once the value shown is for the tire that will reach a given depth soonest.

The second plot is an approximation of the tread wear rate I would have seen had I not had the tires rotated. Each curve is the running sum of tread wear at a given corner of the vehicle subtracted from the original depth at that corner. From this plot its easy to see that the wear rate has been quite linear over the life of the tires. The estimated tire lifetimes would be 3000-3400 miles less than the actual values due to the lack of rotation. At replacement the rear tires would have lost roughly twice as much tread as the fronts.

39,921 bytes

The last plot shows the wear rate averaged across all the tires. The tire lifetimes thus derived are approximately what would be expected if the tires were not directional such that side-to-side swaps could be included in the tire rotations. This would result in an additional 3600-6200 miles of tire life. The tires have a 40,000 mile warranty so even the largest estimated lifetime would be short by about 20%.

21,939 bytes

Return to Home Page Latest data: July 30, 2011