Beyond the Specifications: Best Practices for OBSI Measurement Dana M. Lodico, PE Lodico Acoustics LLC TRB 89 th Annual Meeting Washington D.C. January 10, 2010
Topics Results of OBSI parameter testing NCHRP 1 44 unless noted (Donavan and Lodico, 2008) Other Studies Mesa Rodeo Measurements (Donavan, 2006) Wind Tunnel Testing (Donavan and Lodico, 2004) Temperature Study (Bendtsen, Lu, and Kohler, 2009) Comparison of results to the AASHTO Procedure Quick discussion of things to look out for and the Data quality checks Further research NCHRP 1 44a, correlation between users, etc
Evaluation and Specification of OBSI Vehicle variables Test Variables Tire inflation pressure, wheel loading Tire variation, wear, & hardness discussed in next presentation, Tire type (SRTT) Environmental variables Air and pavement temperature, wind, air density Test execution variables Probe location, vehicle test speed, probe configuration, measurement reproducibility
Effect of Tire Inflation Pressure AASHTO Procedure : Cold inflation pressure of 30 ± 2 psi NCHRP 1 44: 0.4 db increase per 10 psi increase Range of 4 psi range of 0.16 db OBSI 26 42
Effect of Vehicle Loading AASHTO Procedure : For trailers, 850 ± 100 lbs NCHRP 1 44: +0.1 to 0.3 db per 100 lb increase Mesa Rodeo: Normalization based on NCHRP loading results improved correlation among users Baseline + 200 lbs
Effect of Air Temperature AASHTO Procedure : Currently no standard adjustment NCHRP 1 44: 0.026 db per o C(r 2 <0.42) Temperature Study: Recommends 0.027 db/ o Cair temperature correction for SRTT on asphalt 30 o C 40 o C 86 o F 104 o F
Effect of Pavement Temperature AASHTO Procedure : Currently no standard adjustment NCHRP 1 44: No clear correlation (r 2 <0.2) Temperature Study: Recommends 0.018 db/ o C pavement temperature correction for SRTT on asphalt 36 o C 60 o C 96 o F 140 o F
Wind Effect: Wind Tunnel Testing Direction of Travel Crosswind Direction + Crosswind Direction Horizontal Single Probe with Windscreen OBSI Fixture Horizontal Dual Probe with Commercial Windscreen Vertical Dual Probe with Windscreen (On Road)
Wind Induced Background Noise for Various Speeds & Crosswinds (Based on Wind Tunnel Testing) Sound Pressure Level, dba 105 100 95 90 85 80 75 70 60 mph with 11 mph (+) crosswind 56 km/hr, Vert. DP 72 km/hr, Vert. DP 97 km/hr, Vert. DP 113 km/hr, Vert. DP 56 km/hr, Horiz. DP 72 km/hr, Horiz. DP 97 km/hr, Horiz. DP 113 km/hr, Horiz. DP 56 km/hr, Single 72 km/hr, Single 97 km/hr, Single 113 km/hr, Single 65 60 +20 +10 0-10 -20 Yaw Angle, degrees *Based on Donavan and Lodico 2008
Sound Level, dba 95 90 85 80 75 70 Example of Wind Noise Contamination (60 mph with 11 mph + Crosswind) 60 mph, 11 mph +Crosswind Single Probe, SPL Vert. DP, SPL Horz. DP, SPL Single Probe, IL Vert. DP, IL Horz. DP, IL AC Pavement 65 60 55 Potential for Wind Noise Contamination 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000 1/3 Octave Band Center Frequency, Hz *Based on Donavan and Lodico 2008
Effects of Wind Condition AASHTO: Currently no limit on wind condition Wind Tunnel Results: In the absence of crosswind or in the presence of negative crosswind, all probe fixture configurations should measure OBSI without concerns of windinduced background noise from 400 to 5000 Hz Potential for error in tire noise OBSI measurements exists for positive crosswind conditions of 11 mph or greater at 60 mph For SPL, wind induced noise contamination would be expected except for a very limited range Use of SPL not advisable when microphones are exposed to airflow
Windscreen Configurations for Wind Tunnel Testing Commercial 3 ½ Windscreen Custom Shaped Windscreen
Wind Noise Reduction Devices AASHTO: Use of windscreens is required, no specification on type or size, noise cones not required Wind Tunnel Common, commercial 3½ diameter spherical windscreen performed better than shaped (smaller) windscreen below 500 Hz for SPL & IL No benefit of nose cones when 3½ diameter windscreen is used Not advisable to use nose cones only
Effect of Fixture Configuration
Dist from tire Fore/Aft Vertical Probe Location Distance from pavement AASHTO: 3 ± ¼ NCHRP: 0.4 db per ¼ upward movement Distance from tire AASHTO: 4 ± ¼ NCHRP: 0.2 db per ½ away from tire Fore/aft direction AASHTO: 8 ¼ ± ¼ apart NCHRP: No clear correlation
Effect of Vehicle Speed AASHTO Procedure : 60 ± 1 mph NCHRP 1 44: +0.3 db per 1 mph increase Range of 2 mph range of 0.6 db OBSI 56 mph 64 mph
Reproducibility AASHTO Procedure 2 runs minimum, 3 runs suggested Standard deviation of overall levels (400 to 5000 Hz) for 2 or more runs must be 0.6 db or less, 1/3 OB levels 1.2 db or less Run to run repeatability NCHRP 1 44: 0.8 db range & 0.3 db standard deviation for 10 consecutive runs Comparison of four vehicles max range of 0.6 db (AC), 0.8 db (PCC) User to user reproducibility Mesa Rodeo: Four Teams Average range 1.3 db, maximum range 2.2 db Normalizing for factors of tire hardness and loading: average range 1.1 db, maximum range 1.5 db Expected to improve as understanding of precision and bias improves (NCHRP 1 44a)
Data Quality Checks PI Index must be < 5 db in each ⅓ OB, 400 to 5000 Hz NCHRP 1 44: All 578 runs met criteria from 500 to 5000 Hz, 3% of runs exceeded criteria at 400 Hz (data discarded) SI must be positive, 400 to 5000 Hz Coherence must be 0.8, 400 to 4000 Hz NCHRP 1 44: <1% (4 out of 578) runs did not meet criteria due to equipment overloads, etc (data discarded)
AASHTO Instrumentation Requirements Test Parameter Vehicle Test speed Wheel Loading (Trailers) Tire Inflation Pressure (Cold) Tire Durometer Hardness Probe Distance from Pavement Probe Location, Fore/Aft Probe Distance from Tire Air/Pavement Temperatures AASHTO Requirements 60 ± 1 mph 850 ± 100 lbs 30 ± 2 psi 64 ± 2 hardness numbers 3 ± ¼ above pavement 8¼ ± ¼inch 4 ± ¼inch No standard adjustment
Data Quality Criteria Acoustic Data Checks Run to Run Range, Overall A Wtd OBSI level Run to Run Range, ⅓ Octave Band Levels (per OB) Coherence PI Index AASHTO Requirements Standard dev 0.6 db Standard dev 1.2 db > 0.8 for 400 to 4000 Hz < 5 db for 400 to 5000 Hz
Data Quality Parameters What does bad data look like?
Learning from Experience Some things to look out for Dirt or debris in the windscreen Strong crosswind conditions Interfering items in the measurement path (tape, etc) Passage of loud or heavy vehicle Loosening of microphone from preamp Extreme temperatures (equipment overload) Large reflecting objects nearby
References Measuring Tire Pavement Noise at the Source. Paul Donavan and Dana M. Lodico. NCHRP Report 630. National Cooperative Highway Research Program. Transportation Research Board. Washington 2009. Donavan, P., and Lodico, D., Wind Tunnel Assessment of Flow Induced Background Noise on On Board Tire Pavement Measurement, prepared for California Department of Transportation, Division of Environmental Analysis, Sacramento, CA, prepared by Illingworth & Rodkin, Inc., Petaluma, CA, February 2008 Donavan, P., Comparison of OBSI Measurements for Four Research Teams: The Mesa Rodeo, prepared for California Department of Transportation, Division of Environmental Analysis, Sacramento, CA, prepared by Illingworth & Rodkin, Inc., Petaluma, CA, January 2006. Bendtsen, H., Lu, Q., and Kohler, E., Temperature influence on road traffic noise Californian OBSI measurement study, March 2009. Standard Practice for Measurement of Tire/Pavement Noise Using the On Board Sound Intensity (OBSI) Method, Draft National Standard TP 78 10, American Association of State Highway and Transportation Officials, Washington D.C., January 2010.