Search Results

In the late 1970’s and early 1980’s, Jing-Yau Chung along with Joseph Pope published several external General Motors reports on the then novel measurement of sound intensity (SI) using the two-microphone, cross-spectral method. Application of this measurement method was then extended to sound intensity measurements in flow. Through component wind tunnel measurements, it was determined that the intensity of noise sources could be accurately measured up to a level of 15 dB below the sound pressure level generated by flow noise on microphones. An initial application of this method was to the identification of noise sources alongside rolling truck tires. It was then extended to the measurement of the aerodynamic noise generated by protrusions added to automotive vehicle designs. These included items such as outside rearview mirrors, windshield wipers, A-pillar offsets, grille whistles, roof racks, underbodies, and fixed-mast radio antennas.

Noise and vibration measurements were conducted on eight light vehicles ranging from small compact passenger cars to a large sport utility vehicle on and off shoulder rumble strips of two different designs to assess the input to a vehicle operator when the vehicle departed from the travel lane. The first design was a more conventional design, consisting of cylindrical indentions ground into the pavement at regular 30 cm intervals, and a continuous sinusoidal profile with a peak-to-peak length of 36 cm. Triaxial vibration measurements were made at six locations, including the steering wheel and column, the seat cushion and track, and the front and rear spindles. Interior noise was measured at six locations, one at the operator’s outward ear and five at the front seat passenger (three in the fore/aft locations of the seat and at outboard and inboard ear locations). In addition to the in/on vehicle measurements, pass-by noise levels were made.

Acoustic beamforming was used to localize noise sources on heavy trucks operating on highways in California and North Carolina at a total of 20 sites. Over 1,200 trucks were measured under a variety of operating conditions, including cruise on level highways, on upgrades, down degrades, low speed acceleration, and for various speeds and pavements. The contours produced by the beamforming measurements were used to identify specific source contributions under these conditions and for a variety of heavy trucks. Consistently, the highest noise levels were seen at the tire-pavement interface, with lesser additional noise radiated from the engine compartment. Noise from elevated exhaust stacks was only documented for less than 5% of the trucks measured. The results were further reduced to produce vertical profiles of noise levels versus height above the roadway. The profiles were normalized to the highest noise level at ground level.

A measurement program was completed to assess driver input versus exterior noise generation for four vehicle designs and two different rumble strip designs. The vehicles included a small compact car, an immediate size car, a full sport utility vehicle, and a medium duty dump truck. The first rumble strip was a conventional design providing shorter wavelength input to the tire. The second was designed to provide longer wavelength, more harmonic input to the tire. The measurements included exterior pass-by noise and on-board exterior noise and interior measurements of sound pressure level and vibration level at the seat track and steering column. In general, the results indicated that the longer wavelength strips produced less overall A-weighted pass-by noise with little or no reduction in interior noise and vibration.

The results of a series of tests were performed that are used to investigate the contribution of aerodynamic noise to lower frequency passenger car interior and exterior cruise noise levels. Wind tunnel measurements were used to isolate aerodynamic noise from tire-pavement and engine noise and to indicate that the vehicle underbody is a significant source region for both interior and exterior noise. Comparing interior on-road measurements to the wind tunnel results, it was found that aerodynamic noise was slightly less than an equal contributor to cruise noise averaging 4.8 dB lower than the road levels between 50 and 400 Hz at a speed of 80 km/h. At 140 km/h, the difference dropped to 2.3 dB indicating that the aerodynamic noise was the major contributor. For exterior pass-by, aerodynamic noise levels were found to account for almost all of the noise measured during coast-by conditions in the frequency range from 50 to 400 Hz at 97 km/h.

Developing common methods of noise evaluation and facilities can present a number of challenges in the area of tire/pavement noise. Some of the issues involved include the design and construction of pavements globally, the change in pavement over time, and variation in the noise produced with standard test tires used as references. To help understand and address these issues for airborne tire/pavement noise, acoustic intensity measurement methods based on the On-board Sound Intensity (OBSI) technique have been used. Initial evaluations have included measurements conducted at several different proving grounds. Also included were measurements taken on a 3m diameter tire noise dynamometer with surfaces replicating test track pavements. Variation between facilities appears to be a function of both design/construction and pavement age. Consistent with trends in the literature, for smooth asphalt surfaces, the newest surface produced levels lower than older surfaces.

With increasing use of the constant speed pass-by conditions to capture the noise generated by this portion of the vehicle operating cycle, knowledge of the contributing sources of noise was become increasingly important. For frequencies above 400 Hz, the noise is dominated by tire/pavement noise as can be demonstrated by comparing on-board sound intensity (OBSI) measurements to constant speed pass-by noise levels. At lower frequencies, direct on-board measurements become more difficult as the tire/pavement noise source strength decreases with decreasing frequency and microphone induced wind noise increases. To investigate the contribution of sources at these lower frequencies, cruise and coast pass-by measurements were made for a number of different pavement types and two different tire designs at test speeds of 56, 72, and 97 km/h over a frequency range from 50 to 10,000 Hz. OBSI measurements were also conducted for these same conditions.

There is a growing interest in using a standard reference tire for both assessing changes in test track pavement over time and rank ordering of the performance of different highway pavements. Because of longer-term availability, the ASTM Standard Reference Test Tire (SRTT) is the primary candidate for these applications. Issues of concern for the SRTT include tire-to-tire variation, the relation of the SRTT to other tires currently in use, and the “break-in” period required for stable test tires. To address tire-to-tire variability, seven SRTT’s were tested on variety of asphalt concrete (AC) and Portland cement concrete (PCC) surfaces on two occasions. These included five new tires and two that had been in use for some time. Two of the new tires were re-tested with increasing use to examine any break-in period effect.

Acoustic beamforming was used to visualize the sound radiation of trucks under test track passby and actual highway operating conditions. The purpose of these measurements was to obtain an understanding of which sources contribute to the overall passby noise level and to determine the vertical distribution of noise sources. For trucks, drive axle tires were found to be the major contributor to passby noise at highway speeds, followed by powertrain noise to a much less degree, and very occasionally, exhaust stack outlet noise. For medium and heavy trucks, the acoustic mean source height was found to be about 0.5m and about 0.3m for light vehicles.

Typically, the noise emission from trucks under highway cruise conditions is not reduced as much as it is for light vehicles when quieter pavements are used. Potential reasons for this are that other noise sources beside tire/pavement noise are more significant for trucks than light vehicles and/or the effect of pavement on truck tire noise generation is different than it is for light vehicle. As the cruising passby noise levels of trucks are about 10 dB greater than for light vehicles, this becomes an important issue for highway noise abatement when trucks make up even a relatively small percentage of the traffic flow. To investigate this issue, beam forming and conventional passby testing methods were used to investigate the contribution of both tire/pavement noise and the other noise sources for common types of heavy trucks.

At speeds of 50 km/h or greater, the exterior noise emission of light vehicles is typically dominated by tire/pavement noise for operating conditions of cruise and moderate acceleration. At a test speed of 56 km/h, it has been found that pavement type can create a 10 dB or more variation in tire/pavement noise. This has significant implications for both community noise and vehicle noise emission testing. In this paper, the results of tire/pavement noise measurements for over 80 different pavements in Europe and the United States are reported. These pavements include research surfaces, existing roadways, and ISO 10844 passby test surfaces. Measurements were conducted using an on-board sound intensity methodology that has been correlated to cruise-by noise levels. These results are discussed in terms of the revisions being considered for the ISO 362 passby test procedure and the ISO 10844 test surface specification.

With the growing recognition that pavement selection can be an effective traffic noise abatement tool, there has been increased need for developing methods to characterize tire/road noise generation for existing and experimental highway surfaces. To address this need, sound intensity measured on-board a test vehicle has been developed as an alternative technique to wayside, passby, or trailer methods. As part of this development, the relationship between sound intensity measured close to a moving tire contact patch and coastby sound pressure data measured at stationary point 7.5 meters away has been demonstrated for different tires and road surfaces. A protocol for sound intensity measurement on existing highways in traffic has also been developed. Using these, a library of the tire/road noise levels has been assembled for California State Highways and experimental highway test sections.

A road surface complying with the new International Standards Organization (ISO) specification was installed at an Arizona test facility (DPG site) in the winter of 1995/96. As part of the acoustic qualification of this site, comparative tests were conducted between this new surface, a Society of Automotive Engineers (SAE) sealed asphalt surface and an existing ISO surface in Michigan (MPG site). Initial testing with one vehicle and tire combination indicated that the new ISO surface produced ISO 362-1994 passby and coastby levels about 2 dB lower than sealed asphalt. Relative to the Michigan surface, the levels for the new Arizona ISO surface were 3 to 3½ dB lower. These differences were much greater than expected based on previously published studies of these two test surface types.

Errors in sound intensity measurements are examined for situations in which a high reactive component is present. This occurs when the indicated sound intensity level is much smaller than the sound pressure level. In these situations, significant errors in the sound intensity measurement can occur for even very small phase mismatch between microphone channels. High reactive components may be encountered in the near field of an extended source such as a body panel, when the intensity vector does not coincide with the sensitive axis of the probe, or when measurements are made in the presence of high background noise, reverberant sound fields, air flow, or standing waves. Expressions for calculating sound intensity error due to phase mismatch with high reactive components present are developed and examples of calculated error for common instrumention are provided. Cases of sound intensity measurements with known amounts of phase mismatch are also examined.

In previous literature, sound intensity has been used to quantify the strength of tire-pavement interaction noise sources very near an operating tire under non-driven, cruise conditions as measured on trailers or actual vehicles. In the current investigation, the relationship between such on-board sound intensity data and coast-by sound pressure levels measured 7.5 meters away from the centerline of vehicle travel were examined. When compared either in terms of overall A-weighted levels or 113 octave band spectra, these data demonstrate a strong correlation between the two types of measurements. Given this correlation, the sound intensity technique was then used to quantify the tire/pavement interaction noise for the driven tires of a passenger car under accelerating conditions such as those specified in the ISO 362/SAE 51470 or SAE 5986 passby noise procedures.