sealPurdue News

March 25, 2003

Machine harnesses sound science to probe causes of road noise

WEST LAFAYETTE, Ind. – Innovative types of pavements may help to reduce traffic noise on future highways, according to initial research using a new, one-of-a-kind machine custom-made for Purdue University.

Download Photo Here
Photo caption below

"This is by no means a comprehensive study, but it does represent the beginning of work that eventually will lead to a precise understanding of what causes highway noise," said Robert Bernhard, co-director of Purdue's Institute for Safe, Quiet and Durable Highways.

Road noise is a major environmental nuisance for residents and businesses located near highways. Researchers are trying to design pavements and tires that produce less noise.

A 38,000-pound, 12-foot-diameter circular machine designed at Purdue makes it possible to test numerous types of pavement surfaces and compositions in combination with various tire designs. Curved test-pavement sections fit together to form a circle, and two tires are rolled over the surfaces at varying speeds while microphones and other sensors record noise and data.

Because no other equipment in the United States can test any combination of pavements and tires, researchers expect their new "Tire/Pavement Test Apparatus" to yield a wealth of data, said Bernhard, a professor of mechanical engineering and director of the Ray W. Herrick Laboratories at Purdue.

Preliminary findings will be discussed for the first time today (Tuesday, 3/25) during the 89th Purdue Road School, an annual gathering for government officials, traffic experts and engineers to discuss a range of transportation issues.

Unlike most other test equipment, the Purdue machine is designed so that pavement test samples remain stationary while tires roll over the surfaces. Keeping the pavements stationary is critical to accurately mimic actual road conditions. Many other machines use designs in which a stationary tire rides on top of motorized steel rollers.

The rollers cannot be made of actual pavements, however, because the spinning motion would cause the pavements to disintegrate. The steel rollers are covered with various surfaces, but those surfaces do not possess the exact characteristics of real pavements and are difficult to replace.

Researchers also measure highway noise by towing a tire behind a vehicle. That method, however, is limited to the type of pavement that is available on existing roads or test tracks, and the test results are affected by uncontrollable environmental conditions, such as other traffic, wind noise and changing weather. The Purdue engineers are able to analyze numerous types of pavements by changing test samples, and the experiments can be carried out inside a laboratory where acoustics, temperature and other environmental conditions can be carefully controlled.

"The thing we can do that nobody else can do is test a large range of pavements and tires in a way that promises to isolate the precise causes of highway noise," Bernhard said.

A report on the first findings is available online.

Engineers have initially used the machine to test three types of concrete surfaces – smooth, textured and porous – and four tire designs, which had varying degrees of stiffness in their sidewalls and tread bands, or the steel belts that encircle the tire underneath the treads.

"We tested all four combinations: tires with a weak sidewall and a weak tread band, a stiff sidewall and a weak tread band, a weak sidewall and a stiff tread band and a stiff sidewall and stiff tread band," Bernhard said.

An array of five microphones recorded the noise levels at various distances and at several frequencies, or tones, emanating from the interface of tire and pavement.

"There were pretty significant differences between the three pavements, but we saw pretty much no difference in the sound radiation from the tires," Bernhard said. "All four tires were pretty much the same, which means the sidewalls and the tread bands we tested really didn't have much of an effect on the tire noise. However, other tire characteristics, such as the rubber compounds used, may have more effect in controlling noise."

Preliminary information showed that the porous pavement generated the least noise.

"What we think is happening is that certain elements of the pavement and tire amplify sound, but we really don't understand this completely yet," Bernhard said.

More work is needed to discover precisely how the noise is generated. The researchers will soon begin testing tires and pavements that contain embedded sensors, which will provide more data about noise generation.

"We are going to go after what's happening at the interface of the tire and road surface," Bernhard said.

Engineers have several theories for the root causes of highway noise. One theory is that as a tire's tread blocks strike pavement, the rubber blocks vibrate, radiating noise.

"We want to find out whether a tread does in fact vibrate, and if so, at what frequency it is vibrating," Bernhard said.

Another theory is that grooves between tread blocks act as tiny organ pipes as the tire spins, producing noise. Still another theory is that the tread blocks stick momentarily after striking the pavement, making a noise as they pull away from the surface.

The Purdue machine, which cost about $250,000, was paid for with a combination of funding, including money from the U.S. Department of Transportation, contracts with several tire companies and a gift from Ford Motor Co.

Writer: Emil Venere, (765) 494-4709,

Source: Robert Bernhard, (765) 494-2141; Home (765) 463-5463;

NOTE TO JOURNALISTS: A publication-quality photograph and video B-roll showing graduate student William Thornton working with Purdue's new Tire/Pavement Test Apparatus are available. The video is available by contacting Jesica Webb, Purdue News Service marketing coordinator, at (765) 494-2079,


William Thornton, a mechanical engineering doctoral student at Purdue University, positions a microphone needed to gather data using the Tire/Pavement Test Apparatus at Purdue's Institute for Safe, Quiet and Durable Highways. The machine, the only one of its kind, is housed at the university's Ray W. Herrick Laboratories. (Purdue News Service photo/David Umberger)

A publication-quality photograph is available at


The Effects of Varying the Tire Cap Ply, Sidewall Filler Height and Pavement Surface Texture on Tire/Pavement Noise Generation

Robert J. Bernhard, Principal Investigator
William D. Thornton, Research Assistant
Jonathon Baumann, Research Assistant
The Institute for Safe, Quiet and Durable Highways
Schools of Civil and Mechanical Engineering,
Purdue University

In order to better understand the effect of tire carcass construction and pavement texture on tire/pavement noise generation, a measurement program as conducted on a group of four automobile tires on three pavement textures. The tires included all combinations of cap ply, no cap ply, sidewall filler height of 1.75 inches and sidewall filler height of 0.35 inches. The tires, General AmeriG4S, P205/65R15, were nominally identical in all other respects including tread pattern and rubber compound properties. During this investigation, other tire design variables such as tire pressure and rim size were held constant. The Tire/Pavement Test Apparatus at the Ray W. Herrick Laboratories at Purdue University was used for the tests. Five microphones were used to measure the sound generated by the tire/pavement interaction. Three microphones were located according to the specifications of the Close Proximity Method. The other two microphones were located such that one was in front of the tire 5 cm from the leading edge of the contact patch and one was behind the tire 5 cm from the trailing edge of the contact patch, along the centerline of the tire. The tires all produce similar overall sound pressure level and frequency spectra for a specific test surface. The three pavements produced significantly different spectra. The porous pavement produced higher levels below 1,000 Hz but lower levels than the other two pavement types above 1,000 Hz. The pavement surface textures investigated had a much more significant effect on sound generation and radiation of the tires than the variations investigated for cap ply and sidewall filler height.

* To the Purdue News and Photos Page