This Rumble Strip Noise Evaluation study presents results of sound level monitoring of three types of longitudinal rumble strips installed along the edge of two-lane rural roads in Polk County, Minnesota. The study is in response to objections raised by some landowners about the unwanted noise caused by vehicles traveling over rumble strips when they drift over the edge or centerline of the roadway. By changing and modifying the design, the ultimate goal is to provide the maximum safety by capturing the driver’s attention through tactile and sound levels while minimizing the associated external noise generated by the rumble strips. A limited number of references with substantial data on interior or exterior sound levels were identified for review. Most of the studies reviewed were based only on the overall A-weighted decibel (dBA) which in itself, is not sufficient to determine audibility or annoyance caused by rumble strip noise. Only two of the papers involved measurement of the spectral characteristics of rumble strips, using one-third octave band levels. Testing was performed on three rumble strip designs — California (CA), Pennsylvania (PA), and Minnesota (MN) - which were installed on rural highways near Crookston, Minnesota. Tests were performed with three different vehicles — passenger car, pickup truck and a semi-trailer truck. Three speeds were used for each vehicle/rumble strip combination — 30 mph, 45 mph and 60 mph. One test with each vehicle was made on the normal roadway surface, but only at 45 mph due to time and personnel constraints. Typical pass-by levels without rumble strips were extracted from the FHWA Traffic Noise Model (TNM) data base, which compared favorably with the 45 mph test. One-third octave band sound levels were taken 50 feet and 100 feet from the edge of roadway, as well as inside the vehicle adjacent to the driver. Video recordings were taken 50 feet from the edge of roadway. Digital audio recordings were captured for each of the sound level readings. The maximum observed pass-by level has been used in the comparative analysis. Overall A-weighted (dBA) levels showed proportional increases with traffic speed and vehicle weight. Exterior sound levels with California and Minnesota designs were fairly similar, but levels increased more rapidly with speed with the Minnesota design, with a higher exterior sound level than the California design at 60 mph. Interior sound levels were similar for both the California and Minnesota designs for the passenger car and pickup. Interior levels in the tractor cab increased somewhat with speed but were difficult to isolate from normal road noise. In general, the Pennsylvania design had lower exterior and interior sound levels. When comparing sound level with and without a rumble strip, the same patterns described above were observed. While comparisons of overall dBA levels between interior and exterior are somewhat complex, the data tend to show that the Minnesota levels are consistently greater than the California and Pennsylvania levels. While California and Minnesota levels are similar, the exterior California levels and frequency are generally lower than the Minnesota levels and frequency, suggesting an improved exterior-to-interior sound level ratio for the California design. The Pennsylvania design does not appear to show much change between interior and exterior levels, compared with the California and Minnesota designs. Estimates of sound level with distance from the roadway were made using a typical outdoor sound propagation model, with one-third octave band source levels taken from the maximum pass-by levels at 50 ft. These were then compared with the background sound level spectrum that was measured during the California and Pennsylvania rumble strip tests. Using the concept of sound detectability developed originally for the Army Tank Automotive Command in the early 1970s, the detectability of the rumble strips was calculated. “Detectability” level is normally lower than “audibility” level since it is associated with actively listening for a sound compared with passively hearing a sound. For example, in a restaurant, one can hear people at the next table but not pay much attention to what is being said. This can be called “passive” hearing. On the other hand, when one tries to understand carefully what is being said at the next table, this can be called “active” listening. In summary, it is the authors’ opinion that the California strip provided adequate driver feedback while generating less exterior noise than the Minnesota strip. The Pennsylvania strip did not provide much driver feedback, although it did generate less exterior noise than either the California or Minnesota strips. Based on the study results, potential future studies could include wider strips to address the heavy commercial tire bridging that occurred with the 8-inch strips. Additional studies could also evaluate different width centerline strips and other vehicle types. (Author/publisher)
Abstract