February 2008
Research Report: UCPRC- RR- 2007- 03
IInnvveessttiiggaattiioonn ooff Nooiissee,, Duurraabbiilliittyy,,
Peerrmeeaabbiilliittyy,, aanndd Frriiccttiioonn Peerrffoorrmaannccee
Trreennddss ffoorr Asspphhaallttiicc Paavveemeenntt SSuurrffaaccee
Tyyppeess:: Fiirrsstt-- aanndd SSeeccoonndd-- Yeeaarr Reessuullttss
Authors:
Aybike Ongel, John T. Harvey, Erwin Kohler,
Qing Lu, and Bruce D. Steven
Partnered Pavement Research Center Strategic Plan Element No. 4.16: Investigation of Noise, Durability,
Permeability, and Friction Performance Trends for Asphaltic Pavement Surface Types
PREPARED FOR:
California Department of Transportation
Division of Research and Innovation
Office of Materials and Infrastructure
PREPARED BY:
University of California
Pavement Research Center
UC Davis, UC Berkeley
ii UCPRC- RR- 2007- 03
DOCUMENT RETRIEVAL PAGE Research Report: UCPRC- RR- 2007- 03
Title: Investigation of Noise, Durability, Permeability, and Friction Performance Trends for Asphaltic Pavement
Surface Types: First- and Second- Year Results
Authors: Aybike Ongel, John T. Harvey, Erwin Kohler, Qing Lu, and Bruce D. Steven
Prepared for:
Caltrans Division of
Research and Innovation
FHWA No:
CA091200A
Report Date:
February 2008
Work Submitted:
June 27, 2008
Contract No.:
65A0172
Status:
Final Stage 6, approved version
May 20, 2009
Version:
Final
Abstract: The central purpose of the research is to support the California Department of Transportation ( Caltrans)
Quieter Pavement Research Program, which has as its goals and objectives the identification of quieter, safer asphalt
pavement surfaces. The research conforms with Federal Highway Administration ( FHWA) guidance provided to
state departments of transportation ( DOTs) that conduct tire/ pavement noise research. Results from this research are
intended to identify best practices for selecting asphaltic surfaces based on performance trends identified from field
measurements for noise, permeability, friction, and durability.
This report evaluates the first two years of measurements of noise ( on- board sound intensity), permeability, skid
resistance ( friction), roughness, and surface distresses of the most common asphalt pavement surface types in
California: open- graded asphalt concrete, which includes conventional mixes ( OGAC), rubberized mixes ( RAC- O),
and F- mixes; rubberized gap- graded asphalt concrete ( RAC- G); and dense- graded asphalt concrete ( DGAC).
The sample of pavement surfaces in this study includes three age categories, two traffic types, and two rainfall
regions. This report presents results for the sections in the factorial experiment in the detailed Work Plan for this
project. This report also presents similar results for the Division of Environmental Analysis sections and other
special test sections, which are referred to as ES sections. This report presents all results for the factorial sections
and ES sections, some of which were presented in previous interim reports.
Conclusions are made regarding the performance of open- graded mixes and RAC- G compared with DGAC; the
variables affecting tire/ pavement noise; the correlation of laboratory absorption values with field- measured noise
levels; and the performance of mix types included in the study in addition to DGAC, OGAC, RAC- O, and RAC- G.
Preliminary recommendations are made for practice based on the results, and recommendations are made for future
work.
Keywords:
asphalt concrete, decibel ( dB), noise, absorption, macrotexture, microtexture, open- graded, gap- graded, dense-graded,
onboard sound intensity
Proposals for implementation:
No proposals for implementation are presented in this report.
Related documents:
Work Plan for project 4.16, “ Investigation of Noise, Durability, Permeability, and Friction Performance Trends for
Asphaltic Pavement Surface Types”
Signatures:
A. Ongel
First Author
J. Harvey
Technical Review
D. Spinner
Editor
J. Harvey
Principal
Investigator
T. Joe Holland
Caltrans Contract
Manager
UCPRC- RR- 2007- 03 iii
DISCLAIMER
The contents of this report reflect the views of the authors who are responsible for the facts and accuracy
of the data presented herein. The contents do not necessarily reflect the official views or policies of the
State of California or the Federal Highway Administration. This report does not constitute a standard,
specification, or regulation.
PROJECT OBJECTIVES
The research presented in this report is part of the California Department of Transportation ( Caltrans)
Quieter Pavement Research ( QPR) Work Plan. The central purpose of the research is to support the
Caltrans Quieter Pavement Research Program, which has as its goals and objectives the identification of
quieter, safer asphalt pavement surfaces. The program Road Map and Work Plan outline the tasks carried
out in this research.
The research conforms with Federal Highway Administration ( FHWA) guidance provided to
state departments of transportation that conduct tire/ pavement noise research.
Results from this research are intended to identify best practices for selecting asphaltic surfaces
based on performance trends identified from field measurements for noise, permeability, friction, and
durability. This work includes the following objectives:
1. Provide a literature survey of U. S. and European practice and research regarding the performance
of asphaltic surfaces.
2. Develop an operational capacity at the Partnered Pavement Research Center ( PPRC) to measure
field on- board sound intensity, laboratory noise impedance, and field surface friction.
3. Develop a database structure for field and laboratory measurements collected in this project.
4. Measure the properties of as- built surfaces ( sound intensity, permeability, friction, and distresses)
over time, with trends in data summarized annually. Continue data collection as authorized by
Caltrans. Measure the same properties for mixes from outside California and summarize data and
trends reported from outside California.
5. Perform statistical analyses on measurement results from the field and laboratory and report
performance trends and modeling results.
6. Prepare a report summarizing the work related to completion of the first five objectives.
This report presents results from completion of these objectives, including the first two years of
field measurements and laboratory test results.
iv UCPRC- RR- 2007- 03
CONVERSION FACTORS
SI* ( MODERN METRIC) CONVERSION FACTORS
APPROXIMATE CONVERSIONS TO SI UNITS
Symbol Convert From Multiply By Convert To Symbol
LENGTH
in. inches 25.4 millimeters mm
ft feet 0.305 meters m
AREA
in. 2 square inches 645.2 square millimeters mm2
ft2 square feet 0.093 square meters m2
VOLUME
ft3 cubic feet 0.028 cubic meters m3
MASS
lb pounds 0.454 kilograms kg
TEMPERATURE ( exact degrees)
° F Fahrenheit 5 ( F – 32)/ 9 Celsius C
or ( F – 32)/ 1.8
FORCE and PRESSURE or STRESS
lbf poundforce 4.45 newtons N
lbf/ in. 2 poundforce/ square inch 6.89 kilopascals kPa
APPROXIMATE CONVERSIONS FROM SI UNITS
Symbol Convert From Multiply By Convert To Symbol
LENGTH
mm millimeters 0.039 inches in.
m meters 3.28 feet ft
AREA
mm2 square millimeters 0.0016 square inches in. 2
m2 square meters 10.764 square feet ft2
VOLUME
m3 cubic meters 35.314 cubic feet ft3
MASS
kg kilograms 2.202 pounds lb
TEMPERATURE ( exact degrees)
C Celsius 1.8C + 32 Fahrenheit F
FORCE and PRESSURE or STRESS
N newtons 0.225 poundforce lbf
kPa kilopascals 0.145 poundforce/ square inch lbf/ in. 2
* SI is the symbol for the International System of Units. Appropriate rounding should be made to comply with Section 4 of ASTM E 380.
( Revised March 2003.)
UCPRC- RR- 2007- 03 v
EXECUTIVE SUMMARY
With traffic noise becoming a growing concern, the public is expecting “ quieter pavements” to be
constructed to abate traffic noise levels. Quieter pavement, a new concept intended to reduce the impact
that tire/ pavement noise has on the highway environment, offers new options for minimizing the impact
of traffic noise levels on neighborhoods adjacent to highways.
Over the past several years, the concept of using quieter pavement to reduce noise impact has
received increasing attention, in California and across the nation. Much of the research on queiter
pavements has generally established their short- term benefits; therefore, most of the recent attention has
turned to the development of a better understanding of long- term acoustic benefits. The California
Department of Transportation ( Caltrans) has initiated several studies to evaluate the acoustic properties of
pavements and pavement surface characteristics on tire/ pavement noise levels. The research presented in
this report is part of one of those studies and is part of the Caltrans Quieter Pavement Research ( QPR)
Work Plan.
The Caltrans QPR Plan is a systematic research proposal intended to examine the impact of
quieter pavements on traffic noise levels and to establish which pavement characteristics have the greatest
impact on tire/ pavement noise. The goal of the QPR program is to evaluate the acoustic properties and
performance characteristics of both flexible and rigid pavements and of bridge decks used by the state.
This report only covers flexible pavement surfaces. Additionally, the QPR program is intended to identify
surface treatments, materials, and construction methods that will result in quieter pavements that are also
safe, durable, and cost- effective. The information gathered will be used to develop quieter pavement
design features and specifications for noise abatement throughout California.
For the flexible pavement part of the QPR program, Caltrans identified the need for research in
the areas of acoustics, friction, and pavement performance of asphalt pavement surfaces for the state
highway network. In November 2004, the Caltrans Pavement Standards Team ( PST) approved a new
research goal for the Partnered Pavement Research Center ( PPRC) Strategic Plan; it was numbered
Element 4.16 and titled “ Investigation of Noise, Durability, Permeability, and Friction Performance
Trends for Asphaltic Pavement Surface Types.”
The central purpose of this research is to support the Caltrans QPR program Road Map and Work
Plan with goals and objectives that address identification of asphalt pavement surfaces that are both
quieter and safer. The research conforms with FHWA guidance provided to state departments of
transportation that conduct tire/ pavement noise research.
vi UCPRC- RR- 2007- 03
The objectives and deliverables from the Work Plan for this project completed with this report are
shown in the following table.
Work
Plan
Section Objective Deliverables
1.2.1 1. Literature survey
• Literature survey of U. S. practice, performed by PPRC
• Literature survey of European practice performed by Swiss
Federal Laboratories for Materials Testing and Research ( EMPA)
1.2.2 2. PPRC test
capability
PPRC operational capability to measure field sound intensity, lab
noise impedance, and field surface friction
1.2.3 3. Database structure
• Database structure
• Populated database at completion of data collection objective
( Objective 4)
4. Data collection
4a. Field sections in
California
• Properties of as- built surfaces ( sound intensity, permeability,
friction, and distresses) over time, with trends in data summarized
annually
• Continued data collection as authorized by Caltrans.
1.2.4
4b. Field and lab data
from outside
California
Database of measurements of outside mixes and report summarizing
data collected and trends
1.2.5 5. Performance- trend
statistical analysis
Report documenting statistical analysis results from data collection
objective ( Objective 4), with summary of performance trends and
modeling results
1.2.6 6. Two- year summary
report
Report summarizing all of the work completed; this document
constitutes the report
Results from this research are intended to identify best practices for selecting asphaltic surfaces
based on performance trends identified from field measurements for noise, permeability, friction, and
durability. This report presents the results from the entire two- year research study involving the pavement
noise, permeability, durability, roughness, surface distress, and friction characteristics of the most
common asphalt pavement surface types in California: open- graded asphalt concrete, which includes
conventional mixes ( OGAC), rubberized mixes ( RAC- O), and F- mixes; rubberized gap- graded asphalt
concrete ( RAC- G); and dense- graded asphalt concrete ( DGAC). The sample of pavement surfaces in this
study includes three age categories, two traffic types, and two rainfall regions.
Age categories consist of the following: less than one year old, one to four years old, and four to
eight years old. Traffic types are based on Caltrans 2004 annual average daily traffic ( AADT) data for
highways and freeways. Traffic was categorized as high if the AADT ( two- way) was greater than 32,000
vehicles per day, with lower amounts categorized as low. Rainfall is based on annual average rainfall in
California from 1960 to 1990, with amounts greater than 620 mm ( 24.4 inches) categorized as high and
smaller quantities as low.
UCPRC- RR- 2007- 03 vii
The pavement sections in the experiment are referred to as QP sections. This experiment design
includes 52 QP sections and 10 Stantec sections ( identified as 01- N* and 06- N*). The F- mixes were
placed only in low- trafficked areas, so traffic levels could not be evaluated for these sections. Similar
results for 23 Division of Environmental Analysis pavement sections and other special test sections
( referred to as ES sections) are also included in this report.
UCPRC field crews visited each of the test sections twice with Caltrans Maintenance traffic
closures — approximately one year apart with a rainy season in between — and collected data on
permeability, skid resistance ( friction), roughness, noise, and surface distress. Cores were also taken from
each 125- or 150- m ( 410- or 492- ft) section. The cores were used to determine air- void content and were
burned in an ignition oven to obtain aggregate gradations.
The measured results and the qualitative and statistical analyses from this testing program are
documented in this report. This information is organized in this report as follows:
• Chapter 1 introduces the QPR study, presents the background to the study, the objectives, and the
performance parameters for pavement surfaces.
• Chapter 2 provides a summary literature survey covering U. S. and European sources.
• Chapter 3 presents the methodology used for the study.
• Chapter 4 describes the independent variables and abbreviations used in the analysis.
• Chapter 5 presents an analysis of the permeability and air- void content data and an evaluation of
clogging.
• Chapter 6 contains an analysis of skid resistance data, including data on microtexture and
macrotexture.
• Chapter 7 provides an analysis of ride- quality data in terms of the International Roughness Index
( IRI).
• Chapter 8 presents an analysis of the condition survey data.
• Chapter 9 presents an analysis of the on- board sound intensity ( OBSI) data collected on the test
sections, laboratory sound absorption data from field cores, and the correlation between OBSI
and absorption.
• Chapter 10 presents an analysis of the data collected on the ES sections.
• Chapter 11 presents an overall evaluation of the performance models developed in this study, and
an assessment of the life spans of the different surface mixes for different conditions and failure
criteria based on the models.
• Chapter 12 lists the conclusions from the analyses and includes preliminary recommendations.
• Appendices provide various data corrections used and detailed condition survey information.
viii UCPRC- RR- 2007- 03
The data presented in this report is included in a relational database that will be delivered to Caltrans
separately. Specific data in the database includes:
• Microtexture and macrotexture data that affect skid resistance;
• Ride quality in terms of International Roughness Index ( IRI);
• On- board sound intensity ( OBSI), a measure of tire/ pavement noise;
• Sound intensity for different frequencies; and
• Surface distresses, including bleeding, rutting, raveling, transverse cracking, and cracking in the
wheelpaths;
• Climate data; and
• Traffic data.
The analyses presented for each performance variable in Chapters 5 through 9 include a summary of
the expected trends from the literature, descriptive statistics, and where the data is sufficient, statistical
models. Several appendices provide data corrections used and detailed condition survey information. The
data presented in this report is included in a relational database that will be delivered to Caltrans
separately.
Conclusions
In this report, the performance of open- graded mixes as well as of other asphaltic mixes used in
California was evaluated in terms of safety, noise, and durability. Data was collected on the OGAC,
RAC- O, RAC- G, and DGAC mixes as well as on some new mixes such as bonded wearing course
( BWC), gap- graded asphalt rubber ( RUMAC- GG), gap- graded rubber modified asphalt ( Type G- MB),
and dense- graded rubber modified asphalt ( Type D- MB). The objectives of this study were to:
• Evaluate the durability and effectiveness of open- graded mixes in increasing safety and reducing
noise compared to other asphalt surfaces
• Determine the pavement characteristics that affect tire/ pavement noise
• Correlate sound absorption with tire/ pavement noise
• Evaluate the performance of new mixes compared to the asphaltic mixes currently used in
California
Performance of Open- Graded Mixes
The results showed that current open- graded mixes reduced tire/ pavement noise compared to the
dense- graded mixes included in the study by almost 2 dB ( A) on average for all sections over the eight-
UCPRC- RR- 2007- 03 ix
year range of ages, which according to the literature is near the limit of what the human ear can discern.
Twenty- five percent of the open- graded mixes provided noise reduction above 3 dB ( A) compared to the
average noise level of a DGAC mix, which is 104 dB ( A) for the sections tested. Over the entire set of
sections including all ages, the open- graded mix noise levels were between 1 dB ( A) greater and 4 dB ( A)
less than the average DGAC noise level. The noise levels of the DGAC mixes in the study were similar
across all ages of pavement.
Noise reductions between 2 dB ( A) and 6 dB ( A) were reported in the literature for open- graded
mixes. The results presented in this report are from comparisons between different surfaces of similar
ages. Greater noise reductions would be expected when new open- graded surfaces are placed on existing
DGAC surfaces that have widespread and severe distresses than when comparing the noise levels of
open- graded mixes with those of DGAC surfaces of similar ages as was done in this study.
Also note that noise levels above 1,000 Hz are generally considered more annoying, and that
increasing air- void content and increasing macrotexture reduce the noise levels at higher frequencies.
Since open- graded mixes have higher air- void content and macrotexture, they may reduce the noise levels
at higher frequencies and so may be perceived by the human ear as quieter and the noise as less annoying
than dense- graded mixes, even though the overall A- weighted noise levels are not significantly different
from each other.
Open- graded mixes have higher permeability and friction than dense- graded and RAC- G mixes;
therefore, they can reduce hydroplaning and spray and splash and hence improve safety. Based on the
results of the conditions surveys for pavements less than nine years old included in this study, they also
may be less prone to transverse cracking. However, although it could not be revealed statistically in this
study, it is expected that open- graded mixes would be more prone to raveling since their high
permeability would be expected to increase the oxidation rate of the binder, in comparison to the less
permeable DGAC and RAC- G mixes.
Open- graded mixes lose their noise- reducing properties with time mainly due to clogging and also
due to the presence of distresses on the pavement surface. The work in this study predicts their noise
levels to reach those of dense- graded mixes within seven years. Clogging occurs at the top part of the
surface layer and reduces its permeability. There is also some indication that thicker mixes, above 50 mm,
may be less clogged and hence have higher permeabilities than thinner mixes. The longevity of benefits
provided by open- graded mixes varies with mix properties, rainfall, and presence of raveling.
From this small sample of pavements in California, there does not appear to be a major difference in
performance between RAC- O and OGAC mixes with respect to noise and permeability benefits across the
age ranges. However, the rate of increase in IRI is slower for rubberized mixes. Although the data was not
x UCPRC- RR- 2007- 03
statistically significant in this study, rubberized mixes tended to have better cracking performance, which
would be expected to slow the rate of noise in later years.
Performance of RAC- G Mixes
It appears from the data that RAC- G mixes provide some noise benefit compared to DGAC mixes.
Most of the noise benefits from RAC- G appear to come from the fact that they have higher air- void
content than DGAC mixes when they are built ( compaction of RAC- G is by method specification, where
the compaction method is specified and the relative density is not specified, rather than by an end- result
specification where the relative density is specified and the contractor chooses the compaction method.).
However, they lose their permeability faster than the open- graded mixes, and hence their noise- reducing
properties. Based on the descriptive statistics, the noise levels from RAC- G mixes appear to approach
those of DGAC within four years. The sound intensity model overpredicts the noise levels of RAC- G
mixes; therefore, this model cannot be used to estimate the lifetime of RAC- G mixes in terms of noise
reduction.
Variables Affecting Tire/ Pavement Noise
The study showed that tire/ pavement noise is greatly influenced by pavement surface characteristics
such as gradation, macrotexture, age, and presence of distresses. Coarser gradation and increasing air-void
content reduce the overall noise levels, and the presence of distresses and increasing macrotexture
and age increase the overall noise levels, confirming the previous findings of other researchers. However,
this study found that the overall A- weighted noise levels are insensitive to changes in air- void content for
open- graded mixes with air- void content above 15 percent. This insensitivity occurs because air- void
content above 15 percent is usually associated with higher macrotexture ( MPD) values, and for large
texture depths, increasing air- void content does not reduce the overall noise levels, and its effects are
surpassed by those of increased tire vibrations.
Since California mixes are placed in thin layers ( around 30 mm), thickness was not found to affect the
noise levels of the sections studied. However, there is some indication that increasing thickness may
lower the noise levels for thicknesses above 50 mm ( 2 inches).
The pavement temperature was not found to significantly affect the noise levels.
The use of rubber asphalt binders was also not found to significantly affect the noise levels, although
the noise levels of RAC- O mixes were somewhat less than those of OGAC mixes.
The low frequencies of tire/ pavement noise were found to be governed by tire vibrations due to high
macrotexture, and the higher frequencies were found to be governed by air- pumping mechanisms that can
be reduced by the presence of air voids on the pavement surface, confirming the findings of previous
researchers on the noise- generation mechanisms. However, increasing air- void content was found to
UCPRC- RR- 2007- 03 xi
increase the noise levels at a given macrotexture at lower frequencies, probably due to increased tire
vibrations.
At frequencies around 800 to 1,000 Hz, where the tire/ pavement noise is highest, the air- pumping
cannot be reduced by increasing air- void content above 15 percent, and tire vibrations govern the noise
generation for mixes with high air- void content and high macrotexture values. This trend can also be seen
in the overall noise levels.
At frequencies above 1,000 Hz, higher air- void content and higher macrotexture values reduce the
air- pumping noise. Therefore, open- graded mixes have significantly lower noise levels at frequencies
above 1,000 Hz.
Correlation of Absorption Values with Noise Levels
The noise levels of dense- and gap- graded mixes decrease with increasing absorption. However, no
correlation was found between the overall A- weighted sound intensity and absorption for open- graded
mixes. Correlations between sound intensity ( noise) measured in the field and laboratory absorption
values depended on frequency. Noise levels around 500 Hz are governed by tire vibrations; therefore,
absorption has no effect on the noise levels for any mix type. At frequencies above 630 Hz, absorption
reduces the noise levels caused by air pumping for dense- and gap- graded mixes, and there are clear
trends relating noise to absorption.
Tire vibrations may cause significant noise levels for open- graded mixes with high macrotexture
values at lower frequencies ( less than 1,000 Hz), and there is no trend between noise and absorption. The
noise- reducing effect of absorption can be seen at 1,000 Hz for open- graded mixes, if macrotexture is also
considered. The noise- reducing effects of absorption can be clearly seen at frequencies above 1,000 Hz
for open- graded mixes. Air- pumping noise governs noise generation at frequencies above 1,000 Hz,
confirming the earlier findings, as increasing absorption reduces the noise levels regardless of the
macrotexture values. This trend is stronger for higher frequencies, which are considered more annoying to
humans.
Performance of New Mixes
The bituminous wearing course ( BWC) mix placed on the LA 138 sections has lower permeability
and friction, higher noise levels, and almost the same distress development as current Caltrans open-graded
mixes in the LA 138 section study.
Based on the Fresno 33 ( Firebaugh) sections, the RUMAC- GG and Type G- MB mixes did not
perform as well the RAC- G mix when placed in thin lifts ( 45 mm); the RUMAC- GG and Type G- MB
mixes have higher noise levels and are more susceptible to bleeding. However, RUMAC- GG was more
xii UCPRC- RR- 2007- 03
crack resistant when placed in thick layers ( 90 mm). Type D- MB, which may be a candidate as an
alternative to dense- graded mixes after further investigation, has performance characteristics very similar
to those of DGAC mixes, and it may provide better crack resistance; however, it was more susceptible to
bleeding.
The European gap- graded ( EU- GG) mix placed on LA 19 has performance characteristics very
similar to those of gap- graded mixes ( RAC- G) used in California.
F- mixes have been used only in a wet environment on the north coast. Indications are that they do not
perform as well as OGAC and RAC- O with regard to noise, probably because of their large NMAS values
and raveling.
Other Conclusions
Note that the conclusions presented here are valid within the range of the air- void content, thickness,
age, and gradation properties of the mixes used in this study and under California climate and traffic
conditions. The OBSI measurements were conducted using an Aquatread 3 tire and a passenger car. The
conclusions may differ for trucks and vehicles with different tires as noise- generation mechanisms are
highly dependent on the vehicle and tire type. Also note that the OBSI method is a near- source
measurement; therefore, it captures only the tire/ pavement noise. Since the noise levels next to highways
are also affected by noise propagation and noise absorption under propagation, the greater absorption
values measured as part of this project may indicate that the open- graded mixes provide higher levels of
noise reduction at the side of the highway than these results may show.
The comparison of pass- by measurements made by Volpe ( unverified by UCPRC with regard to wind
speeds and other factors) with OBSI measurements indicated that absorption may provide additional noise
reduction next to highways since 75- mm OGAC shows higher noise reduction than dense- graded mixes
when measured using the pass- by method. However, the pass- by measurements found no additional noise
reduction for the 30- mm OGAC and RAC- O sections.
The effects of NMAS and thickness could not be fully evaluated as these variables have different
specifications for different mix types. Open- graded mixes have NMAS values of 9.5 and 12.5 mm, and
dense- and gap- graded mixes have NMAS values of 12.5 and 19 mm. F- mixes are the only open- graded
mixes with an NMAS value of 19 mm. Open- graded mixes are placed in thin layers, while RAC- G and
DGAC mixes are usually placed in a thicker lift. RAC- G mixes are usually placed at half the thickness of
DGAC mixes as the rubber content allows for reduced thickness, providing structural and reflection crack
retardation equivalency. Therefore, NMAS and thickness effects were identified only within each mix
type. Also, rubberized mixes are usually overlays of pavements with more extensive cracking than the
UCPRC- RR- 2007- 03 xiii
pavements on which DGAC mixes are placed. Therefore, the effects of rubber on crack retardation could
not be fully evaluated.
The effects of stiffness on noise levels were evaluated by comparing the noise levels of rubberized
and nonrubberized mixes as well as by comparing shear modulus values with noise levels. The study’s
preliminary conclusion is that stiffness does not play a major role in determining noise levels for mixes of
the types included in this study.
Recommendations
Based on the findings, none of the other asphalt mix types evaluated in this study can provide an
alternative to current Caltrans open- graded mixes in terms of noise reduction and safety. However,
durability of open- graded mixes compared to other mix types depends on the climatic conditions and
traffic.
The results indicate that the current recommendation for the best approach to noise reduction is to use
thin layers of open- graded mixes with nominal maximum aggregate sizes of 12.5 or 9.5 mm. The smaller
aggregate sizes will somewhat reduce air- void content and permeability; however, open- graded mixes
with smaller aggregate sizes will likely have greater durability because of their lower air- void content and
will likely cost less than open- graded mixes with larger aggregate sizes because they can be constructed
as thinner lifts. The results indicated that the desired air- void content for open- graded mixes for noise
reduction could be limited to a maximum of 15 percent since higher air- void content does not provide any
additional noise reduction and reduces durability. Mixes with lower air- void content would also be more
resistant to clogging.
There do not appear to be noise- reduction benefits from increasing the thickness of open- graded
mixes for thicknesses less than 50 mm. However, the results gave some indication that thicknesses greater
than 50 mm ( 2 inches) reduce noise. Placing open- graded mixes in thicker lifts would also help reduce the
IRI value and increase cracking resistance for overlays of PCC. The results also gave some indication that
thicker lifts may be less susceptible to clogging.
Open- graded mixes have longer lives in terms of noise and permeability with low levels of truck
traffic and rainfall. High truck traffic increases clogging, and mixes under low rainfall are also more
susceptible to clogging, although they are less likely to show raveling and polishing. When placing open-graded
mixes, the air- void content and thickness will need to be balanced with the permeability
requirements needed to reduce hydroplaning for a given site.
Overall preliminary recommendations for open- graded mix design based on the results of this study
are shown in the table below. These recommendations are also the basis for recommendations for further
xiv UCPRC- RR- 2007- 03
work to improve the performance of open- graded mixes, discussed in the next section of this report
( Section 12.3).
Table Preliminary Recommendation for Open- Graded Mix and Thickness Design
to Achieve Performance Goals
Performance Criteria
( relevant section of report)
Mix and
Thickness
Design
Variables
Noise
( Sections 9.1.4
and 9.2.3)
Permeability
( Section 5.4)
Durability**
( Section
8.6)***
Ride Quality
( Section 7.3)
Friction
( Section 6.7)
Air- Void
Content
15 percent or
greater
Maximize* Minimize Maximize
Nominal Max
Aggregate
Size
Minimize 12.5 mm
instead of 9.5
mm
Maximize
Gradation Greater
fineness
modulus
( coarser
gradation)
Greater
fineness
modulus
( coarser
gradation)
Greater
fineness
modulus
( coarser
gradation)
Binder Type Rubberized Rubberized
Overlay
Thickness
Greater than 50
mm may help
* Permeability recommendations should be based on expected rainfall events for a particular project
location. Development of these criteria are outside the scope of this project.
** Durability is defined as resistance to distress development.
*** Few sections had significant distresses, and results were not statistically significant.
Recommendations regarding durability are based on judgment as well as the results of this study.
Recommendations for Further Work
In this study, pavement characteristics and noise were observed for two years. However, two
years is a short time to observe any trends. Therefore, permeability, friction, IRI, and sound intensity
measurements and condition surveys should be conducted on the given sections for at least two or three
years to develop better time histories and see more sections reach failure.
Open- graded mixes have lower noise levels than dense- and gap- graded mixes at higher
frequency levels, which may be a benefit that A- weighted measurement does not capture well in terms of
annoyance rather than audibility. Since the human ear is most sensitive at frequencies between 1,000 and
4,000 Hz, the open- graded mixes may be perceived as quieter than dense- graded mixes with the same
overall noise levels. The noise levels should be correlated with the human perception of annoyance to
better evaluate noise- mitigation strategies.
Since the reason for placing open- graded mixes is to reduce the noise levels next to highways, the
way- side measurements should be better correlated with OBSI levels than was possible in this study to
understand the actual noise reduction provided by open- graded mixes.
UCPRC- RR- 2007- 03 xv
At 500 Hz, increasing air- void content was found to increase noise levels along with
macrotexture; however, the noise- generation mechanism is unknown. The further effects of air- void
content on noise levels at lower frequencies should be evaluated. In addition, a new parameter that
correlates better with the sound intensity levels should be developed. This parameter can be a
combination of MPD, RMS, and air- void content as well as a new measure of macrotexture.
The results gave some indication that open- graded mixes with finer gradations ( lower fineness
modulus) may provide lower noise levels, particularly at higher frequencies of noise. In this study, only a
few open- graded mixes had fine gradations. The effects of fineness modulus on the noise levels should be
further evaluated, particularly for mixes with the same NMAS.
This study could not fully evaluate the effects of NMAS and thickness on pavement performance.
Therefore, a laboratory study should be performed to consider the durability, sound absorption ( correlated
with high- frequency noise), and permeability for a full factorial experiment considering these variables.
Some optimization of the mixes based on initial results should also be performed. Since the presence of
polymer- modified binders could not be identified for the OGAC sections in this study because of a lack of
reliable as- built records for many sections, polymer and conventional binders, as well as rubberized
binders used by Caltrans, should also be included in the factorial. Macrotexture should also be measured,
since the results indicate that absorption and macrotexture provide an indication of noise at 1,000 Hz.
The results of the laboratory study will provide a basis for designing a factorial for field- test
sections to verify the laboratory results regarding the effects of thickness, NMAS, fineness modulus, and
binder type on clogging, cracking, and noise levels. Permeability and noise measurements as well as
condition surveys should be conducted on these test sections. The air- void content should also be
measured using CT scans with a higher resolution than used in this study. A resolution around 15 microns
( based on the results of this study) would be enough to see fine particles clogging the mix. The effects of
pavement temperature on noise levels were evaluated measuring nine sections at three temperatures. No
correlation was found between pavement temperatures and noise levels. A larger data set, with open-,
gap-, and dense- graded mixes, should be obtained, and measurements should be conducted using a wider
range of pavement temperatures. It would be useful to analyze the effects of pavement temperature on
noise levels separately for each mix type.
xvi UCPRC- RR- 2007- 03
UCPRC- RR- 2007- 03 xvii
TABLE OF CONTENTS
PROJECT OBJECTIVES..................................................................................................................... ... iii
CONVERSION FACTORS ...................................................................................................................... iv
EXECUTIVE SUMMARY ........................................................................................................................ v
LIST OF TABLES ............................................................................................................................... . xxiii
LIST OF FIGURES ............................................................................................................................... xxv
LIST OF ABBREVIATIONS ............................................................................................................... xxxi
1 INTRODUCTION................................................................................................................... .......... 1
1.1 Background..................................................................................................................... ................ 1
1.2 Quieter Pavement Research Objectives Addressed in This Report ................................................. 2
1.3 Traffic Noise ............................................................................................................................... .... 3
1.4 Desired Properties of Open- Graded Mixes...................................................................................... 6
1.4.1 Performance of Open- Graded Mixes .................................................................................. 7
1.4.1.1 Permeability................................................................................................................... 7
1.4.1.2 Skid Resistance ( Friction).............................................................................................. 7
1.4.1.3 Roughness...................................................................................................................... 8
1.4.1.4 Durability..................................................................................................................... . 8
1.4.1.5 Noise Reduction ............................................................................................................ 8
1.5 Contents of This Report ................................................................................................................... 9
2 LITERATURE SURVEY................................................................................................................ 11
2.1 Permeability ............................................................................................................................... ... 11
2.2 Skid Resistance ( Friction).............................................................................................................. 14
2.3 Roughness ( Unevenness) and Ride Quality................................................................................... 18
2.4 Pavement Distresses..................................................................................................................... . 20
2.4.1 Bleeding....................................................................................................................... .... 20
2.4.2 Rutting .............................................................................................................................. 20
2.4.3 Transverse Cracking ......................................................................................................... 21
2.4.3.1 Transverse Cracking Caused by Thermal Stresses ...................................................... 21
2.4.3.2 Transverse Cracking Caused by Reflection Cracking ................................................. 22
2.4.4 Raveling....................................................................................................................... .... 22
2.4.5 Fatigue Cracking............................................................................................................... 23
2.5 Highway Noise .............................................................................................................................. 24
2.5.1 Pavement Variables Affecting Noise................................................................................ 27
2.5.1.1 Texture........................................................................................................................ 27
xviii UCPRC- RR- 2007- 03
2.5.1.2 Roughness ( Unevenness)............................................................................................. 29
2.5.1.3 Air- Void Content ( Porosity)........................................................................................ 29
2.5.1.4 Stiffness ....................................................................................................................... 32
2.5.1.5 Age .............................................................................................................................. 32
2.5.1.6 Temperature................................................................................................................. 33
3 METHODOLOGY.................................................................................................................... ...... 35
3.1 Site Selection ............................................................................................................................... . 35
3.1.1 Experimental Design of the Test Sections ........................................................................ 35
3.2 Data Collection .............................................................................................................................. 51
3.2.1 Coring ............................................................................................................................... 53
3.2.2 Condition Survey .............................................................................................................. 55
3.2.3 Permeability ...................................................................................................................... 55
3.2.4 Friction....................................................................................................................... ...... 56
3.2.5 Air Temperature and Pavement Temperature................................................................... 57
3.2.6 On- Board Sound Intensity ................................................................................................ 57
3.2.7 International Roughness Index ( IRI) ................................................................................ 58
3.2.8 Macrotexture................................................................................................................... . 59
3.2.9 Air- Void Content .............................................................................................................. 60
3.2.10 Aggregate Gradation......................................................................................................... 61
3.2.11 Extent of Clogging............................................................................................................ 61
3.2.12 Acoustical Absorption ...................................................................................................... 62
4 ANALYSIS OF INDEPENDENT VARIABLES........................................................................... 65
4.1 Variable Definitions.................................................................................................................... .. 65
5 EVALUATION OF PERMEABILITY AND AIR- VOID CONTENT OF ASPHALT
CONCRETE MIXES ....................................................................................................................... 77
5.1 Permeability Analysis .................................................................................................................... 78
5.1.1 Descriptive Analysis ......................................................................................................... 78
5.1.2 Regression Analysis.......................................................................................................... 81
5.2 Air- Void Content Analysis ............................................................................................................ 85
5.2.1 Descriptive Analysis ......................................................................................................... 85
5.2.2 Air- Void Content Determination by CT Scan .................................................................. 89
5.3 Evaluation of Clogging .................................................................................................................. 91
5.3.1 Field Permeability Measurements..................................................................................... 91
5.3.2 CT Scan........................................................................................................................... . 97
UCPRC- RR- 2007- 03 xix
5.4 Summary of Findings................................................................................................................... 101
6 EVALUATION OF SKID RESISTANCE ( FRICTION) OF ASPHALT
CONCRETE MIXES ..................................................................................................................... 105
6.1 Microtexture................................................................................................................... ............. 105
6.2 Descriptive Analysis .................................................................................................................... 106
6.3 Correction of British Pendulum Numbers for Temperature ........................................................ 110
6.4 Statistical Modeling of Microtexture ........................................................................................... 111
6.5 Macrotexture................................................................................................................... ............ 116
6.5.1 Descriptive Analysis ....................................................................................................... 117
6.5.2 Statistical Modeling of Macrotexture ............................................................................. 120
6.6 International Friction Index ( IFI)................................................................................................. 123
6.7 Figure 66: Comparison of F60 values for different mix types at different ages............................ 125
6.7 126
6.8 Summary of Findings................................................................................................................... 126
7 EVALUATION OF ROUGHNESS OF ASPHALT CONCRETE MIXES .............................. 129
7.1 Descriptive Analysis .................................................................................................................... 129
7.2 Regression Analysis..................................................................................................................... 132
7.3 Summary of Findings................................................................................................................... 137
8 PAVEMENT DISTRESS EVALUATION................................................................................... 139
8.1 Bleeding ............................................................................................................................... ....... 139
8.2 Rutting ............................................................................................................................... ......... 140
8.3 Transverse and Reflection Cracking ............................................................................................ 142
8.3.1 Descriptive Analysis ....................................................................................................... 142
8.3.2 Regression Analysis........................................................................................................ 144
8.4 Raveling ............................................................................................................................... ....... 146
8.4.1 Descriptive Analysis ....................................................................................................... 146
8.4.2 Regression Analysis........................................................................................................ 148
8.5 Wheelpath Crack Initiation .......................................................................................................... 150
8.5.1 Descriptive Analysis ....................................................................................................... 150
8.5.2 Survival Analysis............................................................................................................ 151
8.6 Summary of Findings................................................................................................................... 154
9 ANALYSIS OF ACOUSTICAL PROPERTIES OF ASPHALT MIXES................................. 157
9.1 On- Board Sound Intensity Levels................................................................................................ 157
9.1.1 Descriptive Analysis ....................................................................................................... 157
xx UCPRC- RR- 2007- 03
9.1.2 Single- Variable Regression Analysis ............................................................................. 163
9.1.3 Principal Components Regression .................................................................................. 177
9.1.4 Summary of Findings...................................................................................................... 183
9.2 One- Third Octave Band Analysis of Sound Intensity Levels...................................................... 185
9.2.1 Evaluation of Pavement Surface Effects on the Spectral Content of Sound Intensity
Levels 186
9.2.2 Evaluation of Pavement Temperature Effects on the Frequency Content of Sound
Intensity Levels ............................................................................................................................... 207
9.2.3 Summary of Findings...................................................................................................... 213
9.3 Evaluation of Acoustical Absorption Values............................................................................... 215
9.3.1 Descriptive Analysis ....................................................................................................... 215
9.3.2 Correlation of Acoustical Absorption Values with Air- Void Content and Surface
Thickness ............................................................................................................................... ........ 217
9.3.3 Correlation of Absorption Values with A- Weighted Sound Intensity Levels ................ 219
9.3.4 Correlation of Sound Absorption Values with One- Third Octave Band Frequency
Sound Intensity Levels ..................................................................................................................... 222
9.3.4.1 Sound Intensity Levels for 500- Hz One- Third Octave Band .................................... 222
9.3.4.2 Sound Intensity Levels for 630- Hz One- Third Octave Band .................................... 224
9.3.4.3 Sound Intensity Levels for 800- Hz One- Third Octave Band .................................... 225
9.3.4.4 Sound Intensity Levels for 1,000- Hz One- Third Octave Band ................................. 227
9.3.4.5 Sound Intensity Levels for 1,250- Hz One- Third Octave Band ................................. 230
9.3.4.6 Sound Intensity Levels for 1,600- Hz One- Third Octave Band ................................. 231
9.3.5 Summary of Findings...................................................................................................... 233
10 ANALYSIS OF ENVIRONMENTAL NOISE MONITORING SITE SECTIONS................. 235
10.1 Fresno 33 Sections ....................................................................................................................... 235
10.2 Sacramento 5 and San Mateo 280 Sections ................................................................................. 242
10.3 LA 138 Sections....................................................................................................................... ... 251
10.4 LA 19 Section .............................................................................................................................. 261
10.5 Yolo 80 Section ........................................................................................................................... 265
10.6 Summary of Environmental Noise Monitoring Site Sections Analysis....................................... 269
11 EVALUATION OF PERFORMANCE MODELS AND PREDICTION OF LIFETIME
FOR DIFFERENT ASPHALT MIX TYPES............................................................................... 273
11.1 Evaluation of Permeability and Clogging Models....................................................................... 273
11.2 Evaluation of Microtexture Models ............................................................................................. 275
UCPRC- RR- 2007- 03 xxi
11.3 Evaluation of Roughness Models ................................................................................................ 278
11.4 Evaluation of On- Board Sound Intensity Model ......................................................................... 280
11.5 Prediction of Lifetime for Different Asphalt Mix Types ............................................................. 281
12 CONCLUSIONS, RECOMMENDATIONS, AND RECOMMENDATIONS FOR
FURTHER WORK ........................................................................................................................ 285
12.1 Conclusions.................................................................................................................... ............. 285
12.1.1 Performance of Open- Graded Mixes .............................................................................. 285
12.1.2 Performance of RAC- G Mixes ....................................................................................... 286
12.1.3 Variables Affecting Tire/ Pavement Noise ...................................................................... 287
12.1.4 Correlation of Absorption Values with Noise Levels..................................................... 287
12.1.5 Performance of New Mixes ............................................................................................ 288
12.1.6 Other Conclusions........................................................................................................... 288
12.2 Recommendations................................................................................................................ ....... 289
12.3 Recommendations for Further Work ........................................................................................... 291
REFERENCES..................................................................................................................... .................. 293
APPENDIX A: Correction of OBSI Values for Speed ( from 35 mph to 60 mph)............................. 304
APPENDIX B: Air- Density Correction ................................................................................................ 305
APPENDIX C: Regression Analysis for Each Frequency Level ........................................................ 306
APPENDIX D: Condition Survey of Environmental Noise Monitoring Site Sections for
Two Years ............................................................................................................................... ....... 317
xxii UCPRC- RR- 2007- 03
UCPRC- RR- 2007- 03 xxiii
LIST OF TABLES
Table 1: Summary of Objectives and Deliverables in Work Plan ................................................................ 2
Table 2: FHWA Noise Abatement Criteria ( NAC) in dB ( A) ( hourly A- weighted sound level) ................ 5
Table 3: Road Noise Level Regulations in Europe in dB ( A)...................................................................... 6
Table 4: Experimental Design of the Selected Test Sections ..................................................................... 40
Table 5: Caltrans Environmental Noise Monitoring Site ( ES) Sections*.................................................... 49
Table 6: Climatic Information for ES Test Sections*.................................................................................. 50
Table 7: Traffic Volume and Truck Traffic for ES Test Sections* ............................................................. 50
Table 8: Data Collection in the Field During Traffic Closures................................................................... 52
Table 9: Data Collection at Highway Speed............................................................................................... 52
Table 10: Laboratory Measurements and Tests on Cores Collected in the Field ....................................... 53
Table 11: Descriptive Statistics for the Independent Variables .................................................................. 70
Table 12: Regression Analysis of Permeability .......................................................................................... 82
Table 13: Comparison of Air- Void Content by CT Scan and CoreLok Methods....................................... 90
Table 14: Regression Analysis of Clogging ............................................................................................... 93
Table 15: Weighted Least Squares Regression Analysis of Clogging........................................................ 93
Table 16: Regression Analysis of Clogging for the Center of the Lane ..................................................... 95
Table 17: Weighted Least Squares Regression Analysis of Clogging for the Center of the Lane ............. 95
Table 18: Regression Analysis of Clogging in Wheelpath ......................................................................... 96
Table 19: Weighted Least Squares Regression Analysis of Clogging in Wheelpath ................................. 97
Table 20: Regression Analysis of Microtexture ....................................................................................... 112
Table 21: Wheelpath BPN Difference ...................................................................................................... 116
Table 22: Regression Analysis of Macrotexture....................................................................................... 120
Table 23: Regression Analysis of Difference in Macrotexture Between Two Years ............................... 123
Table 24: Regression Analysis of IRI Values........................................................................................... 133
Table 25: Regression Analysis of Difference in IRI Values Between Two Years ................................... 137
Table 26: Sections Showing Rutting in the Second Year ......................................................................... 141
Table 27: Regression Analysis of Presence of Transverse Cracking........................................................ 145
Table 28: Regression Analysis of Presence of Raveling .......................................................................... 149
Table 29: Number of Previously and Currently Cracked Sections ........................................................... 151
Table 30: Single- Variable Cox Regression Model for Wheelpath Crack Initiation ................................. 153
Table 31: Regression Analysis of Single- Variable Models for Sound Intensity Levels........................... 164
Table 32: Comparison of Shear Modulus and A- Weighted Sound Intensity Levels for
Selected Cores ............................................................................................................................... .. 173
xxiv UCPRC- RR- 2007- 03
Table 33: Regression Analysis of Change in Sound Intensity Levels ...................................................... 177
Table 34: Correlation Matrix of Significant Variables ............................................................................. 179
Table 35: Pattern Matrix Extracted by Principal Components Analysis ................................................. 181
Table 36: Structure Matrix Extracted by Principal Components Analysis .............................................. 181
Table 37: Pattern Matrix Extracted by Principal Axis Factoring............................................................. 182
Table 38: Structure Matrix Extracted by Principal Axis Factoring ......................................................... 182
Table 39: Pavement Characteristics Affecting Noise Levels at Different Frequencies ............................ 188
Table 40: Absorption Values and Noise Reduction of LA 138 Open- Graded and BWC Sections .......... 259
Table 41: Predicted Lifetime of Different Mix Types for BPN Values.................................................... 277
Table 42: Predicted Lifetime of Different Asphalt Mix Types with Respect to Performance
Variables...................................................................................................................... .................... 283
Table 43: Preliminary Recommendation for Open- Graded Mix and Thickness Design to Achieve
Performance Goals ........................................................................................................................... 291
UCPRC- RR- 2007- 03 xxv
LIST OF FIGURES
Figure 1: Addition of two sound sources. ..................................................................................................... 4
Figure 2: Mastic distribution of the open- graded mixes through the thickness ( 31). ................................. 13
Figure 3: Positive texture. ........................................................................................................................... 15
Figure 4: Negative texture. ......................................................................................................................... 15
Figure 5: IRI roughness scale ( WAPA [ 54] from Sayers, 1986 [ 55])......................................................... 19
Figure 6: Noise- generation mechanisms ( 14). ............................................................................................ 26
Figure 7: Pavement texture and roughness ( unevenness). .......................................................................... 28
Figure 8: Noise reflection on reflective surface.......................................................................................... 31
Figure 9: Noise reflection on porous surface. ............................................................................................. 32
Figure 10: Map of the test sections. ............................................................................................................ 35
Figure 11: Dense- graded asphalt concrete ( DGAC). .................................................................................. 45
Figure 12: Open- graded asphalt concrete ( OGAC). ................................................................................... 45
Figure 13: Rubberized open- graded asphalt concrete ( RAC- O)................................................................. 46
Figure 14: Rubberized open- graded asphalt concrete F- mix ( RAC- O F- mix). .......................................... 46
Figure 15: Rubberized gap- graded asphalt concrete ( RAC- G)................................................................... 47
Figure 16: Typical aggregate gradations for different mix types from ignition oven, with 12.5- mm
NMAS. ............................................................................................................................... ............... 47
Figure 17: Typical aggregate gradations for different mix types from ignition oven, with
19- mm NMAS........................................................................................................................... ........ 48
Figure 18: Typical field sampling layout. ( Note: Core locations in the second year are within 1 m
upstream of the first- year locations and core locations 1 and 2, 5 and 6, and 9 and 10.)................... 54
Figure 19: Falling- head permeameter. ........................................................................................................ 56
Figure 20: British Pendulum skid- resistance tester..................................................................................... 57
Figure 21: On- board sound intensity ( OBSI) microphone setup. ............................................................... 58
Figure 22: Laser profilometer beam............................................................................................................ 59
Figure 23: MPD calculations ( ASTM E 1845). .......................................................................................... 60
Figure 24: CoreLok seal of specimen. ........................................................................................................ 60
Figure 25: Computed Tomography scanner ( 106). ..................................................................................... 62
Figure 26: Impedance tube system.............................................................................................................. 63
Figure 27: Distribution of average annual rainfall...................................................................................... 71
Figure 28: Distribution of average annual minimum daily temperature of the coldest month (° C). .......... 72
Figure 29: Distribution of average annual maximum daily temperature of the hottest month (° C). .......... 72
Figure 30: Distribution of annual ESALs in the coring lane. ..................................................................... 73
xxvi UCPRC- RR- 2007- 03
Figure 31: Distribution of annual freeze- thaw cycles. ................................................................................ 73
Figure 32: Pairwise comparison of surface and mix properties.................................................................. 74
Figure 33: Pairwise comparison of climate variables. ................................................................................ 75
Figure 34: Pairwise comparison of cold temperature variables and temperature differences..................... 75
Figure 35: Pairwise comparison of traffic data. .......................................................................................... 76
Figure 36: Box plot of permeability values for different mix types............................................................ 78
Figure 37: Box plot of permeability values for different mix types at different ages ( Note: Age
category 0 is less than one year old, category 1 is one to four years old, and category 2 is greater
than four years old)........................................................................................................................... . 79
Figure 38: Comparison of permeability values for different mix types at different ages for first
and second years.......................................................................................................................... ...... 80
Figure 39: Comparison of permeability differences between first- year and second- year measurements
for different mix types at different ages ( positive value indicates a reduction in permeability). ....... 81
Figure 40: Permeability ( cm/ sec) versus air- void content for different mix types. .................................... 83
Figure 41: Permeability variation for open- graded mixes with different NMAS values............................ 85
Figure 42: Box plot of air- void content for different mix types ( dots show the mean values). .................. 86
Figure 43: Box plot of air- void content for different mix types at different ages....................................... 87
Figure 44: Box plot of fineness modulus for different mix types at different ages. ................................... 87
Figure 45: Comparison of air- void content for different mix types at different ages for first and second
years. ............................................................................................................................... .................. 88
Figure 46: Difference in air- void content between first- year and second- year measurements for
different mix types at different ages ( positive values indicate a reduction in air- void content)......... 89
Figure 47: Permeability difference between the center and the right wheelpath for different mix types
( positive value indicates greater permeability in the center of the lane than in the wheelpath)......... 91
Figure 48: CT scan image of an open- graded mix ( top view). ................................................................... 98
Figure 49: Air- void distribution of open- graded mixes and EU gap- graded mix through the thickness
of the surface layer. ............................................................................................................................ 99
Figure 50: Air- void distribution of dense- graded mixes and BWC mix through the thickness of the
surface layer. ............................................................................................................................... .... 100
Figure 51: Air- void content trend for open- graded mixes and EU gap- graded mix through the thickness
of the surface layer. .......................................................................................................................... 101
Figure 52: Box plot of BPNs for different mix types including all ages. ................................................. 107
Figure 53: Box plot of BPNs for different mix types at different ages..................................................... 108
Figure 54: Comparison of BPNs for different mix types at different ages for first and second years. ..... 109
UCPRC- RR- 2007- 03 xxvii
Figure 55: Difference in BPNs between first- year and second- year measurements for different mix
types at different ages ( positive values indicate reduction in friction)............................................. 110
Figure 56: Scatter plot of BPN versus AADT. ......................................................................................... 113
Figure 57: Box plot of BPNs at the center and the right wheelpath.......................................................... 115
Figure 58: Box plot of MPD values for different mix types with F- mixes separated............................... 117
Figure 59: Box plot of MPD values for different mix types at different ages. ......................................... 118
Figure 60: Comparison of MPD values for different mix types at different ages for first and second
years. ............................................................................................................................... ................ 119
Figure 61: Difference in MPD values between first- year and second- year measurements for different
mix types at different ages ( positive values indicate increase in MPD values)................................ 119
Figure 62: MPD values for different NMAS values and for open- graded and dense- and gap- graded
mixes. ............................................................................................................................... ............... 121
Figure 63: Comparison of F60 values for different mix types. .................................................................. 124
Figure 64: Comparison of Sp values for different mix types. ................................................................... 124
Figure 65: Comparison of F60 values for different mix types, F- mixes separated. ................................... 125
Figure 66: Comparison of F60 values for different mix types at different ages. ...................................... 125
Figure 67: Variation in IRI values for different mix types. ...................................................................... 130
Figure 68: Variation in IRI values for different mix types at different ages............................................. 130
Figure 69: Comparison of IRI values for different mix types at different ages for first and
second years. ............................................................................................................................... .... 131
Figure 70: Difference in IRI values for different mix types at different ages ( positive values
indicate an increase in IRI)............................................................................................................... 132
Figure 71: Number of sections with and without bleeding categorized by mix type. ( Note: Year 1
refers to the first year of measurement and Year 2 to the second year of measurement.)................ 140
Figure 72: Sections with and without rutting............................................................................................ 142
Figure 73: Number of sections with and without transverse cracking for different mix types................. 143
Figure 74: Number of sections with and without transverse cracking for rubberized and nonrubberized
mixes. ( Note: 1st Year refers to the first year of measurement and 2nd Year to the second year of
measurement.) ............................................................................................................................... .. 144
Figure 75: Number of sections with and without raveling for different mix types. ( Note: Year 1
refers to the first year of measurement and Year 2 to the second year of measurement.)................ 147
Figure 76: Number of sections with and without raveling for rubberized and nonrubberized mixes.
( Note: Year 1 refers to the first year of measurement and Year 2 to the second year of
measurement.) ............................................................................................................................... .. 148
xxviii UCPRC- RR- 2007- 03
Figure 77: A- weighted sound intensity levels for different mix types...................................................... 158
Figure 78: A- weighted sound intensity levels for different mix types with F- mixes separated. .............. 159
Figure 79: A- weighted sound intensity levels for different mix types at different ages. .......................... 160
Figure 80: Cumulative distribution function of noise reduction of OGAC, RAC- O, and RAC- G mixes
across an eight- year range of ages ( positive value indicates a reduction in noise). ......................... 161
Figure 81: Comparison of A- weighted sound intensity levels for different mix types at different
ages for first and second years. ........................................................................................................ 162
Figure 82: Difference in A- weighted sound intensity levels between first and second years for
different mix types at different ages ( positive value indicates an increase in noise). ...................... 163
Figure 83: Scatter plot with a best- fit line of A- weighted sound intensity levels versus NMAS ( mm). .. 165
Figure 84: Scatter plot of A- weighted sound intensity levels versus NMAS ( mm) without QP- 48......... 166
Figure 85: Scatter plot of sound intensity levels versus MPD for different mix types. ............................ 167
Figure 86: Scatter plot of A- weighted sound intensity levels versus air- void content for different
mix types. ............................................................................................................................... ......... 168
Figure 87: Scatter plot of A- weighted sound intensity levels versus log ( permeability) for different
mix types. ............................................................................................................................... ......... 168
Figure 88: Scatter plot of A- weighted sound intensity levels versus fineness modulus for different
mix types. ............................................................................................................................... ......... 169
Figure 89: Scatter plot of A- weighted sound intensity levels versus surface layer thickness for
different mix types. .......................................................................................................................... 170
Figure 90: Scatter plot of air- void content versus surface thickness for different mix types.................... 171
Figure 91: Scatter plot of sound intensity levels for rubberized and nonrubberized mixes...................... 172
Figure 92: Relationship between A- weighted sound intensity levels and surface temperatures ( º C)....... 174
Figure 93: Scatter plot of A- weighted sound intensity levels versus log ( permeability) for different
mix types and for different MPD categories. ................................................................................... 176
Figure 94: Scree plot for principal components analysis. ......................................................................... 180
Figure 95: Example of one- third octave band spectrum of OBSI............................................................. 186
Figure 96: 500- Hz band sound intensity levels versus MPD.................................................................... 192
Figure 97: 500- Hz band sound intensity levels versus air- void content. .................................................. 192
Figure 98: Sound intensity levels at 630- Hz band versus age. ................................................................. 193
Figure 99: Sound intensity levels for 800 Hz versus air- void content for different mix types. ................ 194
Figure 100: Sound intensity levels for 800 Hz versus MPD values for different mix types..................... 195
Figure 101: Air- void content versus fineness modulus for different mix types........................................ 197
Figure 102: Sound intensity levels for 1,000- Hz band versus air- void content for different mix types... 198
UCPRC- RR- 2007- 03 xxix
Figure 103: Sound intensity levels for 1,000- Hz band versus MPD values for different mix types. ....... 198
Figure 104: 1,250- Hz band sound intensity levels versus MPD for different mix types .......................... 201
Figure 105: 4,000- Hz band sound intensity levels versus air- void content for different mix types. ........ 205
Figure 106. 4,000- Hz band sound intensity levels versus MPD for different mix types. ......................... 205
Figure 107: Relationship between sound intensity at 500 Hz and surface temperatures ( º C). ................. 207
Figure 108: Relationship between sound intensity at 630 Hz and surface temperatures ( º C). ................. 208
Figure 109: Relationship between sound intensity at 800 Hz and surface temperatures ( º C). ................. 208
Figure 110: Relationship between sound intensity at 1,000 Hz and surface temperatures ( º C). .............. 209
Figure 111: Relationship between sound intensity at 1,250 Hz and surface temperatures ( º C). .............. 209
Figure 112: Relationship between sound intensity at 1,600 Hz and surface temperatures ( º C). .............. 210
Figure 113: Relationship between sound intensity at 2,000 Hz and surface temperatures ( º C). .............. 210
Figure 114: Relationship between sound intensity at 2,500 Hz and surface temperatures ( º C). .............. 211
Figure 115: Relationship between sound intensity at 3,150 Hz and surface temperatures ( º C). .............. 211
Figure 116: Relationship between sound intensity at 4,000 Hz and surface temperatures ( º C). .............. 212
Figure 117: Relationship between sound intensity at 5,000 Hz and surface temperatures ( º C). .............. 212
Figure 118: Example of one- third octave band sound intensity levels for different mix types at
different ages. ............................................................................................................................... ... 213
Figure 119: Box plots of absorption values for different mix types. ........................................................ 216
Figure 120: Comparison of wheelpath absorption values for different mix types at different ages. ........ 217
Figure 121: Correlation of absorption values with air- void content for all mixes pooled together. ......... 218
Figure 122: Correlation of absorption values with surface layer thickness for different mix types. ........ 219
Figure 123: A- weighted sound intensity levels versus absorption values for different mix types. .......... 220
Figure 124: Correlation of sound intensity levels with wheelpath absorption for dense- and
gap- graded mixes. ............................................................................................................................ 221
Figure 125: Correlation of sound intensity levels with wheelpath absorption for open- graded mixes..... 221
Figure 126: Sound intensity levels for 500- Hz band versus absorption values. ....................................... 223
Figure 127: Sound intensity levels for 630- Hz band versus absorption values. ....................................... 224
Figure 128: Sound intensity levels for 800- Hz band versus absorption values. ....................................... 226
Figure 129: Sound intensity levels for 1,000- Hz band versus absorption values. .................................... 228
Figure 130: Sound intensity levels for 1,000- Hz band versus absorption values for different
mix types and different macrotexture values.................................................................................... 229
Figure 131: Sound intensity levels for 1,250- Hz band versus absorption values. .................................... 230
Figure 132: Sound intensity levels for 1,600- Hz band versus absorption values. .................................... 232
Figure 133: Layout of Fresno 33 sections................................................................................................. 237
xxx UCPRC- RR- 2007- 03
Figure 134: First- year and second- year air- void content for Fresno 33 sections...................................... 238
Figure 135: First- year and second- year permeability values for Fresno 33 sections. ( Note: The
scale for permeability was selected for comparison of permeability values across different ES
sections)...................................................................................................................... ..................... 239
Figure 136: First- year and second- year MPD values for Fresno 33 section............................................. 240
Figure 137: First- year and second- year sound intensity levels for Fresno 33 sections............................. 241
Figure 138: Bleeding for 45- mm Type G- MB mix................................................................................... 242
Figure 139: First- year and second- year air- void content for Sacramento 5 sections................................ 244
Figure 140: First- year and second- year permeability values for Sacramento 5 sections.......................... 244
Figure 141: First- year and second- year air- void content for San Mateo 280 section. .............................. 245
Figure 142: First- year and second- year permeability values for San Mateo 280 section. ........................ 245
Figure 143: First- year and second- year IRI values for Sacramento 5 sections......................................... 246
Figure 144: First- year and second- year IRI values for San Mateo 280 section........................................ 247
Figure 145: First- year and second- year MPD values for Sacramento 5 sections. .................................... 248
Figure 146: First- year and second- year MPD values for San Mateo 280 section..................................... 248
Figure 147: First- year and second- year sound intensity levels for Sacramento 5 sections....................... 249
Figure 148: First- year and second- year sound intensity levels for San Mateo 280 section...................... 250
Figure 149: Layout of the test sections. .................................................................................................... 252
Figure 150: Comparison of first- year and second- year air- void content for LA 138 sections. ................ 253
Figure 151: Comparison of first- year and second- year permeability values for LA 138 sections............ 253
Figure 152: First- year and second- year IRI values for LA 138 sections. ................................................. 254
Figure 153: First- year and second- year MPD values for LA 138 sections............................................... 255
Figure 154: Comparison of sound intensity levels for LA 138 sections................................................... 256
Figure 155: Noise reduction from pass- by measurements by Volpe National Transportation Systems
Center for LA 138 mixes.................................................................................................................. 257
Figure 156: Comparison of gradation of LA 19 section with RAC- G gradation...................................... 262
Figure 157: First- year and second- year air- void content for LA 19 section............................................. 262
Figure 158: Comparison of first- year and second- year permeability values for LA 19 section. .............. 263
Figure 159: First- year and second- year MPD values for LA 19 section. ................................................. 264
Figure 160: First- year and second- year sound intensity levels for LA 19 section.................................... 265
Figure 161: First- year and second- year air- void content for Yolo 80 section. ......................................... 266
Figure 162: First- year and second- year permeability values for Yolo 80 section. .................................. 267
Figure 163: First- year and second- year MPD values for Yolo 80 section................................................ 268
Figure 164: First- year and second- year sound intensity levels for Yolo 80 section. ................................ 269
UCPRC- RR- 2007- 03 xxxi
LIST OF ABBREVIATIONS
AADT average annual daily traffic
AADTCL average annual daily traffic in the coring lane
AADTT average annual daily truck traffic
AC asphalt concrete
ADT average daily traffic
ADTT average daily truck traffic
AV air- void content
BWC bonded wearing course
BPN British Pendulum Number
BS British Standard
Caltrans California Department of Transportation
Caltrans PCS Caltrans Pavement Condition Survey
CDIM Climate Database for Integrated Model
CTM Circular Texture Meter
CT scan Computed Tomography scan
DEA Division of Environmental Analysis
DFT Dynamic Friction Tester
DGAC dense- graded asphalt concrete
DMI Distance Measuring Instrument
EB eastbound ( used for pavement traffic direction)
EMPA Swiss Federal Laboratories for Materials Testing and Research
ESAL equivalent single- axle load
ES sections environmental noise monitoring site sections
ETD estimated texture depth
EU- GG European gap- graded asphalt concrete
FHWA Federal Highway Administration
F- mixes open- graded gradation mixes with 19 mm NMAS ( originally developed by
the Oregon DOT)
Fre 33 Fresno 33
Gbulk bulk specific gravity
Gmm maximum specific gravity
IFI International Friction Index
IRI International Roughness Index
IQR interquartile range
LA 138 Los Angeles 138
LA 19 Los Angeles 19
LTTP Long- Term Pavement Performance
MPD mean profile depth
MTD mean texture depth
NAC noise abatement criteria
NB northbound ( used for pavement traffic direction)
NCAT National Center for Asphalt Technology
NMAS nominal maximum aggregate size
OBSI on- board sound intensity
OGAC open- graded asphalt concrete
OLS regression models ordinary least squares regression models
PCC portland cement concrete
PCR principal components regression
PCS Pavement Condition Survey
PMS Pavement Management System
xxxii UCPRC- RR- 2007- 03
QP sections quiet pavement experimental design sections
RAC- G rubberized gap- graded asphalt concrete
RAC- O rubberized open- graded asphalt concrete
RMS root mean square
RUMAC- GG rubber- modified asphalt concrete ( dry process, a local- government
specification)
Sac 5 Sacramento 5
SB southbound ( used for pavement traffic direction)
SM 280 San Mateo 280
SMA stone mastic asphalt
Type D- MB dense- graded mix with binder meeting MB specification
Type G- MB gap- graded mix with binder meeting MB specification
UCPRC University of California Pavement Research Center
VMA voids in the mineral aggregate
WIM weigh in motion
UCPRC- RR- 2007- 03 1
1 INTRODUCTION
1.1 Background
With traffic noise becoming a growing concern, the public is expecting quieter pavements to be
constructed to abate traffic noise levels. Quieter pavements may offer new options for minimizing the
impact of traffic noise levels on neighborhoods adjacent to highways.
Quieter pavement is a new concept intended to reduce the impact that tire/ pavement noise has on
the highway environment. The concept of using quieter pavements to reduce noise has received
increasing attention in California and nationwide over the past several years. With the short- term benefits
of quieter pavements somewhat documented, most of the new attention has focused on developing a
better understanding of the long- term acoustic benefits of quieter pavements. In response to public
expectations, the California Department of Transportation ( Caltrans) has initiated several studies to
evaluate the acoustic properties of pavements and the role of pavement surface characteristics on
tire/ pavement noise levels. The research presented in this report is part of one of these studies and is part
of the Caltrans Quieter Pavements Research ( QPR) Work Plan ( 1)
The Caltrans QPR plan is a systematic research proposal intended to examine the impact of
quieter pavements on traffic noise levels and to establish which pavement characteristics have the greatest
impact on tire/ pavement noise. The goal of the QPR study is to evaluate the acoustic properties and
performance characteristics of both flexible and rigid pavements and of bridge decks used by the state.
Additionally, the QPR study is intended to identify surface treatments, materials, and construction
methods that will result in quieter pavements that are also safe, durable, and cost effective. The
information gathered in the study will be used to develop quieter- pavement design features and
specifications for noise abatement throughout the state. ( 1)
For the flexible pavement part of the QPR study, Caltrans identified a need for research in the
areas of acoustics, friction, and pavement performance of asphalt pavement surfaces for the state highway
network. In November 2004, the Caltrans Pavement Standards Team ( PST) approved a new research goal
for the Partnered Pavement Research Center ( PPRC) Strategic Plan; it was numbered Element 4.16 and
titled “ Investigation of Noise, Durability, Permeability, and Friction Performance Trends for Asphaltic
Pavement Surface Types.”
The goals of this research as described in the Work Plan ( 2) approved by the PST are to:
1. Develop a database for lifetime performance trends to identify best practices. Trends will
be determined for California open- graded ( including mixes with and without rubberized asphalt
binder), rubberized gap- graded asphalt concrete ( RAC- G). and dense- graded asphalt concrete
2 UCPRC- RR- 2007- 03
( DGAC) mixes with regard to sound intensity, durability ( raveling, rutting, and cracking),
friction, and permeability. Performance trends will be analyzed as a function of gradation, binder
type, traffic [ speed, average daily traffic ( ADT), average daily truck traffic ( ADTT), and
equivalent single- axle load ( ESAL)], climate ( rainfall, temperature, and freezing), and roughness
[ International Roughness Index ( IRI)].
2. Gather and summarize information on laboratory tests that are correlated with these
performance measures ( sound intensity, durability, friction, and permeability), gather information
on mix design methods, and identify best practices that can potentially be brought to California.
3. Survey practice and research in other states and in Europe on the lifetime performance of
their open- graded mix types with respect to sound intensity, durability, friction, and permeability.
Gather and summarize performance data and identify promising mixes that can be brought to
California.
4. Determine whether a relationship exists between a laboratory noise absorption test, the
impedance tube, and field sound intensity measurements using field cores.
1.2 Quieter Pavement Research Objectives Addressed in This Report
The objectives in the Work Plan for PPRC Strategic Plan Element 4.16 ( PPRC SPE 4.16) are
shown in Table 1 ( 2).
Table 1: Summary of Objectives and Deliverables in Work Plan
Work
Plan
Section Objective Deliverables
1.2.1 1. Literature survey
• Literature survey of U. S. practice, performed by PPRC
• Literature survey of European practice performed by Swiss
Federal Laboratories for Materials Testing and Research ( EMPA)
1.2.2 2. PPRC test
capability
PPRC operational capability to measure field sound intensity, lab
noise impedance, and field surface friction
1.2.3 3. Database structure
• Database structure
• Populated database at completion of data collection objective
( Objective 4)
4. Data collection
4a. Field sections in
California
• Properties of as- built surfaces ( sound intensity, permeability,
friction, and distresses) over time, with trends in data summarized
annually
• Continued data collection as authorized by Caltrans.
1.2.4
4b. Field and lab data
from outside
California
Database of measurements of outside mixes and report summarizing
data collected and trends
1.2.5 5. Performance- trend
statistical analysis
Report documenting statistical analysis results from data collection
objective ( Objective 4), with summary of performance trends and
modeling results
1.2.6 6. Two- year summary
report
Report summarizing all of the work completed; this document
constitutes the report
UCPRC- RR- 2007- 03 3
This report completes Objective 6. The results presented in this report complete Objectives 4a
and 5.
A summary of the results of Objective 1, the literature survey completed earlier, have been
included in this report, and it includes information regarding performance properties. A separate technical
memorandum summarizes a literature survey completed in 2006 regarding open- graded mix design
practices and specifications. The data and analysis presented for Objectives 4a and 5 are for the first two
years of data collection on the quiet pavement sections ( QP sections) in the experimental factorial in the
PPRC SPE 4.16 Work Plan, and for the Division of Environmental Analysis ( DEA) sections and other
special test sections, which are collectively referred to as ES sections. Among the DEA sections are
sections that DEA has monitored for noise properties for a number of years, as well as other test sections
placed by Caltrans over the past five years that are outside the main factorial for this project. The data
collected on all sections over the first two years of the study has been entered into the project database
that will be submitted separately. Data was not available to complete Objective 4b, however. Caltrans, its
consultants, and UCPRC have been working with other states and other consultants to develop common
noise measurement protocols so that data can be compared in the future. This report constitutes
completion of Objective 6.
1.3 Traffic Noise
Noise is defined as unwanted or unpleasant sound. Like all other sounds, noise is produced by
vibrating objects and transmitted by pressure waves in a compressible medium such as air. As pressure
waves travel through a medium, they produce sound. Sound waves are characterized by three parameters.
• Wavelength: The distance of the crest of one wave to the crest of the one following.
• Frequency: The number of waves that pass a particular point each second
• Amplitude: The measure of the energy present in a sound wave; the greater the amplitude of the
sound energy, the louder the sound
Sound pressure or sound intensity levels are used to quantify the loudness of an ambient sound.
The frequencies of sounds audible to humans range from 20 to 20,000 Hz, and sound pressures range
from 20 micropascals ( μPa), the threshold of hearing, to 120 pascals ( Pa), the threshold of pain ( 3). Since
it is hard to work with such a broad range of sound pressure, the linear sound pressure, p, is converted to a
logarithmic sound pressure level, SPL, which compresses the scale of numbers into a manageable range.
The conversion from linear sound pressure to logarithmic scale is given in Equation ( 1):
4 UCPRC- RR- 2007- 03
SPL = 10 × log ( p/ pref) ( 1)
where pref is an international standardized reference sound pressure of 2 × 10– 5 Pa. The unit of
SPL is called the decibel ( dB).
Noise levels can also be expressed in terms of sound intensity, which is a measure of energy flow
through a unit area. Sound intensity is converted to a logarithmic sound intensity level, LI, according to
Equation ( 2):
LI = 10 × log ( I/ Iref) ( 2)
where Iref is 10– 12 W/ m2.
Iref is chosen to obtain the same reading in decibels regardless of whether SPL or LI is used to
define the sound wave, and irrespective of whether pressure or intensity in an acoustic free field is
measured. The unit of LI is also the decibel ( dB).
The human ear is not equally sensitive to all sound frequencies; the ear can hear high- frequency
noise better than low- frequency noise that has the same sound pressure ( dB). Therefore, noise
measurement readings can be adjusted to incorporate this difference in sensitivity. The adjustment of
sound measurements according to human sensitivity is called A- weighting, and adjusted noise levels are
written as dB ( A).
Since the decibel scale is nonlinear, resultant noise levels from two different sound sources that
emit two incoherent sounds with the same sound pressure would increase the noise level by 3 dB ( 70 dB
+ 70 dB = 73 dB, not 140 dB). The formula used to add together multiple sources of sound is given in
Equation ( 3). The effect of two incoherent sound levels is shown in Figure 1.
Σ
=
=
i soundsources
Total
dB A i dB( A) log 10 10
10
( )
10 ( 3)
Figure 1: Addition of two sound sources.
73 dB
together
70 dB
70 dB
UCPRC- RR- 2007- 03 5
Noise levels are mainly affected by the distance from the source. As sound waves travel through a
medium, they spread out over a spherical or circular surface; hence, their energy is distributed over a
greater surface area. Since the energy of waves is conserved and the area through which the waves travel
increases, sound intensity decreases. Intensity variation is proportional to the square of the distance from
the source. If the distance from the source is doubled, the sound intensity level is decreased by a factor of
four.
The measurement of noise is adjusted to reflect the sensitivity of human hearing because noise
disturbance can affect the quality of human lives. In recent decades, noise pollution has become a major
concern in the United States and the world. Noise pollution can impair hearing, cause sleep disturbances,
have cardiovascular effects, interfere with social behavior ( aggressiveness, protest, and helpfulness) and
verbal communication, and cause annoyance ( 4). The economic consequences of these health
impairments include loss of property value in areas subject to noise, reduced work performance by those
affected by noise ( 5) and medical costs for improving the health of those affected by noise ( 6). Among all
environmental noises ( construction, rail, road, and airplane noise), road noise has been identified as the
most annoying ( 7). Road noise mostly affects people in residences and businesses next to highways and
people in road vehicles.
Because of increases in motorization and the number of highways, the problem of traffic noise
has begun receiving a lot of attention. The adverse effects of traffic noise on health and the economy have
forced communities to seek solutions to improve quality of life by reducing this noise. Most industrialized
countries have introduced noise emission regulations. The Federal Highway Administration ( FHWA)
specifies noise levels for different types of land zoning where noise abatement should be considered.
FHWA Noise Abatement Criteria ( 8) are given in Table 2.
Table 2: FHWA Noise Abatement Criteria ( NAC) in dB ( A) ( hourly A- weighted sound level)
Activity
Category
NAC, Leq( h)* Description of Activity Category
A 57 ( exterior) Lands on which serenity and quiet are of extraordinary
significance and serve an important public need and where
the preservation of those qualities is essential if the area is to
continue to serve its intended purpose
B 67 ( exterior) Picnic areas, recreation areas, playgrounds, active- sports
areas, parks, residences, motels, hotels, schools, churches,
libraries, and hospitals
C 72 ( exterior) Developed lands, properties, or activities not included in
Category A or B
D - Undeveloped lands
E 52 ( interior) Residences, motels, hotels, public meeting rooms, schools,
churches libraries, hospitals, and auditoriums
* Leq( h) is the sound pressure averaged over one hour.
6 UCPRC- RR- 2007- 03
European countries have also imposed regulations regarding noise levels, given in Table 3 ( 9).
Additionally, the European Union specifies noise emission limits for new tires for passenger cars and for
heavy and light trucks ( 10). The EU has created a project called Coordination of European Research for
Advanced Transport Noise Mitigation ( CALM) that supports research and development of new
technologies to reduce all transport- related noise ( 11).
Table 3: Road Noise Level Regulations in Europe in dB ( A)
Country Planning Value
Leq * Maximum Limit Leq * Remarks
Austria 55 - -
Switzerland Day 50
Night 40
Day 55
Night 45 -
France Day 60
Night 55 65
Average for day from 8
a. m. to 8 p. m.
Average for night from
10 p. m. to 6 a. m.
Denmark 55 - -
UK Day 55
Night 45
Day 72
Night 66
Average for day from 7
a. m. to 11 p. m.
Netherlands Day 55
Night 45
Day 58
Night 48
35 dB ( A) inside
25 dB ( A) inside at night
Sweden 55 - 30 dB ( A) inside
* Leq sound pressures are averaged over 24 hours unless otherwise indicated in Remarks column.
Highway noise arises from automobiles, buses, trucks, and motorcycles in motion. Vehicle noise
has three components: aerodynamic noise, power- unit noise, and tire/ pavement noise. At lower speeds,
the power train and its ancillaries generate the major component of traffic noise, while at higher speeds—
approximately above 50 km/ h for passenger cars and 70 km/ h for heavy vehicles— tire/ pavement
interaction noise dominates the other mechanisms ( 12, 13). Tire/ pavement noise depends on pavement
surface characteristics, vehicle speed, environmental conditions, type of tire, and the dynamics of the
rolling process ( 14). The tire/ pavement noise level increases logarithmically with increasing speed ( 13).
1.4 Desired Properties of Open- Graded Mixes
For open- graded mixes to be accepted as a noise mitigation tool, they should have good
performance and lower life- cycle costs than competing alternatives such as gap- and dense- graded mixes.
Rubberized gap- graded asphalt concrete ( RAC- G) mixes have been proposed as an alternative to open-graded
mixes with respect to noise reduction, while both dense- graded asphalt concrete ( DGAC) and
RAC- G mixes can be alternatives to open- graded mixes with respect to durability and ride quality.
Neither RAC- G nor DGAC can be considered an alternative to open- graded mixes with respect to
permeability. However, RAC- O and open- graded asphalt concrete ( OGAC) mixes are the competing
UCPRC- RR- 2007- 03 7
alternatives for improved permeability. The following section explains the expected behavior of open-graded
mixes in terms of performance factors.
1.4.1 Performance of Open- Graded Mixes
1.4.1.1 Permeability
Permeability is the most important performance variable for open- graded mixes as they are
primarily placed to improve wet- weather surface friction. Open- graded mixes have higher air- void
content, and hence higher permeability, than conventional asphalt mixes, which enables them to remove
standing water. Open- graded mixes can reduce hydroplaning and water spray and splash by draining the
water into the mix and hence enhance wet- weather safety.
Hydroplaning is the loss of directional control when a vehicle is moving fast enough that the tires
lose contact with the surface and ride up on the water film present on the pavement surface. Splash is the
mechanical impact of tires that forces the water out of the tire/ pavement contact area. Splash results in
reduced visibility. Hydroplaning and splash are affected by the water depth on the surface.
Open- graded mixes are susceptible to clogging at least at the surface. When the surface air voids
are clogged with fine materials, air- void content and hence permeability decreases, and the benefits of
open- graded mixes diminish. Therefore, the performance of open- graded mixes is governed mainly by the
length of time that they maintain their permeability.
1.4.1.2 Skid Resistance ( Friction)
Skid resistance is the force required to prevent a vehicle tire from sliding along the pavement
surface. It is important for safety because inadequate skid resistance can result in loss of control and
longer stopping distances, and hence skid- related accidents. Higher friction or skid numbers mean safer
pavements. Skid resistance depends on the pavement surface’s microtexture and macrotexture.
Microtexture refers to small- scale irregularities of the pavement aggregate, and macrotexture refers to
large- scale irregularities of the pavement surface that are affected by aggregate orientation. Macrotexture
can be measured by the sand patch method as well as by laser measurements; microtexture can be
measured by British Pendulum Tester or Dynamic Friction Tester ( DFT) values. A British Pendulum
Number ( BPN) above 45 indicates a satisfactory surface according to a Caltrans ( 15) research document
believed to be from the 1960s. This value is used as a criterion for discussion and comparison of different
test sections in this research study, but it is not a Caltrans standard and should not be construed as an
official standard. The performance of open- graded mixes can be evaluated by the length of time that they
provide satisfactory friction and by comparison of their friction values to those of alternative mixes.
8 UCPRC- RR- 2007- 03
1.4.1.3 Roughness
Roughness refers to surface irregularities with wavelengths greater than 0.5 m. It is associated
with ride quality. Ride quality is an indication of the comfort level of the ride over a pavement surface.
Road users judge a road condition mainly based on its ride quality.
On a rough pavement, the vehicle vertical movements are high. The roughness of the pavement
surface is related to the vibration of the vehicle, tire wear, operating speed, and vehicle operating costs.
Roughness is currently typically measured by a standardized scale called the International Roughness
Index ( IRI). IRI is obtained by performing a quarter- car simulation on a longitudinal profile in the
wheelpath. According to the FHWA, pavements with an IRI value greater than 95 inches per mile and
less than or equal to 170 inches per mile are classified as acceptable, and pavements with an IRI value
less than or equal to 95 inches per mile are classified as good ( 16). The performance of open- graded
mixes can be evaluated by the length of time that they provide at least acceptable ride and by comparison
of their roughness progression to that of alternative mixes.
1.4.1.4 Durability
Durability is the capacity of a pavement to keep its functionality over time. It can be evaluated in
terms of distress development. Surface distresses can be in the form of cracks; deformation such as
rutting, corrugation, bleeding, or shoving; or disintegration such as raveling, stripping, and spalling.
When cracks are present on the pavement surface, water may enter the pavement structure. Since the
base, subbase, and subgrade lose their load- carrying capacities when they are wet, the water entering
through the cracks may lead to more severe pavement failures. Presence of distresses on the pavement
surface may lead to rougher pavements and hence poor ride quality. Additionally, the presence of
bleeding on the pavement surface may reduce friction.
Common distresses of open- graded mixes that have been observed or reported in the literature are
rutting, transverse cracking, reflection cracking, bleeding, raveling, and fatigue cracking, most of which
are also identified in the Caltrans Maintenance Technical Advisory Guide ( 17). The performance of open-graded
mixes can be evaluated in terms of how long the mixes take to develop distresses and how the
distresses progress compared to distresses in alternative mixes.
1.4.1.5 Noise Reduction
Open- graded mixes may provide noise reduction due to their higher air- void content. However,
they may lose their noise- reducing properties with time due to clogging, raveling and cracking. When the
surface air voids are clogged, not only is the permeability reduced but so is noise reduction. Open- graded
mixes should lower the traffic noise at least 3 dB ( A) compared to conventional road surfaces without
UCPRC- RR- 2007- 03 9
jeopardizing pavement safety and durability ( 10). A reduction of 3 dB ( A) has the same effect as reducing
the traffic volume by half. A 3 dB ( A) change is just noticeable to the human ear. The performance of
open- graded mixes can be evaluated in terms of the amount of noise reduction they can provide compared
to alternative mixes and the length of time that they can maintain their noise- reducing properties.
1.5 Contents of This Report
Chapter 1 presents an introduction and background information about asphalt mix types and noise
and summarizes the objectives and scope of this report. Chapter 2 presents a review of the literature
pertinent to permeability, skid resistance ( friction), roughness/ ride quality, pavement distresses, and
highway noise and pavement characteristics affecting noise levels. Chapter 3 describes the selection of
the test sections and equipment and test methods used in the study. Chapter 4 describes the variables used
in the study and presents the descriptive statistics for the independent variables. Chapter 5 evaluates the
permeability and air- void content variable of different mix types and presents the analysis of variables
affecting permeability and clogging. Chapter 6 presents the frictional properties of different mix types and
the variables affecting friction. Chapter 7 describes the roughness of different mix types and the variables
affecting roughness. Chapter 8 evaluates pavement distresses, including bleeding, rutting, raveling,
transverse and reflection cracking, and fatigue cracking. Chapter 9 evaluates the acoustical properties of
pavements, including sound intensity levels and acoustical absorption properties. Chapter 10 compares
the performance of rubber- modified asphalt concrete ( RUMAC- GG), gap- graded mix with modified
binder ( Type G- MB), and dense- graded mix with modified binder ( Type D- MB) asphalt mixes with the
mixes currently used and evaluates the effects of thickness and age on pavement performance for
California Department of Transportation ( DOT) environmental noise monitoring site ( ES) sections. It also
compares pass- by noise measurements using the on- board sound intensity ( OBSI) method. Chapter 11
evaluates the performance models and predicts the lifetime for different types of mixes. Chapter 12
summarizes the findings, makes recommendations for open- graded mix design, and suggests future
research.
10 UCPRC- RR- 2007- 03
UCPRC- RR- 2007- 03 11
2 LITERATURE SURVEY
2.1 Permeability
Permeability is the most important functional performance criterion for open- graded asphalt
concrete mixes, as was discussed in Section 1.4.1. Open- graded pavements let water drain into the surface
mix through the air voids rather than keeping it on the surface of the pavement, which results in reduced
hydroplaning, water splash, and spray, and hence improved safety. Also, glare from the road surface is
reduced and visibility is improved ( 18). Open- graded mixes have been used primarily to improve wet-weather
skid resistance by removing stagnant water from the pavement surface. However, open- graded
mixes can lose their permeability and hence their noise- reducing properties over time due to clogging.
Various factors that affect permeability have been reported in the literature. Air- void content has
been shown to be the most important factor affecting permeability of asphalt pavements ( Kanitpong,
2001; Brown, 2004; Mallick et al., 2003). As air- void content increases, permeability increases. However,
a study by Huang ( 1999) indicated that the interconnectivity of air voids is more important in determining
permeability than the total volume of the air voids.
Aggregate gradation and size also affect permeability. Mixes with coarser gradations were shown
to have higher permeability values than those with finer gradations ( 19; 20; 21; 22). Coarse- graded mixes
have larger voids, and hence greater potential for connected air voids, which results in greater
permeability. The nominal maximum aggregate size ( NMAS) has also been found to affect the field
permeability of asphalt mixes ( 21, 23). Pore size increases as NMAS increases; hence the possibility of
connected air voids increases. Therefore, mixes with larger NMAS values are expected to be more
permeable at a given air- void content than mixes with smaller NMAS values.
Another factor that may affect permeability of DGAC and RAC- G, which are intended to be
impermeable, is lift thickness. Although the Florida DOT ( 22) and Mallick ( 21) concluded that increased
lift thicknesses could lead to better pavement compaction and hence lower permeability, the Wisconsin
DOT ( 24) could not find any relationship between lift thickness and permeability. There have also been
studies looking at the effects of the thickness- to- NMAS ( t/ NMAS) ratio on permeability ( 20, 24). The
Wisconsin DOT ( 24) found that permeability increased for smaller t/ NMAS ratios for a mix with
limestone aggregate; however, no trend was observed for a gravel- aggregate mix. Brown ( 20) concluded
that higher t/ NMAS ratios provide lower air- void content, which may result in lower permeability values.
Because open- graded mixes have higher air- void content, and hence higher permeability than
conventional asphalt mixes, they are susceptible to clogging. Clogging is the blockage of air voids with
fine particles generated by vehicles and deposited from elsewhere by wind and vehicles. When air voids
12 UCPRC- RR- 2007- 03
are clogged with fine materials, air- void content, and hence permeability, decreases, and the benefits of
open- graded mixes diminish.
A minimum permeability value of from 0.01 to 0.4 cm/ sec is specified for open- graded mixes by
European standards for porous asphalt ( 25). The requirement for in- situ permeability of open- graded
mixes in Switzerland is 0.11 cm/ sec ( 15 l/ min) ( 26). No permeability requirements for open- graded mixes
could be found in the literature from the United States.
According to Sandberg et al. ( 10), traffic and rainfall are the most important factors affecting in-service
permeability of open- graded mixes. Fine particles that lodge in the voids of the surface layer can
be suctioned out by the hydraulic action of traffic. This cleaning effect is more pronounced under heavy
rainfall and fast traffic. Due to the suction effect of traffic, wheel tracks were found to remain more
permeable than road shoulders ( 10, 27). However, lower air- void content of open- graded mixes in the
wheelpaths may also be caused by densification of the mix under traffic loading, which has been observed
in Arizona ( 28).
However, according to Bendsten ( 29), the most important factors affecting the in- service
permeability of open- graded mixes are the age of the pavement, maximum aggregate size, air- void
content and distribution, speed of vehicles, and shape of aggregates. Another study, conducted in
Denmark ( 30), evaluated the air- void content of open- graded mixes in horizontal planes taken through the
thickness using Computed Tomography ( CT) scan technology, and no significant difference was found
between the air- void content of the wheelpath and shoulder. The bottom part of these older open- graded
mixes was found to have at least twice the air- void content of the top part, which suggests that fine
particles accumulate only in the top part of the surface layer ( the top 20 to 25 mm) which was typically 50
mm thick. Figure 2 shows the mastic ( composed of asphalt binder, sand, and dirt) distribution of open-graded
asphalt cores that are six years old from another study in Denmark ( 31). The trend lines shown in
the figure indicate that mastic content decreases from top to bottom. The mastic content in the top 20 mm
of the surface layer is higher than that in the lower part. The higher mastic content in the upper part of the
pavement surface was explained by the clogging of air voids by fine particles. However, note that the
mixes shown in the figure are two- layer porous asphalt, where the top layer is 20 to 25 mm thick and has
smaller- size aggregate, and the bottom layer is thicker and has larger- size aggregate.
UCPRC- RR- 2007- 03 13
Figure 2: Mastic distribution of the open- graded mixes through the thickness ( 31).
In the United Kingdom, a study found that open- graded mixes using larger maximum aggregate
size keep their porosity longer compared to mixes with smaller aggregate size ( 32). On the basis of this
result, the United Kingdom specifies 20- mm maximum aggregate size for its open- graded mixes.
Sandberg et al. ( 10) suggested that binder type may affect clogging. There is some evidence that
dirt does not stick to polymer- modified asphalt binder as much as it sticks to unmodified asphalt binder
( 10). This effect may be due to the higher softening point of polymer- modified binder. However, this
effect was identified as needing further investigation. If the softening point at high temperatures affects
the dirt accumulation in the voids, the temperatures experienced by the pavement may also affect
clogging.
The location of the pavement, defined as urban or rural, has also been found to affect the clogging
of open- graded mixes ( 10, 33, 29). Rural roads were found to be more likely to get clogged by mud and
sand carried by agricultural trucks.
All the research on clogging was conducted in Europe. Therefore, the results are limited to
asphalt- mix designs and traffic levels and climatic conditions in Europe, which are different from those in
the United States. Factors affecting clogging of California open- graded mixes, which experience different
traffic and climate conditions than those in Europe, still need to be identified.
Earlier studies have shown the effects of air- void content and gradation on permeability.
However, research is still needed to clarify the effect of thickness on the permeability of different asphalt
mix types and the changes in permeability of different mixes over time. Permeability is controlled not
14 UCPRC- RR- 2007- 03
only by total air- void content, but also by interconnectivity of the air- void system and by distribution of
air voids within the layer.
2.2 Skid Resistance ( Friction)
Skid resistance is the force required to prevent a vehicle tire from sliding along the pavement
surface. It is important in terms of safety because inadequate skid resistance may result in skid- related
accidents. Skid resistance is generally quantified in terms of friction measurements such as friction or skid
numbers. A skid number is actually the coefficient of friction. Skid resistance can be measured by locked-wheel
tests, spin- up tests, and surface- texture measurements.
Skid numbers above 30 are acceptable for low- volume roads, and skid numbers above 35 are
acceptable for heavily traveled roads ( 34). Higher friction or skid numbers result in shorter stopping
distances. Skid resistance is controlled by the microtexture and macrotexture of the surface as well as the
geometrical design of the road.
Microtexture is the deviation of a pavement surface from a true planar surface with a maximum
dimension of 0.5 mm ( 35). It is associated with microscopic properties of the surface and controlled by
the individual aggregate surface properties, such as shape and harshness. Microtexture controls the
adhesion component of friction between the tire and the road surface. Therefore, it is important for
providing a good grip, and hence skid resistance between the tire and the pavement surface, at low speeds
and under dry road conditions; although microtexture contributes to skid resistance at all speeds, it has the
most influence at speeds less than 30 mph.
Macrotexture is the deviation of a pavement surface from a true planar surface with a dimension
between 0.5 and 50 mm ( 35). The visible irregularities of a pavement surface caused by large aggregate
particles control the texture wavelength. Macrotexture facilitates water drainage by providing water
channels on the pavement surface, preventing a film of water from developing between the pavement and
the tire and loss of contact between the pavement and the tire ( hydroplaning). Macrotexture also
contributes to friction of the pavement surface and controls skid resistance at higher speeds.
Macrotexture can be reported as mean profile depth ( MPD) or mean texture depth ( MTD). MPD
is a two- dimensional estimate of three- dimensional MTD. Laser measurements give the MPD, and sand-patch
measurements give the MTD. In addition to the MPD, laser measurements provide the root mean
square ( RMS), which is a statistical value showing how much the measured profile ( actual data) deviates
from a modeled profile of the data.
Surface texture can be positive or negative depending on t