National Academies Press: OpenBook

Measuring Tire-Pavement Noise at the Source (2009)

Chapter: Chapter 5 - Demonstration Testing of OBSI Procedure

« Previous: Chapter 4 - Evaluation of OBSI Test Parameters
Page 17
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 17
Page 18
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 18
Page 19
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 19
Page 20
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 20
Page 21
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 21
Page 22
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 22
Page 23
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 23
Page 24
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 24
Page 25
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 25
Page 26
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 26
Page 27
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 27
Page 28
Suggested Citation:"Chapter 5 - Demonstration Testing of OBSI Procedure." National Academies of Sciences, Engineering, and Medicine. 2009. Measuring Tire-Pavement Noise at the Source. Washington, DC: The National Academies Press. doi: 10.17226/14212.
×
Page 28

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

17 Introduction The investigation was conducted to demonstrate the ability of the OBSI measurement method to quantify the relative effect of different pavement types in comparison to total vehi- cle noise emissions measured with the CPB and SPB methods. It included the simultaneous measurement of (1) OBSI on spe- cific candidate test tires, (2) controlled passbys on test vehicles equipped with the test tires, and (3) statistical passbys of both light- and heavy-duty vehicles on in-service payments. Using the SPB data, the relationship between OBSI tire-pavement noise and average vehicle noise emissions was examined for both light vehicles and heavy trucks. To account for site-to-site variation, passby results were normalized using the measured relationships between OBSI and CPB levels for each site. This chapter describes the measurements performed, the results of the OBSI and passby testing, and the relationships between the OBSI and passby data. Description of Field Measurements Measurement Sites The OBSI and passby testing was conducted at a total of 12 sites (five in Iowa and seven in California). The sites in Iowa were located along U.S. Highway 30 between mileposts 178 and 198, near Marshalltown. Portions of this section of highway were recently constructed to include many different types of surface texturing (17). Along this section of highway, four sec- tions with different PCC texture were selected as test sites, including a burlap drag surface, a random transverse tined surface, a uniformly tined surface, and a longitudinally tined surface. In addition, a nearby hot-mix AC pavement section was selected as a test site. In California, four of the Caltrans test sections on LA 138 (18), including DGAC, OGAC, rubber- ized, and bonded wearing course AC pavements were tested. Two PCC sections, a grooved and a ground pavement, on the Caltrans research sites on the Mojave Bypass (KN 58) (19) were also tested. In addition, a highly porous rubberized AC pavement along Shasta 299, about 5 miles east of Redding was tested. The site location, photographs of the pavement sec- tions, and the average 1⁄3 octave band spectrum for each sur- face under baseline conditions are provided in Appendix D. All test sites followed the applicable criteria stipulated in the FHWA document on highway noise measurement (20) and the ISO 11819-1 procedure (21). To the degree reasonable, sites were selected to have acoustically hard (non-sound absorb- ing) surface characteristics between the vehicle lane of travel and at least half of the distance to the 25-ft microphone fol- lowing the ISO recommendation. Because the ISO procedure only addresses 25-ft distances, the FHWA criteria were used for the 50-ft microphone positions. To obtain comparable results, sites were selected along roadways where the posted and typi- cal vehicle speeds were 55 mph or higher. The designation, location, and description of each of the test sites are listed in Table 6 (details of these sites are also provided in Appendix D). Measurement Protocol OBSI and controlled passby measurements were conducted using the 2007 Pontiac Grand Prix and the 2007 Chevrolet Impala with Michelin/Uniroyal SRTT and Dunlop SP Winter Sport M3 tires (Dunlop). The OBSI measurements followed the measurement protocol discussed for the parameter inves- tigation using the two-probe approach (2) with a vehicle load consisting of two people and the OBSI instrumentation. Mea- surements were conducted at 60 mph and at test speeds vary- ing from 50 to 70 mph, in 5-mph increments, depending on the typical speed of on-road vehicles, in order to match the speeds of the SPB measurements. The microphone signals were acquired using the same instrumentation and data analysis sys- tem described in Chapter 4. A standard 5-second averaging time was used for test Sites 1 through 9 and 12; a 4-second averaging time was used for Sites 10 and 11 because of the shorter lengths of pavement sections. During post-analysis, C H A P T E R 5 Demonstration Testing of OBSI Procedure

18 the 60 mph samples were reanalyzed into shorter sample seg- ments to assess the variation of OBSI level over the standard 440-ft test section. The microphones were calibrated at the beginning and end of the measurement period. Three passes were made for each test speed, which were averaged together during post analysis. During the data acquisition, OBSI qual- ity metrics of coherence between the two microphones com- prising each probe and the difference between sound pressure and sound intensity level were monitored. The actual time sig- nals of the four microphones were monitored in order to identify any data acquisition abnormalities. Statistical vehicle passby measurements were done generally following the procedures described in the report “Measure- ment of Highway-Related Noise” (20). Two microphone posi- tions were used: one at a distance of 25 ft from the centerline of travel and one at a distance of 50 ft at heights of 5 ft above the height of the pavement. An example passby measurement setup is shown in Figure 12 for the longitudinally tined PCC pavement (Site 5). The passby sound pressure levels were measured using two-channel real time analyzers (RTA). The analyzers were set to “fast” response (1⁄8 second exponential average) and the 1⁄3 octave band and overall A-weighted sound pressure levels occurring in 1⁄10th-second intervals were log- ged during the passby. The maximum level for each passby was then determined from a plot of sound level versus time provided from the RTA. The microphone output signals were also captured on a solid-state digital recorder as backup and later use as required. Vehicle speed and type (light vehi- cle, medium or heavy duty truck, or “other”) were manually recorded for each measured passby. Vehicle speed was mea- sured with a radar gun and wind speed was documented from an anemometer at regular intervals. Prior to testing, the test vehicle speedometer was calibrated with the radar gun “shot” in a direction very nearly parallel to the centerline of vehicle travel. Vehicle speed measurements during the CPB tests were later used to adjust the recorded speed to the actual speed at the measurement angle used for passby data collection. The total number of passby events acquired at each site ranged between 178 and 259 depending on the occurrence of sufficient (nom- inally 40) heavy truck passby events. Controlled vehicle passby measurements of the test vehicles followed the same procedures as those described for the statis- tical passbys. A minimum of three passbys was measured at each test speed. Meteorological Conditions Testing in Iowa (Sites 1 through 5) was conducted over a 3-day period from September 26th to 28th, 2007, where clear skies and calm to light winds (0 to 7 mph) prevailed. Temper- atures ranged from 48 to 60°F for the morning testing on the 26th and 28th, from 60 to 68°F on the afternoon of the 26th, from 59 to 73°F on the morning of the 27th, and from 76 to 77°F on the afternoon of the 27th. Measurements along LA 138 (Sites 6 through 9) and Mohave Bypass (Sites 10 and 11) were conducted on November 6th to 9th, 2007. The sky was clear during the testing with winds from the east of up to 9 mph. Air temperatures ranged from about 60°F in the morning to highs of about 75 to 77°F in the after- noon on all 4 days. The Shasta 299 (Site 12) monitoring was conducted on November 30, 2007, where clear skies and air temperatures ranging from 45°F to about 55°F, with winds speeds of up to 6.7 mph prevailed. Table 6. Passby test sites. Figure 12. Typical passby measurement setup. Site # Location Description 1 US 30 near Marshalltown, IA PCC-Burlap Drag 2 US 30 near Marshalltown, IA PCC-Random Transverse Tine 3 US 30 near Marshalltown, IA PCC-Uniform Transverse Tine 4 US 30 near Marshalltown, IA AC-Dense Graded 5 US 30 near Marshalltown, IA PCC-Longitudinal Tine 6 LA 138 near Lancaster, CA AC-Dense Graded Overlay 7 LA 138 near Lancaster, CA AC-Open Graded Overlay 8 LA 138 near Lancaster, CA AC-Rubberized Open Graded Overlay 9 LA 138 near Lancaster, CA AC-Bonded Wearing Course 10 KN 58 near Mojave, CA PCC-Grooved on Burlap Drag 11 KN 58 near Mojave, CA PCC-Ground (2.67 Blade Spacing) 12 Shasta 299 near Redding, CA AC-Rubberized Open Graded Overlay 25 ft Mic. Position 50 ft Mic. Position

19 Data Reduction and Analysis Data Reduction To produce tire OBSI levels for each run, the leading and trailing edge levels were averaged together on an energy basis. For each condition (vehicle speed, pavement, and tire), the 1⁄3 octave band levels between 250 and 5,000 Hz for the three runs were arithmetically averaged to represent that condition. The overall level between 400 and 5,000 Hz was then calculated. Although the levels in the 400 Hz band were sufficiently low so as to have minimal effect on the overall level, they were included as data quality indicator requirements were met. These overall levels were then used for comparison to the over- all passby levels. The 60-mph samples were also reanalyzed into shorter sample segments to assess the variation of OBSI level over the standard 440-ft test section (results are provided in Appendix D). Most of the sections exhibited fairly homo- geneous noise levels throughout the length of the test section. Two of the pavement sections, Site 2 and Site 12, showed notable variation. Based on the review of several passes over the same pavement section, the level variation occurred in the same locations as the vehicle traveled over the pavement. It was determined that the 4-and 5-second average levels for each site were appropriate for the remainder of the OBSI/ passby analysis. In processing the passby data, each event for both the 25-ft and 50-ft microphone positions was reviewed to verify that it was acoustically “clean.” The ISO 11819-1 (21) criteria was used to define clean passby events; only events where a single peak rose at least 6 dB above any surrounding data were included in the analysis. The maximum level for each such event was paired with the recorded vehicle type and speed. For the SPB events, these levels were plotted against vehicle speed for both light vehicles and heavy trucks. The data were then fit with a standard logarithmic regression producing an equation and r2 value and plots were reviewed for “outlier” points (points seemingly out- side the normal range of data). Points that could be associ- ated with field notes of unusual noises were dropped from the data set. For each data set, the usable speed range was also determined. The clean passby vehicle counts and speed ranges for the test sites are given in Table 7. For comparison to OBSI and CPB data, the logarithmic regression equations for each data set were used to calculate the average SPB level for usable speed range in 5-mph steps (plots of the SPB data and the regression curves, equations, and r2 values are given in Appendix D). The data from the CPB events were processed similar to the SPB data. However, the data for the passby levels were processed in 5-mph increments over the speed range within 50 to 70 mph. Average levels at each speed, site, and tire configu- rations were calculated and used in comparison to the OBSI and SPB data. An example of SPB and CPB levels versus speed is shown in Figure 13 for Site 5, the longitudinally tined PCC pavement on U.S. 30 (similar plots for all of the test sites are included in Appendix D). Data Analysis Once the overall SPB, CPB, and OBSI levels were estab- lished, data analysis was performed to investigate specific aspects of the results. For the light vehicle overall levels, cross- plots of CPB versus OBSI, SPB versus OBSI, and SPB versus CPB were constructed for both test tires and for each micro- phone distance. These included levels for all speeds and all sites. For the heavy truck overall levels, plots of SPB versus OBSI and SPB versus CPB were constructed. For each cross-plot, the data were fit with linear regressions and best fit 1-to-1 slope lines. From these, the slope of the linear regression and r2 values were Table 7. Number of clean SPB events and vehicle speed range at each test site. Light Vehicles Heavy Trucks 25 ft Microphone Distance 50 ft Microphone Distance 25 ft Microphone Distance 50 ft Microphone Distance Site # # of Events Speed Range (mph) # of Events Speed Range (mph) # of Events Speed Range (mph) # of Events Speed Range (mph) 1 103 55–70 44 55–70 30 60–70 15 60–70 2 136 50–70 129 50–70 31 50–65 31 50–65 3 120 55–70 120 55–70 42 60–65 42 60–70 4 37 55–65 14 55–65 5 123 55–70 118 55–70 56 60–70 54 60 – 70 6 123 50–70 123 50–70 54 50–65 54 50–65 7 123 50–70 123 50–70 56 50–65 56 50–65 8 108 50–70 108 50–70 49 50–65 49 50–65 9 89 50–70 89 50–70 42 50–65 42 55–65 10 136 60–70 135 60–70 80 50–65 80 50–65 11 119 60–70 118 60–70 52 50–70 52 55–70 12 126 50–70 119 50–70 33 50–65 33 50–65

20 determined as was the offset, standard deviation, and average deviation from the 1-to-1 line resulting in data similar to that shown in Table 1. Additionally, the difference between OBSI and CPB levels was calculated for each site, each speed, and each tire for those cases where the corresponding data or both types were available (for the 25-ft and 50-ft data). The average and deviations were also determined for these data. The differ- ence between the 25-ft and 50-ft passby levels was also calcu- lated for each passby event for which the corresponding data were available. The differences between the OBSI and CPB data were used to develop normalization coefficients to account for site-to-site geometry and propagation differences. The need for site nor- malization has been noted in previous work (15). Earlier SPB studies on LA 138 reported site biases ranging from −0.6 to 1.4 dB relative to the reference site factors for both the 25-ft and 50-ft microphone locations (15). The measurements con- ducted in Phase I displayed site-specific effects of up to 4 dB even for measurement distances of 25 ft. These effects were evi- denced both by propagation testing (see Appendix B) and cor- responding differences between the OBSI and CPB. Due to the traffic volumes at the measurement sites, propagation tests could not be made. As a result, site normalization factors were determined by first determining the average difference between OBSI and CPB levels at each site and overall average for the 12 test sites. The average OBSI/CPB difference for each site was then subtracted from the average of all sites to deter- mine the normalization factor for each site. These factors were then applied to the SPB data on a site-by-site basis, and back to the CPB data for confirmation. The normalization coefficients were also applied to the heavy-truck SPB data. Results and Discussion The primary results of these measurements are presented in this section (more complete results, including the remainder of the cross-plots, spectral comparisons of OBSI and CPB lev- els, and level versus speed plots for the CPB and SPB data are given in Appendix D). Normalized SPB and CPB Data versus OBSI In order to demonstrate the applicability of the OBSI data to in-service pavements, a series of cross-plots were considered in which the relationships between OBSI and the passby data could be quantified. The first step was to develop the normal- ized relationship between the CPB and OBSI data for all sites and speeds for each tire using the analysis discussed in the pre- vious section. The effect of the normalization can be seen by comparing the raw cross-plots of CPB levels versus OBSI for the SRTT tire in Figure 14 to the normalized results shown in Figure 15. In Figure 14, the data points from any one site tend to fall below or above the regression line and the 1-to-1 line. Ignoring these offsets, the points for each site tend to follow a constant slope similar to the regression and 1-to-1 lines. When the data are normalized as shown in Figure 15, these offsets collapse to follow a 1-to-1 slope with considerably less scatter. In this exam- ple, the slope of the regression is decreased from 1.31 to 1.06 and r2 value is increased from 0.91 to 0.96 with normalization. The 1-to-1 offset remains virtually the same with normalization (24.2 dB with it and 24.3 dB without), however, the standard deviation about the line is reduced substantially from 1.4 to 0.6. Figure 13. SPB and CPB levels versus speed for heavy and light vehicles at 25 and 50 ft for Site 5 (longitudinally tined PCC pavement). 65 70 75 80 85 90 95 100 40 45 50 55 60 65 70 75 80 Speed, mph So un d Pr es su re L ev el , d BA Dunlop 25ft CPB SRTT 25ft CPB Dunlop 50ft CPB SRTT 50ft CPB Lt Vehicles 25ft SPB Lt Vehicles 50ft SPB Hvy Trucks 25ft SPB Hvy Trucks 50ft SPB Hvy Trucks 25ft SPB Hvy Trucks 50ft SPB Lt Vehs 25ft SPB Lt Vehs 50ft SPB

21 Normalization produced similar effects on the data from the Dunlop tire and both tires for the 50-ft microphone locations as indicated in Table 8. In each case, the value of 1-to-1 offset remained virtually unchanged while the standard deviations are reduced by more than 50%. This finding confirms that the normalization does not affect the relationship between the CPB and OBSI data but it reduces the scatter attributed to site-to-site variation, and therefore, the coefficients were also applied to SPB data. Site normalization produced a similar effect on the SPB data as it did on the CPB data. Invariably, the value of the 1-to-1 line offset was virtually unaffected while the scatter was reduced as shown in Figures 16 and 17 for the 25-ft, light vehicle SPB data and the SRTT OBSI data. The effect on the plot metrics for 50-ft data and the Dunlop tire are given in Table 9. As noted in Table 9, the reduction in scatter for the SPB data is not as pronounced as it was for the CPB data (see Table 8) partially due to the appreciable scatter was seen between the CPB and SPB data as illustrated in Figure 18 for the SRTT data at 25 ft. In this plot, any site bias is effectively cancelled out leav- ing only the ability of the test tire to replicate the behavior of the SPB data that spans many different tires and other light vehicles. Figure 14. Controlled vehicle passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—raw data. Figure 15. Controlled vehicle passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—normalized data. 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 10 2 1 04 106 108 110 OBSI Level, dBA So un d Pr es su re L ev el , d BA PCC 1 PCC 2 PCC 3 AC 4 PCC 5 AC 6 AC 7 AC 8 AC 9 PCC 10 PCC 11 AC 12 1-to-1 Line Linear (24.3 dB offset) 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 10 4 1 06 108 110 OBSI Level, dBA So un d Pr es su re L ev el , d BA PCC 1 PCC 2 PCC 3 AC 4 PCC 5 AC 6 AC 7 AC 8 AC 9 PCC 10 PCC 11 AC 12 1-to-1 Line Linear (24.2 dB offset)

22 Figure 16. Statistical light vehicle passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—raw data. Figure 17. Statistical light vehicle passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—normalized data. Table 8. Metrics for CPB versus OBSI relationship. 25 ft Microphone Distance 50 ft Microphone Distance SRTT Dunlop SRTT Dunlop Cross-Plot Metrics Raw Norm Raw Norm Raw Norm Raw Norm Slope 1.31 1.06 1.17 0.95 1.41 0.97 1.38 0.97 r 2 0.91 0.96 0.85 0.94 0.86 0.96 0.86 0.96 Offset, dB 24.3 24.2 24.8 24.7 31.4 31.0 31.7 31.3 Std Dev, dB 1.4 0.6 1.4 0.6 1.9 0.6 2.0 0.5 Avg Dev, dB 1.1 0.6 1.1 0.5 1.6 0.4 1.6 0.4 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 104 106 108 110 OBSI Level, dBA So un d Pr es su re L ev el , d BA US 30 BD US 30 TT US 30 UT US 30 AC US 30 LT LA138 S1 LA138 S2 LA138 S4 LA138 S5 Moj S1 Moj S2 Redding 1-to-1 Line Linear (21.9 dB offset) 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 104 106 108 110 OBSI Level, dBA So un d Pr es su re L ev el , d BA US 30 BD US 30 TT US 30 UT US 30 AC US 30 LT LA138 S1 LA138 S2 LA138 S4 LA138 S5 Moj S1 Moj S2 Reddin g 1-to-1 Fi t Linear (21.8 dB offset)

23 The metrics for the SPB versus CPB cross-plots for the 25-ft and 50-ft and SRTT and Dunlop data are given in Table 10; the stan- dard deviations range from 0.7 to 1.0. Since the relationship of the OBSI data to the SPB data is linked to the correlation of the CPB results to the SPB data, the scatter between the normal- ized SPB and OBSI data was similar to the scatter reported in Table 8 between the CPB and OBSI data. In applying this approach to heavy trucks, the normalization coefficients developed for the light vehicles were applied directly in the SPB to OBSI comparison. Arguably, the effect of site-to-site variations may be different for trucks than light vehicles due to differences in effective source height. However, as with the light vehicles, normalizing the SPB data produced a reduction in the deviations about the 1-to-1 line with mini- mal change in offset for each microphone distance and test tire. Cross-plots of the SPB versus OBSI data for the SRTT mea- sured at 25 ft are shown in Figures 19 and 20 for the raw and normalized data, respectively. The metrics for 50 ft and the Dunlop tire are shown in Table 11. Unlike the light vehicle data, normalizing the truck data did not result in the regression line slope more closely approaching 1. Instead, the slope decreases even more because of the increase in the relative contributions from engine/ exhaust noise as the speed decreases as noted in the Refer- ence Energy Mean Emission Levels (REMELs) database (22). This would have the effect of causing higher overall levels at the lower speeds than levels due to tires alone resulting in a decreased slope. Comparing Tables 9 and 11, the offsets between the SPB and OBSI data for trucks are 8.9 to 9.3 dB less than for light vehicles indicating that trucks are much louder than light vehicles on average throughout the data range of the measurements. Further comparing the standard deviations for light vehicles and heavy trucks, it is seen that values for trucks are not much larger than those of light vehicles, typically no more than 0.2 dB. This suggests that the SPB levels could be estimated from OBSI data measured with either test tire with almost the same confidence for both light vehicles and heavy trucks. Prediction of SPB Data from OBSI Data The applicability of OBSI data to assessing the noise per- formance of in-service pavement is demonstrated by how well SPB levels can be predicted from OBSI data. To demonstrate Figure 18. Statistical light vehicle passby levels at 25 ft versus controlled vehicle passby level for the SRTT at all test sites and speeds. Table 9. Metrics for light vehicle SPB versus OBSI relationship. 25 ft Microphone Distance 50 ft Microphone Distance SRTT Dunlop SRTT Dunlop Cross-Plot Metrics Raw Norm Raw Norm Raw Norm Raw Norm Slope 1.31 0.92 1.29 0.93 1.32 0.91 1.32 0.91 r2 0.95 0.88 0.89 0.87 0.84 0.87 0.84 0.89 Offset, dB 21.9 21.8 23.9 23.8 28.5 28.3 30.5 30.3 Std Dev, dB 1.2 0.9 1.4 0.9 1.7 0.9 1.9 0.9 Avg Dev, dB 1.2 0.8 1.2 0.7 1.5 0.7 1.7 0.6 70 72 74 76 78 80 82 84 86 88 90 70 72 74 76 78 80 82 84 86 88 90 Controlled Passby Level, dBA Lt V eh ic le S ta tis tic al P as sb y Le ve l, dB A PCC 1 PCC 2 PCC 3 AC 4 PCC 5 AC 6 AC 7 AC 8 AC 9 PCC 10 PCC 11 AC 12 1-to-1 Line Linear (27 dB offset)

24 Figure 19. Statistical heavy-truck passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—raw data. Table 10. Metrics for CPB versus light vehicle SPB relationship. Figure 20. Statistical heavy truck passby levels at 25 ft versus OBSI level for the SRTT at all test sites and speeds—normalized data. 25 ft Microphone Distance 50 ft Microphone Distance Cross-Plot Metrics SRTT Dunlop SRTT Dunlop Slope 0.94 1.02 0.92 0.94 r2 0.93 0.96 0.94 0.96 Offset, dB 2.7 1.0 2.6 1.1 Std Dev, dB 0.9 0.7 1.0 0.8 Avg Dev, dB 0.7 0.6 0.7 0.7 80 82 84 86 88 90 92 94 96 98 10 0 96 98 100 102 104 10 6 1 08 11 0 1 12 114 116 OBSI Level, dBA So un d Pr es su re L ev el , d BA PCC 1 PCC 2 PCC 3 AC 4 PCC 5 AC 6 AC 7 AC 8 AC 9 PCC 10 PCC 11 AC 12 1-to-1 Line Linear (12.9 dB offset) 80 82 84 86 88 90 92 94 96 98 100 96 98 100 102 104 106 108 110 112 114 116 OBSI Level, dBA So un d Pr es su re L ev el , d BA PCC 1 PCC 2 PCC 3 AC 4 PCC 5 AC 6 AC 7 AC 8 AC 9 PCC 10 PCC 11 AC 12 1-to-1 Line Linear (12.9 dB offset)

25 this applicability in a less abstract manner than cross-plots, the offsets for the 1-to-1 lines from Tables 8 and 10 can be sub- tracted from the OBSI data for each tire. This yields a predicted SPB level based on either the SRTT or Dunlop tire at whatever OBSI test speed is selected. Depending on which offset is selected, the SPB levels for light vehicles or trucks at 25 ft or 50 ft can be predicted. Further, the predicted SPB levels can be compared to both the raw and normalized SPB levels. This is illustrated for the primary test speed of 60 mph in Figures 21 through 24. For light vehicle and the 25-ft microphone distance (Fig- ure 21), several features are noted. First, there is virtually no difference whether the predicted SPB levels are generated from the SRTT or Dunlop tires. Second, the normalized (measured) SPB values compare quite well to the predicted levels with an average difference of only 0.1 dB and standard deviations of 0.8 dB. Third, when site-to-site differences are not normalized, the variation between the predicted and measured SPB levels varies more with standard deviations of 1.2 and 1.4 dB depend- ing on the tire. For heavy trucks at 25 ft (Figure 22), the behav- ior is almost identical to that for the light vehicles with an aver- age difference of 0.1 dB and standard deviations of 0.7 and 0.8 dB for the normalized SPB data and 1.3 and 1.5 dB for the uncorrected SPB data. For light vehicles at the 50-ft microphone positions (Fig- ure 23), the results are similar to the 25-ft results except that a larger variance occurs between the predicted SPB and the uncorrected data. The normalized SPB maintains an average difference from the predicted of 0.1 dB with a standard devi- ation of 0.7 and 0.8 dB. For the uncorrected data, the standard deviations are 2.0 and 2.2 dB depending on the tire. This larger variance of the uncorrected 50-ft data is expected based on the average difference in level between the 25-ft and 50-ft micro- phone. These differences were found to vary as much as 3.6 dB between sites (see Appendix D). The same trends are seen for the trucks at 50 ft (Figure 24) with the exception of a larger standard deviation (0.9 to 1.0 dB) between the predicted and normalized SPB data. This SPB prediction methodology was applied to OBSI and SPB data obtained at other vehicle speeds and resulted Table 11. Metrics for heavy truck SPB versus OBSI relationship. Figure 21. Predicted SPB based on SRTT and Dunlop tires and measured light vehicle SPB levels at 60 mph and 25 ft—raw and normalized. 25 ft Microphone Distance 50 ft Microphone Distance SRTT Dunlop SRTT Dunlop Cross- Plot Metrics Raw Norm Raw Norm Raw Norm Raw Norm Slope 1.13 0.84 1.14 0.85 1.17 0.76 1.22 0.79 r2 0.82 0.89 0.77 0.85 0.81 0.79 0.78 0.77 Offset, dB 12.9 12.9 14.9 14.9 19.6 19.2 21.6 21.0 Std Dev, dB 1.4 0.9 1.6 1.0 1.6 1.2 1.8 1.2 Avg Dev, dB 1.2 0.7 1.3 0.8 1.3 1.0 1.5 1.1 70 75 80 85 90 95 100 PC C 1 PC C 2 PC C 3 AC 4 PC C 5 AC 6 AC 7 AC 8 AC 9 PC C 1 0 PC C 1 1 AC 12 N oi se L ev el , d BA SPB - Predicted Based on SRTT SPB - Predicted Based on Dunlop SPB - Measured Raw Data SPB - Measured Normalized Data

26 Figure 22. Predicted SPB based on SRTT and Dunlop tires and measured heavy truck SPB levels at 60 mph and 25 ft—raw and normalized. Figure 23. Predicted SPB based on SRTT and Dunlop tires and measured light vehicle SPB levels at 60 mph and 50 ft—raw and normalized. in essentially the same findings (details are provided in Appendix D). In general, SPB can be predicted from OBSI by subtracting the offset values established in this research using either of the two test tires. The offsets appropriate for each tire, vehicle type, and microphone distance are provided in Table 12 along with the standard deviations expected for such predictions. In applying these values using different test tires (SRTT or Dunlop design), it should be realized that standard deviations do not include differences that may be encountered from tire- to-tire variation. For each case, two standard deviations are given: one for the normalized SPB data and one for the uncor- rected SPB data. The first of these can be thought of as the stan- dard deviation that would be expected for an average of sites with the same pavement. The second standard deviation is that which should be applied to a specific site for those that are geo- metrically and acoustically in the range of the sites included in this research. For either the average or site-specific case, the off- 80 85 90 95 100 105 110 PC C 1 PC C 2 PC C 3 AC 4 PC C 5 AC 6 AC 7 AC 8 AC 9 PC C 1 0 PC C 1 1 AC 12 N oi se L ev el , d BA SPB - Predicted Based on SRTT SPB - Predicted Based on Dunlop SPB - Measured Raw Data SPB - Measured Normalized Data 60 65 70 75 80 85 90 PC C 1 PC C 2 PC C 3 AC 4 PC C 5 AC 6 AC 7 AC 8 AC 9 PC C 1 0 PC C 1 1 AC 12 N oi se L ev el , d BA SPB - Predicted Based on SRTT SPB - Predicted Based on Dunlop SPB - Measured Raw Data SPB - Measured Normalized Data

27 set between measured OBSI level and predicted SPB level is the same and only the expected accuracy varies. The standard devi- ations in Table 12 again indicate that SPB levels can be pre- dicted from the OBSI data with virtually the same level of confidence for both light vehicle and heavy trucks at a distance of 25 ft. Test Tires As noted in Figures 21 through 24, SPB levels predicted from OBSI using the SRTT and Dunlop tires are almost identical when the appropriate offset is used. From Table 11, the offsets for the Dunlop tire are 1.8 to 2.0 dB greater than for the SRTT, with the Dunlop producing higher noise levels. These tire dif- ferences are consistent with those measured for the OBSI parameter testing described in Chapter 4. The plot of passby level versus vehicle speed provided in Appendix D indicates that the Dunlop tire typically produced higher passby levels than the SRTT and generally approximated the levels of the light vehicle SPB more closely than the SRTT. In regard to truck SPB results, no evidence was found to suggest that the more aggressive tread of the Dunlop tire would better repre- sent truck SPB variation with pavement than would the SRTT. Since the two tires performed equally well in producing pre- dicted SPB levels for both light vehicles and heavy trucks, the decision on which test tire to be used in the OBSI procedure can be based on other, non-noise related issues (e.g., long-term availability). Summary In order to demonstrate the ability of the recommended OBSI test procedure to characterize the noise performance of in-service pavements, an extensive measurement program was completed. This program included the measurement of 1,343 light vehicle passby events and 539 heavy truck passby events at 12 sites and pavements in the states of Iowa and California, and measurements of controlled test vehicle passby events and OBSI. By comparing the CPB and OBSI data, significant site- to-site variation of up to 4.2 dB was identified. Site variation was Figure 24. Predicted SPB based on SRTT and Dunlop tires and measured heavy truck SPB levels at 60 mph and 50 ft—raw and normalized. Table 12. Offsets for predicting SPB from OBSI with expected standard deviations. 70 75 80 85 90 95 100 PC C 1 PC C 2 PC C 3 AC 4 PC C 5 AC 6 AC 7 AC 8 AC 9 PC C 1 0 PC C 1 1 AC 12 N oi se L ev el , d BA SPB - Predicted Based on SRTT SPB - Predicted Based on Dunlop SPB - Measured Raw Data SPB - Measured Normalized Data Offsets (to be subtracted from OBSI level), dB Application SRTT OBSI Dunlop OBSI Site Average Standard Deviation, dB Specific Site Standard Deviation, dB Light Vehicles at 25 ft 21.8 23.8 0.8 1.3 Heavy Trucks at 25 ft 12.9 14.9 0.8 1.4 Light Vehicles at 50 ft 28.3 30.3 0.8 2.1 Heavy Trucks at 50 ft 19.2 21.0 1.0 1.8

28 also indicated by average differences of up to 3.7 dB between the 25-ft and 50-ft passby levels cross the various sites. The site information from the CPB and OBSI data was used to normal- ize the SPB to OBSI data and establish a 1-to-1 relationship between them for each microphone distance and vehicle type. It was then demonstrated that these relationships could be used to effectively predict SPB results based on OBSI data for the average of sites included in the field testing with a standard deviation of 0.8 dB for both light vehicles and heavy trucks at a distance of 25 ft from the roadway. For 50 ft, a standard deviation of 0.8 and 1.0 dB was maintained for light vehicles and heavy trucks, respectively. Although the SRTT produced lower noise levels than the Dunlop test tire, essentially, no dif- ferences were found when using the data to predict SPB data. Some indication of noise sources other than tire-pavement was found for heavy trucks, particularly at lower speeds (below ∼60 mph) where engine/exhaust noise are expected to become more pronounced. However, the ability to predict SPB levels for heavy trucks from OBSI data alone was almost equal to that for light vehicles. Also, no issues were discovered that would limit the use of the 25-ft microphone distance for heavy trucks.

Next: Chapter 6 - Conclusions, Recommendations, and Suggested Research »
Measuring Tire-Pavement Noise at the Source Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 630: Measuring Tire-Pavement Noise at the Source examines a suggested procedure for measuring tire-pavement noise at the source using the on-board sound intensity (OBSI) method.

The following appendixes to the report are available online.

Appendix A: Review of Literature

Appendix B: Test Evaluation of Candidate Methods and Recommendation for Test Procedure Development

Appendix C: Results of Test Parameter Evaluation

Appendix D: Demonstration Testing of OBSI Procedure

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!