STATE OF CALIFORNIA
California
Department
of
Transportation
FLEXIBLE
PAVEMENT
REHABILITATION
MANUAL
Revised: June 1, 2001
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
ii
DISCLAIMER
This manual is intended for the use of Caltrans personnel. Engineers and agencies
outside of Caltrans may use this manual at their own discretion. Caltrans is not
responsible for any work performed by non- Caltrans personnel using this manual.
ACKNOWLEDGMENT
The information contained in this manual is a result of efforts of many individuals
in the Structural Section Design and Rehabilitation Branch ( SSD& R). Gary Mann,
P. E. and Paul Mason, P. E., with a combined experience of 40 years of pavement
structural section design and rehabilitation, are the principal authors of this
publication. The contributions of engineers and technicians who have worked and
are presently working in SSD& R are also acknowledged.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
iii
TABLE OF CONTENTS
Chapter - Section
1 GENERAL
1 – 10 Background
1 – 20 Foreword
2 PERFORMING A DEFLECTION STUDY
2 – 10 Equipment
2 – 20 Establishing Test Sections
2 – 30 Pavement Background Information
2 – 40 Collecting Field Data
2 – 50 Measuring Deflections
2 – 60 Converting to Equivalent Deflectometer Values
2 – 70 Mean and 80th Percentile Deflections
2 – 80 Preparing a Deflection Map
3 INTRODUCTION TO AC PAVEMENT REHABILITATION DESIGN
3 – 10 Design Governed by Structural Adequacy
3 – 20 Design Governed by Reflective Cracking
3 – 30 Design Governed by Ride Quality
3 – 40 Choosing the Design Recommendation
4 FLEXIBLE PAVEMENT REHABILITATION DESIGN GUIDE
4 – 10 Basic Overlay Using DGAC
4 – 20 Rubberized Asphalt Concrete ( Type G)
4 – 30 Stress Absorbing Membrane Interlayers
4 – 40 Cold Recycled Asphalt Concrete Pavement
4 – 50 Hot Recycled Asphalt Concrete Pavement
4 – 60 Remove and Replace
4 – 70 Asphalt Concrete Overlay Placed on a Cushion Course
4 – 80 Cushion Course Design With Drainage Layer
4 – 90 Asphalt Concrete Overlay With Drainage Layer
5 APPENDIX
5 – 10 Guidelines for Involving Moisture and Temperature in Flexible Pavement Rehabilitation
5 – 20 Identifying and Recording Distress
5 – 30 Abbreviations
5 – 40 Definitions
5 – 50 Bibliography
6 TABLES
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
1- 1
CHAPTER 1
GENERAL
This manual delineates the basic design
strategies of the 1979 “ Asphalt Concrete
Overlay Design Manual” plus the many
changes in procedures, and incorporates
the use of new strategies and materials
presently being used by Caltrans. It is
intended as a tool to provide guidance to
those who are recommending asphalt
concrete pavement rehabilitation for
planning, design and maintenance of the
state’s highways. Caution and
engineering judgment must be exercised
throughout the investigation and design
process.
1 – 10 Background
Since 1938, deflection measurements
have been utilized for the evaluation of
flexible pavement. In 1951, the
Laboratory at the Division of Highways
initiated a series of comprehensive
deflection research studies in an effort to
establish relationships between
pavement deflections and pavement
performance. The results and
conclusions of the first formal study
were published in 1955 ( 1). An
evaluation of the data, with respect to
pavement deflections versus pavement
conditions, permitted the establishment
of the concept of " tolerable deflection"
criteria for a variety of asphalt concrete
( AC) structural sections*. Tolerable
* The term " tolerable deflection," refers to the
level beyond which repeated deflections of that
magnitude would produce fatigue cracking in the
surface prior to the planned design period of the
pavement.
deflections eventually provided the basis
for the application of pavement
deflection data to overlay design.
However, since tolerable deflection
values were collected for roads with
Traffic Indices ( TI’s) of approximately
9, results of laboratory fatigue tests on
asphalt concrete samples were used to
establish a method to adjust tolerable
deflection levels for other TI values ( 2).
In 1960, California began using
deflection data in conjunction with the
tolerable deflection as the basis for
overlay design. Data accumulated on
the deflection values for various
thicknesses of AC pavement with
cement treated base, or aggregate base,
subjected to various traffic loadings
along with the tolerable deflection
criteria already established, provided the
basis of the Caltrans overlay design
procedure. By 1966, approximately 80
overlay projects; including state
highways, county roads, and city streets;
had been designed by deflection
analysis.
After almost 20 years of research into
determining asphalt concrete pavement
deflections and relating these deflections
to pavement performance,( 3) the data
collection and design procedures were
formally adopted in 1969. California
Test Method 356 “ Methods of Test to
Determine Flexible Pavement
Rehabilitation Requirements By
Pavement Deflection Measurements,”( 4)
defined pavement rehabilitation
requirements on state highways in
California. During this time the primary
overlay material was dense graded
asphalt concrete.
In 1974, changes based on the
performance of newly constructed
highway projects under study since 1964
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
1- 2
simplified the procedure for determining
an AC overlay thickness ( 3). Revised
deflection attenuation data and tolerable
deflection levels of AC pavements were
also included.
In 1979, the “ Asphalt Concrete Overlay
Design Manual”( 5) was published. This
manual provided the methods: ( 1) to be
used for acquiring information regarding
the existing asphalt concrete pavement,
( 2) to design AC pavement overlays
using deflections for structural adequacy
based on California Test 356, ( 3) to
retard reflective cracking and ( 4) to
restore ride quality.
Environmental concerns and the State’s
commitment to recycle as much roadway
material as economics permit have also
influenced rehabilitation methods.
Consequently, over the past 21 years,
Caltrans* rehabilitation strategies have
increased in number with new materials,
interlayers, recycling of existing
pavements, and the addition of waste
products ( such as rubber) in asphalt
concrete.
A study was published in 1980 ( 6) that
reviewed the actual service life of
pavement overlays designed by
California Test Method 356. The design
period of the overlays in this study was
10 years. The average service life was
found to be 11.6 years. Judgment as to
length of service was recognized to be
entirely subjective and, thus, susceptible
to variation. However, the term “ service
life" was defined in this report as the
period of time until the extent of load-associated
alligator cracking or patching
reached a combined total of 30 percent
* The California Division of Highways became
the California Department of Transportation
( Caltrans) on July 1, 1973.
of the roadway wheel path areas. ( One
wheel path with continuous alligator
cracking was considered to be 50 percent
and continuous cracking in both wheel
paths would be 100%.)
1 – 20 Foreword
Headquarters Maintenance along with
each district office determines which
portions of the California highway
system are candidates for rehabilitation.
The Pavement Management System
( PMS) is the primary tool used in
determining where repairs are needed
and how available funds will be
apportioned statewide.
Besides rehabilitation for structural
adequacy, when an existing roadway is
being widened the existing pavement
should be brought up to the same life
expectancy as the new pavement.
The Pavement Condition Inventory, a
report generated under the PMS, will
“ trigger” a section of roadway when the
ride quality is poor. The design for
alleviating a poor ride problem should
also provide an increased service life for
the pavement ( normally 10 years).
When the new lanes are added, the
existing shoulder may be called upon to
carry a wheel path. A deflection study
will determine if it will support the new
loads or if any up- grade is necessary.
When construction requires that public
traffic be detoured to an existing street
or roadway for a period of time, a
before- and- after study may be necessary
to determine the extent of added distress
and to develop a recommendation to
bring the pavement back to its intended
service life.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
1- 3
Each design requires an evaluation based
on three components: providing
structural adequacy, retarding reflective
cracking from the underlying layer and
improving the ride quality.
The project engineer should consult the
regional or district materials engineer or
the district pavement engineer early in
the project development process in order
to reduce the lag time between
conception and construction of the
project. Pavement deflection studies and
rehabilitation recommendations should
be requested early in the process to
provide accurate information for
estimating project costs.
Development of a recommendation to
rehabilitate an existing AC pavement
requires collecting background data as
well as collecting field data. Thorough
investigation of the pavement surface,
deflection measurements of the existing
pavement and knowledge of the
subsurface conditions are all necessary.
Finally, all the assembled information
previously acquired, along with the
calculations, are used to determine the
amount of rehabilitation necessary to
return the roadway to an acceptable level
of service.
There are many variations in materials,
traffic loads and environment that affect
the performance of pavement structural
sections. This makes it impossible to
develop hard and fast rules for the
rehabilitation of pavements. Therefore,
the project engineer should rely on the
experience, judgment and guidance of
engineers in pertinent functional
engineering areas who are familiar with
design, construction, materials, and
maintenance of pavements in the
geographical area of the project.
The use of the metric system is
encouraged and prevalent in State
contracts. However, the English system
lends itself better to the use of this
manual since deflections in all previous
research and current field studies are
measured in thousands of an inch
( 0.001- inch) for the California
Deflectometer as well as other devices.
All calculations in this manual are in the
English system and final results are in
metric equivalent.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 1
CHAPTER 2
PERFORMING A
DEFLECTION STUDY
2 – 10 Equipment
Since the early 1960’ s, Caltrans research
data have been based on deflections
obtained by the " California Traveling
Deflectometer” ( 2) ( Photo 1). The trailer
consisted of a mechanical arm that
placed the probe between the dual
wheels on a single rear axle. The dual
wheels were reconfigured so that the
probe was easy to insert. The probe
measured the vertical movement
( deflection) of the pavement as the dual
wheels passed the site.
The Traveling California Deflectometer
built by Caltrans, was one of a kind and
operated for routine work until 1969 and
for research until about 1980. After it
was no longer practical to use the
California Traveling Deflectometer due
to the age of its electronics, the trailer
portion was retained,) and used to apply
loads to pavement measurement sites to
perpetuate the standard deflection
device. This is now referred to as the
" California Deflectometer". A
Benkelman Beam* ( Photo 2) is used to
measure the deflection at the site. Either
the California Traveling Deflectometer
or the California Deflectometer were
used in the development of Caltrans’
flexible pavement overlay design
method and all past research projects.
The California Deflectometer is
currently used to correlate other
deflection devices such as the falling
weight deflectometer ( Photos 3 and 4)
and Dynaflect ( Photo 5). Correlation is
done at least annually. For routine
deflection measurements since 1969
Caltrans has been using the Dynaflect.
For engineers and agencies outside
Caltrans using this design method,
consideration should be given to
correlating testing equipment with a
truck having the axle weight, tire
spacing, and tire pressure that conforms
to the specifications of the California
Deflectometer. ( See California Test
356).
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 2
Photo 1 – California Traveling Deflectometer
Photo 2 - Benkelman Beam
Photo courtesy of Roger Smith
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 3
Photo 3 - KUAB Falling Weight Deflectometer
Photo 4 – JILS Falling Weight Deflectometer
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 4
Photo 5 – Dynaflect
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 5
2 – 20 Establishing Test Sections
Test sections are representative portions
of a roadway being considered for
rehabilitation. They are selected as
being representative of the entire mile,
which helps to keep the amount of
preliminary survey work to a reasonable
level on projects that are several miles
long.
Traffic safety should always be
considered when selecting test sections.
Areas of inadequate sight distance
should be avoided. The district
coordinator or area maintenance
superintendent should be contacted for
assistance and for traffic control.
For two- lane highways, if the project is
less than a mile in length, the entire
project is considered the test section.
Pavement deflections are measured at
approximately 0.01- mile ( 0.02- km)
intervals in the outside wheel path
( OWP) in both lanes. When projects are
greater than a mile in length, a 0.20- mile
( 0.32- km) test section ( 21 deflection
readings at 0.01- mile intervals) is
selected to represent each lane mile. If
possible, test sections are staggered from
lane to lane to obtain a representative
coverage of the roadway.
For multi- lane highways if the project is
less than a mile in length the entire
project is considered the test section.
Pavement deflections are measured at
approximately 0.01- mile intervals in the
outside wheel path in both outside lanes.
If possible, at least one 0.20- mile test
section* is selected for each of the inner
lanes, with pavement deflections
* Normally, 0.20- mile test sections consist of 21
deflection readings at 0.01- mile intervals. The
measurements are usually made in the outside wheel
path or the location with the most distress.
measured in the OWP, wherever
possible. Side clearance to fixed objects
( i. e. guard railing) may make this
unattainable.
If the multi- lane project is greater than a
mile in length, a 0.20- mile test section
should be selected for each mile for both
outer lanes. For each five miles of
roadway, one 0.20- mile test section
should be selected for each of the inner
lanes. Additional test sections will be
required if structural section changes
occur and/ or roadway appearances are
not uniform.
Pavement deflections should be
measured from the beginning to the end
on ramps or connectors using the whole
as a test section. Short ramps require a
short testing interval in order to obtain
sufficient readings. Extremely long
ramps testing may be longer than the
normal 0.2- mile test section. If the ramp
is closed to traffic when testing, the
wheel path with the most distress should
be used. Otherwise the wheel path that
allows the traffic to pass safely is tested.
Shoulders that will carry future traffic
should have test sections established
according to the normal procedure in this
section.
Sometimes state highways are also city
streets. Test section determinations on
city streets are performed in the same
manner as described for two- lane
roadways and multilane facilities. It is
often necessary to select a greater
number of test sections on city streets or
test continuously due to frequent
changes in structural section and/ or
roadway condition.
Engineering judgment should always be
used in selecting the number of test sites
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 6
for pavement deflection measurements.
The suggested test frequencies described
above are the minimum number
recommended. Structural section
changes are not always clearly visible in
the field, but can usually be located from
large changes in deflection
measurements and confirmed by core
data. Therefore, whenever there appears
to be a need for additional information,
make as many deflection measurements
as necessary.
As the need for additional lanes has
occurred, widening of the roadway has
sometimes created two different
structural sections even within a single
lane. These can usually be noticed by a
longitudinal crack at the joint. A test
section on each of the structural sections
should be selected for use in the
rehabilitation study.
Occasionally, a return to a project may
be required for additional testing after
reviewing the initial deflection data in
the office.
2 – 30 Pavement Background
Information
Background information is obtained
from both the Region/ District and the
files of the Structural Section Design and
Rehabilitation ( SSD& R) Branch, Office
of Materials Engineering and Testing
Services.
When requesting a pavement deflection
study from SSD& R, the District
Materials Engineer ( DME) should
provide at least the following
information: the original structural
section data, maintenance overlays and
date of placement, and the project’s
design Traffic Index ( TI).
SSD& R has records on previous
deflection studies. If the District’s
records show no maintenance or
rehabilitation was done for the project in
question, the previous structural section
data may be used. If information is
limited or not available, pavement cores
must be removed to provide this
information.
Previous deflection studies for the
project found in the SSD& R files, where
no maintenance or rehabilitation has
been done, can be used to determine a
rate- of- change in deflections that may be
considered when designing new
rehabilitation strategies. Normally,
deflections increase with age beginning
several months after construction if the
pavement is under traffic loads.
A previous study may have been done
when moisture was in the structural
section. Consequently, those deflections
may be higher than in the current study.
If this happens, the previous, higher
deflections should be used to design the
current rehabilitation.
2 – 40 Collecting Field Data
A pavement condition inspection is as
important to the design engineer as the
deflection values. The function of the
pavement condition inspection is to
obtain the necessary information to be
used in conjunction with the evaluated
deflection values to determine the
appropriate rehabilitation strategies.
The pavement condition inspection
provides data that may convince the
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 7
designer to adjust the rehabilitation to
meet the special requirements of that
section of pavement.
The inspection of the project should
describe the general condition of the
pavement in terms of visual appearance
including the type, severity and extent of
distress. This should include items such
as rutting, bleeding, raveling, patching,
potholes, shoving ( sometimes called
slippage), corrugations, pumping,
delamination, and the various types of
cracking.
Also, an inspection of the project should
include other details that should be
recorded such as the existing structural
section changes; permanent vertical
control features that will limit an
increase in profile grade; any localized
drainage problems; embankment
settlement; and areas of deep cuts and
fills within the test section.
Representative test sections and other
important features recorded, such as
failed areas whether tested or not tested
should be photographed and recorded.
Air and pavement temperatures should
be measured. Date and time of
measurement should be recorded.
2 – 50 Measuring Deflections
California Test 356 should be consulted
when pavement deflection
measurements will be obtained with
different testing devices ( 4). A copy of
the test method can be downloaded from
the following Caltrans’ Internet address:
( Address as of June 2001)
www. dot. ca. gov/ hq/ esc/ ctms/ index. html
2 – 60 Converting to Equivalent
Deflectometer Values
Caltrans, at the present time, uses the
Dynaflect as its primary deflection-measuring
device. Although repeatable
instruments, each Dynaflect has a unique
correlation curve. The correlation curve
for each Dynaflect vs. California
Deflectometer has been determined,
through experience and testing, to be
non- linear and unique. SSD& R has a
conversion chart for each of its
Dynaflects to be used to convert each
individual deflection measurement to the
equivalent California Deflectometer.
When using the falling weight
deflectometer, convert the mean and 80th
percentile deflection values to equivalent
California Deflectometer values using
appropriate correlation curves.
A comparison of each deflection-measuring
device to the California
Deflectometer should be performed at
least once a year. The correlation curve
results can be calculated and placed in a
conversion chart for ease of use.
2 - 70 Mean and 80th Percentile
Deflections
Individual deflection readings for each
test section should be reviewed prior to
determining mean and 80th percentile
values. This review may locate possible
areas that are not representative of the
entire test section.
An example would be a localized failure
with a very high deflection. It may be
more cost effective to repair the various
failed sections prior to rehabilitation
Thus, the high deflection values in the
repaired areas would not be included
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 8
when calculating mean and 80th
percentile values for the representative
test sections.
When similar deflection values are
found within a test section they should
be analyzed as a group. There may be
several groups within the test section or
only one. If all deflections are similar,
the entire test section is analyzed as a
whole. With this in mind, engineering
judgment must always be utilized in
analyzing deflection data.
The mean deflection level for a test
section is determined by dividing the
number of individual deflection
measurements into the sum of the
deflections.
x =
n
D i Σ
Where:
x = mean deflection for a test section
Di = an individual deflection
measurement in the test section
n = number of measurements in the
test section
The 80th percentile deflection value
represents a deflection level at which
approximately 80 percent of all
deflections are less than the calculated
value and 20 percent are greater than the
value. Thus the design will provide
thicker rehabilitation than using the
mean value. The 80th percentile
deflection values are obtained using the
following equation:
D x 0.84s 80 = +
Where:
x = mean deflection for a test section
D80 = 80th percentile of the
deflections at the surface for a
test section, in inches
s = standard deviation of all test
points for a test section
( )
1
2
−
Σ −
=
n
s Di x
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 9
Example 2- 1: Determine the mean and
the 80th percentile from the evaluated
deflection values ( assume no isolated
failures). The Dynaflect data obtained
from a 0.20- mile ( 0.32- km) test section
is converted to the equivalent California
Deflectometer deflection as follows:
Test
Dynaflect
Readings
( in. x 10 - 3)
California
Deflectometer
Deflection
( inch)
1 1.68 0.035
2 1.43 0.031
3 1.21 0.027
4 1.92 0.039
5 2.08 0.041
6 1.66 0.035
7 1.73 0.036
8 1.59 0.034
9 1.83 0.037
10 1.74 0.036
11 1.50 0.032
12 1.40 0.030
13 1.39 0.030
14 1.58 0.033
15 1.63 0.034
16 1.79 0.037
17 1.90 0.038
18 1.66 0.035
19 1.74 0.036
20 1.54 0.033
21 1.73 0.036
Solution 2- 1:
Sum of 21 deflections = 0.725 inch
s = 0.0033 inch
x =
n
D i Σ
= 0.725/ 21 = 0.0345 inch
D x 0.84s 80 = +
= 0.0345 inch + 0.84( 0.0033 inch)
D80 = 0.037 inch ( 0.940 mm).
2 - 80 Preparing a Deflection Map
A deflection map is a sketch of the
project illustrating D80 deflection levels
for each test section. The purpose of the
deflection map is to show a visual
representation in order to determine if
certain areas of the project should be
grouped and analyzed separately ( by
observing the differences in deflection
levels). Deflection values should not
only be looked at along the lane, but
from lane to lane and travel direction.
Traffic can vary considerably from lane
to lane and in opposing directions, thus
causing different distress and deflection
levels. Rehabilitation requirements and
limits can then be determined for each
direction or lane.
See Figure 1 for an example of a
" deflection map."
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
2- 10
Figure 1
EXAMPLE OF DEFLECTION MAP
( Not to Scale)
Project Limits
Avg. 80th percentile
deflection level
= 0.029 in.
Avg. 80th percentile
deflection level
= 0.015 in.
Date Tested: 4/ 21/ 00
Traffic Index ( 10 year) = 9.0
Note: All deflections are in terms of equivalent California Deflectometer values.
PM 59.0
PM 60.0
PM 61.0
PM 62.0
PM 63.0
PM 64.0
PM 65.0
PM 66.0
24
28
30 30
28 34
13
20 18 13
14 15 14
NB
SB
Photo 1
Photo 2
Photo 3
Photo 4
Photo 5
Photo 6
Photo 7
Photo 13
Photo 12
Photo 10
Photo 9
Photo 8
Photo 11
80th percentile deflections ( 10- 3 inches) for
0.02- mile test sections.
Pavement deflections taken at 0.01- mile intervals.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 1
CHAPTER 3
INTRODUCTION TO
AC PAVEMENT
REHABILITATION DESIGN
Currently, Caltrans uses the philosophy
of extending service life of a pavement
for a 10- year period of time for
rehabilitation. However, the design
engineer may request pavement
rehabilitation for a different design
period. The design procedure is the
same for a different design period. It is
accomplished by using the appropriate
Traffic Index ( TI) for the period of time
for the design.
When determining the rehabilitation
alternatives, the engineer must know
both the design period requested and the
TI for the pavement being evaluated.
Traffic Index is a measure of the number
of equivalent 18,000- lb ( 80- kN) single
axle loads ( ESAL’s) expected in the
design lane over the design period. The
TI does not vary directly with ESAL’s
but rather exponentially according to the
following formula as illustrated in Table
603.4A of the “ Highway Design
Manual.” ( 8)
TI = 9.0 ( ESAL / 106 ) 0.119
Where: TI = Traffic Index
ESAL = Equivalent 18,000- lb
Single Axle Loads
If that TI is unknown and the 10- year TI
is known, use Table 603.4A in the HDM
to establish the ESAL’s. Then
proportion from the ten- year ESAL to
the ESAL of the new design life. Finally
select the corresponding TI for the new
ESAL.
There are three components to be
considered when designing flexible
pavement rehabilitation:
1) Structural adequacy upgrade;
2) Reflective crack retardation; and
3) Ride quality improvement.
3 - 10 Designing for Structural
Adequacy
Deflections are used for determining the
thickness requirements for rehabilitation
of asphalt concrete ( AC) pavements
when considering structural section
adequacy. Condition and structural
section of the existing roadbed together
with measured deflections and the
projected TI provide the majority of the
information to be used during
consideration for structural adequacy.
Once the data has been collected and the
deflections of the test sections have been
reduced to 80th percentile deflections
( D80’ s) and placed on a project
deflection map, the design process
involves both calculations and
engineering judgment.
The project deflection map should be
examined for similar D80 values.
Adjacent test sections with similar D80
values should be grouped together.
There may be several groups within the
project or only one. If all D80 values are
similar, the entire project may be
analyzed as a whole.
A group is a collection of adjacent test
sections that have similar values for the:
• Average 80th percentile deflection
( D80).
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 2
• Average existing asphalt concrete
pavement thickness.
• Type of base.
• Traffic Index ( TI).
Each of these has an influence on the
rehabilitation.
Test sections should not be grouped
together if the existing AC thickness
varies more than 0.10 ft ( 30 mm), the
type of base materials is different, or the
TI is different. These influence the
tolerable deflection level that is used in
determining the rehabilitation. Similar
groups of test sections can be analyzed
together.
D80 values should not be examined only
along the lane, but should be examined
from lane to lane and in the direction of
travel. Traffic may vary considerably
from lane to lane and in opposing
directions, thus causing different distress
and deflection levels. Rehabilitation
requirements and limitations should be
considered for each direction or lane as
is appropriate from the data and to meet
the needs of the project site.
Suggestion:
In selecting groups of similar D80 values,
it is suggested that only adjacent test
sections with D80 values that differ less
than about 0.010 inch ( 0.254 mm)
should be grouped together. More than
0.010- inch difference will most likely
produce different thickness
requirements.
Once groups with similar D80 values,
structural sections, types of bases and
TI’s have been identified; average the
D80 values for each group. Use the TI,
average existing AC pavement thickness
and type of base to determine the
tolerable deflection at the surface ( TDS)
for the group using Table 1 ( Chapter 6).
For existing AC pavement over
untreated base material such as
aggregate base ( AB), native material,
etc., use the TDS corresponding to the
thickness of the existing AC pavement
and the appropriate TI.
For existing AC pavement over treated
base or portland cement concrete ( PCC),
use the TDS values in the row for CTB
and the column for appropriate TI.
However, if the underlying CTB
thickness is less that 0.35 ft ( 105 mm),
consider it an untreated base and
determine the TDS from the upper part
of the table corresponding to the
thickness of the existing AC pavement
and the appropriate TI.
In locations where existing AC
pavement is over a treated base and the
measured deflections are high, the
treated base may not be performing as it
should. The treated base layer is no
longer carrying the load as it was
originally designed to do. It may have
deteriorated to the point where this layer
is acting more like an untreated base. In
this case, the rehabilitation should be
designed as though the existing
structural section is AC over untreated
base.
Choosing the appropriate existing
structural section interaction – AC over
treated base or AC over untreated base –
should be made carefully. This choice
will greatly influence the TDS, and the
resulting thickness of rehabilitation
strategies.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 3
Suggestion:
When the D80 value for deflections of a
test section of AC pavement over treated
bases is greater than 0.014 inch, the
rehabilitation may be designed as
though the existing structural section is
AC over untreated base.
Engineering judgment is required when
trying to determine if the treated base is
performing satisfactorily. Besides
deflections, the condition of the
pavement surface should be considered.
The greater the amount of alligator
cracking, the more likely the treated base
layer is not carrying the load as
originally designed. But, if the distress
is mainly transverse cracks without
alligator cracking and localized failures,
the treated base is probably still intact
even though the D80 value is greater than
0.014 inch. D80 values of as high as
0.020 inch have been measured over
intact treated base.
Occasionally on an older pavement, only
minimal distress is apparent from the
condition survey and the average D80 is
less than the TDS. In this case,
corrective repair may not be necessary
other than a seal coat that will seal
cracks, improve appearance, delay
oxidation of the asphalt concrete and
prolong the pavement life.
If the average D80 is greater than the
TDS, determine the required percent
reduction in deflection at the surface
( PRD) to restore structural adequacy as
follows:
PRD = ( 100)
80
80
AverageD
AverageD − TDS
Where:
PRD = Percent Reduction in
Deflection Required at the
surface, as percent
TDS = Tolerable Deflection at the
Surface, in inches
D80 = 80th Percentile of the
Deflections at the surface for a
test section in inches
In Caltrans, structural section design is
based on the concept of gravel
equivalence. This same concept is used
in flexible pavement rehabilitation.
For rehabilitation, the additional gravel
equivalence ( GE) required is determined
from the calculated percent reduction in
deflection and Table 2. It is the amount
of AS gravel that will provide sufficient
strength to reduce the deflections to the
tolerable level.
A gravel factor ( Gf) expresses the
relative value of various materials when
compared to gravel. The gravel factor
( Gf) is given to a material that, when
divided into the gravel equivalence
required, will provide the layer thickness
of the material.
Note that for new pavement design the
Gf for asphalt concrete varies with the
TI [ Table 608.4, of the Highway Design
Manual]. However, for most types of
rehabilitation ( overlay design being
one), the Gf has been established at 1.9
for all Traffic Indices.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 4
Commonly Used Gf for Rehabilitation
Asphalt Concrete 1.9
Hot Recycled
Asphalt Concrete 1.9
Cold Recycled
Asphalt Concrete 1.5
AC Below the
Analytical Depth 1.4
Aggregate Base 1.1
Aggregate
Subbase 1.0
Native Soil 0
When the volume of traffic increases to
the level that new lanes should be added,
the existing shoulder may be called upon
to carry a wheel path. If the shoulder
pavement has not carried traffic loads
and fatigue cracking is absent,
engineering judgment is required to
analyze the measured deflections on the
shoulder.
Oxidized asphalt pavement may be
“ bridging” rather than producing a
deflection basin. The deflections would
be lower than for a normal deflection
basin. To assist in making a
determination on whether the pavement
is bridging, removed cores may be
brought to the lab for testing the in- place
asphalt properties. This should be
emphasized especially when the lighter
deflection equipment is used.
If the design TI is high, a new structural
section designed using the R- value of the
underlying material may be appropriate
for the shoulder turned into a lane. This
is especially applicable when an increase
in the profile grade is limited.
Example3- 1: Determine AC overlay
thickness requirements to restore
structural adequacy. The 10- year Traffic
Index ( TI10) is 11.0.
Location 80th Percentile
Deflection
Existing Structural
Section
PM 1.00 to
PM 3.50
0.025 inch 0.40 foot AC
0.67 foot AB
1.00 foot AS
Solution 3- 1:
Given: TI10 = 11.0
Average D80 = 0.025 inch
AC thickness = 0.40 ft
Step 1:
Obtain tolerable deflection at the
surface ( TDS).
Use Table 1 ( Chapter 6):
AC = 0.40 ft and TI = 11.0
TDS = 0.012 inch
Step 2:
Compare average D80 to TDS.
0.025 > 0.012
Rehabilitation for structural
adequacy is indicated.
Stem 3:
Calculate Percent Reduction in
Deflection ( PRD) required.
( 100) 52%
0.025
012 . 0 025 . 0 =  

 
 −
Step 4:
Determine Gravel Equivalence
( GE) required for deflection
reduction.
Use Table 2; Column A)
GE = 0.68 ft
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 5
Step 5:
Determine the required thickness
of AC overlay.
ft
G
Overlay GE
f
0.36
1.9
= = 0.68 =
Round to 0.35 ft ( 105 mm).
Recommendation: 0.35- ft ( 105- mm)
overlay of DGAC.
3 - 20 Design Governed by Reflective
Cracking
Reflective crack retardation of the new
overlay needs to be considered.
Retarding the propagation of cracks
from the existing pavement into the new
AC overlay will extend its service life.
For AC pavements over untreated bases,
the thickness of a new DGAC overlay
should be at least half the thickness of
the existing asphalt concrete up to a
maximum of 0.35 ft ( 105 mm). Or, if
the existing AC pavement is to be
milled, the thickness of the new AC
should be half the thickness of the
remaining pavement up to a maximum
of 0.35 ft.
For AC pavements over a treated base or
PCC the general guideline ( exceptions
will occur) for a ten- year design is a
minimum overlay of 0.35 ft ( 105 mm) of
new dense graded asphalt concrete
( DGAC). This was developed by
experience and is usually adequate for
retarding reflective cracks. An
exception might be when the underlying
material is a thick PCC such as on an
overlaid PCC freeway that was not
cracked and seated. In this case a
minimum thickness of 0.45 ft. ( 135 mm)
may be appropriate.
For a design life different from a 10- year
design, a slight modification changes the
thickness. For a five- year design,
experience has determined the thickness
should be approximately 75 percent of
the ten- year design thickness. For a
twenty- year design, use 125 percent.
As always, exceptions will occur and
engineering judgment will be necessary
for final design. Factors to be
considered that might influence the
engineer to increase the thickness are:
( 1) Type, sizes, and amounts of surface
cracks.
( 2) Extent of localized failures.
( 3) Existing structural section material
and age.
( 4) Thickness and performance of
previous rehabilitation.
( 5) Environmental factors.
( 6) Anticipated future traffic loads
( Traffic Index).
Unfortunately, there are no set criteria
that will aid the engineer in the decision
process in regards to designing to
prevent reflective cracking. Experience
with similar roadways repaired in the
general area; past overlays and their
performance; and discussions with local
maintenance and construction personnel
are all part of the data gathered to be
considered in the final decision and
engineering judgment process.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 6
3 - 30 Design Governed by Ride
Quality
The Pavement Management System
records ride quality as part of their
pavement condition inventory. The
International Roughness Index ( IRI) for
each lane is measured for the Pavement
Condition Survey. ( IRI has replaced the
Ride Score. The Ride Score of 45 or
more will “ trigger” a project. The
equivalent IRI is not yet determined.)
When ride quality measurements
indicate that the pavement needs
improvement, procedures are needed to
smooth the pavement. At least two
options emerge as viable solutions:
( 1) Place an asphalt concrete overlay
thick enough to be placed in two lifts
[ 0.25- ft ( 75- mm) minimum].
( 2) Cold plane the existing pavement
prior to placing the new asphalt
concrete.
Ride quality will ultimately govern the
rehabilitation strategy design if the
requirements for structural adequacy and
reflective crack retardation are less than
0.25 ft ( 75 mm).
Please note that if the two- lift option is
chosen, the July 1999 Standard
Specification Section 39– 6.01( 9) gives
the contractor the option to place 0.25 ft
( 75 mm) in one layer. Any rehabilitation
report that recommends this overlay
thickness for improving the ride quality,
should point out in the report that the
overlay needs to be placed in two layers
and specified as such in the project
special provisions.
3 - 40 Choosing the Design
Recommendation
The final choice of the recommended
rehabilitation alternative is based on
choosing a strategy that will provide a
total structural section thickness that is
adequate to resist the anticipated loading
it will experience throughout its design
period, the potential for reflective
cracking and to improve ride.
Once the rehabilitation strategies have
been determined to correct for lack of
structural adequacy, to retard reflective
cracking and to improve ride quality, a
single strategy must be chosen which
will be sufficient for all three conditions.
In addition to choosing a rehabilitation
strategy to correct the three criteria listed
earlier, constructability concerns must be
addressed.
Prior to placement of asphalt concrete on
an existing pavement, some preparation
is required besides what is specified in
Standard Specifications 39- 4.01. Cracks
wider than 0.25 inch ( 5 mm) should be
sealed; loose and/ or spalling pavement
removed; and potholes and localized
failures repaired. Routing cracks before
applying crack sealant has been found to
be beneficial. The width of the routing
should be 0.25 inch ( 5 mm) wider than
the crack width. The depth should be
equal to the width of the routing plus
0.25 inch ( 5 mm). In order to alleviate
the potential bump in the overlay from
the crack sealant, leave the crack sealant
0.25 inch ( 5 mm) below grade to allow
for expansion. Design recommendations
should include a reminder of these
preparations.
It has been found that during
construction a dense graded AC layer of
less than 45 mm ( 0.15 ft) may cool
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
3- 7
considerably before adequate
compaction occurs. Therefore, for a
surface course for rehabilitation,
Caltrans generally uses a minimum
thickness of 45 mm ( 0.15 ft). The
minimum thickness for rubberized AC is
30 mm ( 0.10 ft) since it is placed at a
higher temperature.
Structural Section Design and
Rehabilitation Branch ( SSD& R) designs
asphalt concrete thicknesses in 0.05- ft
increments. With the change to metric
values, SSD& R increases the
thicknesses by 15 mm for each 0.05- ft
increment ( Table 5). Please note that
this is not an exact mathematical
conversion.
Example3- 2: Determine the AC
rehabilitation requirements. The 10- year
Traffic Index ( TI10) is 11.0. There are
no restrictions on an increase in profile
grade.
Location 80th Percentile
Deflection
Existing Structural
Section
PM 1.00 to
PM 3.50
0.025 inch 0.40 foot AC
0.67 foot AB
1.00 foot AS
Solution 3- 2:
Recommendations to be considered:
Structural Adequacy:
A 0.35- ft DGAC overlay. Refer
to Example 3- 1. ( Rubber AC
alternatives are discussed in
Section 4- 20 of this manual.)
Reflective Cracking:
A 0.20- ft DGAC overlay. ( One-half
existing AC thickness.)
Ride Quality:
A 0.25- ft DGAC overlay placed
in two layers.
( Section 3- 30).
Discussion 3- 2:
• A second option for ride quality is to
mill the existing rough pavement to
remove much of the surface
undulations prior to placing the new
AC overlay. Milling off 0.10 to 0.20
ft ( 30 to 60 mm) will usually be
sufficient. Milling will change D80
and require additional design
calculations.
• Cold planing and replacing the
existing surface with DGAC or hot
recycled AC to the same grade
would provide a good solution for
reflective cracking and ride quality.
( This is discussed in Section 4- 50 of
this manual.)
• A 0.35- ft DGAC overlay may
increase the profile grade beyond the
allowable if there are restraints such
as are found in urban areas.
Recommendation 3- 2: 0.35- ft ( 105- mm)
overlay of DGAC.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 1
CHAPTER 4
FLEXIBLE PAVEMENT
REHABILITATION DESIGN
GUIDE
4 – 10 Basic Overlay Using DGAC
See Chapter 3 for a complete discussion
of Basic Overlay design using dense
graded asphalt concrete.
Dynaflect deflection values ( Caltrans
primary deflection device) converted to
equivalent California Deflectometer
values are used for determination of
overlay thickness.
1. Calculate Mean *
x =
n
D i Σ
2. Calculate Standard Deviation**
( )
1
2
−
Σ −
=
n
s Di x
where:
x = mean deflection for a test
section
80 D = 80th percentile of the
deflections at the surface for a
test section in inches
* When determining the Mean, omit any
individual measurements on isolated failures
since recommendations in the report will be to
replace these failures.
** ( D x) i − is the difference between each
individual measurement and the mean value.
The number of measurements is designated n.
s = standard deviation of all
deflections for a test section
i D = an individual deflection
measurement in the test section
n = number of measurements in the
test section
3. Calculate the 80th percentile
D x 0.84s 80 = +
4. Determine the Tolerable Deflection at
the Surface ( TDS).
Determine the TDS from the Tolerable
Deflection Chart ( Table 1) with the
design Traffic Index ( TI) and either the
thickness of the existing asphalt concrete
( AC) pavement or the type of base data.
If D80 is at or below the TDS, then the
pavement is considered structurally
adequate and any overlay thickness
should be based on reflective crack
retardation and/ or ride score reduction.
If D80 is greater than the TDS, then the
overlay required for structural adequacy
is determined along with the need for
reflective crack retardation and/ or ride
score reduction.
5. Calculate the Percent Reduction in
Deflection at the surface:
PRD = ( 100)
80
80
D
D − TDS
Where:
PRD = Percent Reduction in
Deflection required at the surface
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 2
TDS = Tolerable Deflection at the
Surface, in inches
D80 = 80th Percentile of the
Deflections at the Surface for a
test section in inches. If test
sections have been grouped, then
average D80 for the group is used.
6. Determine the increase in Gravel
Equivalence ( GE) required to reduce D80
to the TDS. Utilizing the calculated
PRD value, go to Table 2, Column A, to
determine the GE. ( Discussion of an AC
overlay placed on a cushion course is in
Section 4- 70.)
7. Determine the Gravel Factor, Gf. For
a dense graded asphalt concrete ( DGAC)
overlay over an existing AC pavement
use a Gf of 1.9 regardless of thickness
and TI.
8. Determine the overlay thickness for
structural adequacy.
f G
overlay = GE
9. Determine the overlay thickness for
reflective cracking.
overlay = A minimum of half of
the existing AC
thickness ( Section 3- 20)
10. Determine the overlay thickness for
ride quality.
overlay = A minimum of 0.25 ft
placed in two layers
( Section 3- 30)
Example 4- 1: Determine the
recommended AC overlay thickness for
an existing AC pavement.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
10.0 0.030 inch 0.55 foot AC
0.50 foot AB
1.00 foot AS
Existing conditions:
• Occasional to intermittent alligator,
transverse, and longitudinal cracks,
( some 0.5 inch wide).
• Fairly smooth ride.
Calculations 4- 1:
Check for overlay thickness required
for structural adequacy.
Step 1:
Obtain tolerable deflection at the
surface ( TDS).
Use Table 1:
AC = 0.55 ft and TI = 10.0
TDS = 0.012 inch
Step 2:
Compare average D80 to TDS.
0.030 > 0.012
Stem 3:
Calculate Percent Reduction in
Deflection required.
( 100) 60%
0.030
012 . 0 030 . 0 =  

 
 −
Step 4:
Determine Gravel Equivalence
( GE) required for deflection
reduction.
Use Table 2; Column A
GE = 0.85 ft
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 3
Step 5:
Determine the required thickness
of AC overlay for structural
adequacy.
ft
G
Overlay GE
f
0.45
1.9
= = 0.85 =
Check for the overlay thickness
required for reflective crack
retardation.
To retard reflective cracks entering the
new overlay from the pavement below
choose a thickness for the new overlay at
least one- half the thickness of the
existing AC pavement being overlaid
( up to a maximum of 0.35 ft ( 105 mm)
for an underlying aggregate base).
Determine half of the existing pavement
thickness:
overlay =
2
0.55 = 0.275 Round
to 0.30 ft.
Check for smoothness.
The ride quality was previously
determined to be acceptable. If it were
not acceptable, a 0.25- ft ( 75- mm)
DGAC overlay would have to be placed
in two layers.
Discussion 4- 1:
• Since reflective cracking
requirement is less than 0.45 ft ( 135-
mm), and since smoothness is
satisfactory, structural adequacy
governs the overlay design thickness.
• For this overlay example, reflective
cracking could never control, since
the structural requirement of 0.45 ft
( 135 mm) is already above the 0.35-
ft ( 105- mm) maximum for reflection.
• In this example, if the structural
requirement had been less than 0.30
ft ( 90 mm) and the ride quality
needed improvement, then reflective
cracking would be the controlling
criteria with a required overlay of
0.30- ft ( 90- mm).
• To make a rough- riding pavement
smoother by using a minimum of
two procedures, a mill- and- replace
procedure or a procedure that places
an AC overlay in two layers must be
used. The design of the overlay for
ride consideration would be as
follows:
Option 1 - The overlay must be
thick enough to allow for two
layers to be placed. The 0.45- ft
( 135- mm) DGAC overlay for
structural adequacy will provide
the two layers needed for
improving the ride quality.
Option 2 - Mill the existing
rough pavement to remove much
of the surface undulations prior
to placing the new AC overlay.
Milling off 0.10 to 0.20 ft ( 30 to
60 mm) will usually be
sufficient. Milling will change
D80 and require additional design
calculations.
Recommendation 4- 1: 0.45- ft ( 135- mm)
overlay of DGAC.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 4
4 – 20 Rubberized Asphalt Concrete
( Type G) *
Caltrans standard overlay design is a
dense graded asphalt concrete overlay
thickness that will improve the
serviceability for the time frame
specified, usually a ten- year period.
From that design thickness an alternate
design with a thickness of rubberized
asphalt concrete, gap graded ( RAC Type
G) can be determined.
A thickness equivalency of not more
than 1: 2 is given to the RAC Type G
when compared to the dense graded
asphalt concrete ( DGAC) for structural
adequacy or reflective crack retardation.
The equivalencies are tabulated in
Tables 3 and 4
Using RAC Type G instead of DGAC
allows a lower profile grade and reduces
the amount of asphalt concrete materials
used.
The minimum thickness for RAC Type
G is 0.10 ft ( 30 mm). Until further
research, the maximum thickness For
RAC Type G is limited by stability to
0.20 ft ( 60 mm). If the design calls for a
thicker overlay, then a DGAC layer may
be placed prior to placing the RAC Type
G. For example, if the design calls for a
0.55- ft ( 165- mm) DGAC overlay, a
0.15- ft ( 45- mm) layer of DGAC could
be placed first. Then the 0.40- foot ( 120-
mm) DGAC remaining can be replaced
with 0.20- ft ( 60- mm) of RAC Type G
placed as the top layer ( Table 3).
* Data from field and laboratory studies were
used to produce Caltrans internal memorandum
“ Asphalt Rubber Hot Mix – Gap Graded
Thickness Determination Guide” dated March
19, 1992.
A Rubberized Stress Absorbing
Membrane Interlayer ( SAMI- R) may be
used to provide some strength when
placed under RAC Type G. For
structural strength, a SAMI- R is
considered to provide an equivalence of
0.05 ft ( 15 mm) of RAC Type G ( Table
3). For reflective crack retardation from
wide cracks, the SAMI- R is considered
to provide either 0.05 ft ( 15 mm) when
the underlying base is a treated material
or 0.10 ft ( 30 mm) when the underlying
base is an untreated material ( Table 4).
However, it should be noted that RAC
Type G might not prevent cold weather
cracking. A Fabric Stress Absorbing
Membrane Interlayer ( SAMI- F) is not to
be used under RAC Type G because the
high placement temperature of the RAC
Type G is close to the melting
temperature of the SAMI- F material.
Just as with DGAC, prior to placement
of RAC Type G on an existing
pavement, some preparation is required.
Cracks wider than 0.25 inch ( 5 mm)
should be sealed, and potholes and
localized failures repaired.
It is undesirable to place RAC Type G in
areas that will not allow surface water to
drain. As an example, on a surface that
is milled only on the traveled way and
not on the shoulders, thus forming a
“ bathtub” section. To offset that
situation, a combination of materials
might fit the design; for example, place a
layer of DGAC to the original grade
prior to placing the RAC Type G, or mill
the shoulders to slope for drainage.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 5
4 – 30 Stress Absorbing Membrane
Interlayers
Two types of Stress Absorbing
Membrane Interlayers are used for
rehabilitation:
1.) Rubberized ( SAMI- R).
2.) Fabric ( SAMI- F).
Placing a rubberized stress absorbing
membrane interlayer on a pavement
consists of an application of asphalt-rubber
binder on the surface followed
with aggregate screenings that are pre-coated
with paving asphalt.
Placing a fabric stress absorbing
membrane interlayer on a pavement
consists of an application of asphalt
binder on the surface followed with the
fabric. The fabric is manufactured from
polyester, polypropylene or
polypropylene- nylon material that is
non- woven and heat treated on one side.
See Standard Specifications 39- 4.03.
SAMI’s are used to retard reflective
cracks, prevent water intrusion, and in
the case of SAMI- R, enhance structural
strength ( Table 3).
Judgment is required when considering
the use of SAMI’s.
• Consideration should be given to
areas that may prohibit surface water
from draining out the sides of the
overlay, thus forming a “ bathtub”
section.
• Since SAMI’s act as a moisture
barrier, they should be used with
caution in hot environments where
they could prevent underlying
moisture from evaporating.
Moisture trapped within the asphalt
concrete, under wheel loads, may
provide a means by which the asphalt
would be washed off the aggregates.
This action is called stripping. Some
mixes are more susceptible to this action
than others. When AC is to be placed in
these types of locations the aggregates
should be treated prior to mixing.
A SAMI may be placed between layers
of new asphalt concrete ( AC), such as on
a leveling course, or on the surface of an
existing AC pavement. When placed on
an existing AC pavement some
preparation is required to prevent excess
stress on the membrane. This includes
sealing cracks wider than 0.25 inch ( 5
mm), and repairing potholes and
localized failures.
SAMI- R:
Placed Under Rubberized Asphalt
Concrete
Structural Strength – A SAMI- R also
may be used to provide some structural
strength when placed under an RAC
Type G overlay that is designed for
structural adequacy. The SAMI- R in
this case is considered to be
approximately 0.05 ft ( 15 mm) of RAC
Type G for structural strength ( Table 3).
Reflective Cracking – A SAMI- R is
considered to be equivalent to 0.05 ft ( 15
mm) of RAC Type G when the
underlying base of the structural section
is a treated base. When the underlying
base is an untreated base, a SAMI- R is
equivalent to 0.10 ft ( 30 mm) of an RAC
Type G ( Table 4).
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 6
Placed Under Non- Rubberized Asphalt
Concrete
When a SAMI- R is placed under non-rubberized
asphalt concrete designed for
reflective crack retardation, the
equivalence of a SAMI- R depends upon
the type of base material under the
existing pavement. When the base is a
treated material, a SAMI- R placed under
DGAC or open graded asphalt concrete
( OGAC) is considered to be equivalent
to 0.10 ft ( 30 mm) of DGAC. When the
base is an untreated material SAMI- R is
equivalent to 0.15 ft ( 45 mm) of DGAC.
SAMI- F:
A Fabric Stress Absorbing Membrane
Interlayer ( SAMI- F), also called
pavement reinforcing fabric ( PRF),
placed under DGAC designed for
reflective crack retardation provides the
equivalent of 0.10 ft ( 30 mm) of DGAC.
This allows the project engineer to
decrease the new profile grade and also
save asphalt concrete materials.
If the road to be rehabilitated has a high
proportion of small radius horizontal
curves, the use of SAMI- F is probably
not cost effective due to the extra labor
involved during placement.
A SAMI- F should not be placed directly
on coarse surfaces such as a chip seal,
OGAC, areas of numerous rough patches
or on a pavement that has been cold
planed. Coarse surfaces may penetrate
the fabric and/ or the paving asphalt
binder used to saturate the fabric may be
“ lost” in the voids or valleys leaving
areas of the fabric dry. For the SAMI- F
to be effective in these areas, use a
leveling course of DGAC prior to the
placement of the SAMI- F.
Saturating the fabric with asphalt
enhances the properties of the pavement
reinforcing fabric. The fabric is placed
on the asphalt concrete pavement that
has had a heavy tack coat of asphalt
applied. However, on a cool day the
tack coat may cool rapidly, until it
reaches the temperature of the pavement.
In this case, the tack asphalt usually will
remain tacky enough to hold the fabric
in place, but full saturation will not
occur. Therefore, it is up to the heat of
the asphalt concrete overlay to re- melt
the tack coat, allowing it to infiltrate the
fabric. With normal heat and rolling
pressure of the first layer of asphalt
concrete, the fabric should become
saturated. On warm days, the fabric may
come close to full saturation just by
lying on the asphalt tack coat because
the asphalt stays liquid longer.
SAMI- F’s have been found to be
ineffective:
1.) When placed under asphalt rubber-asphalt
concrete. This is due to the high
placement temperature of the RAC Type
G mix, which is close to the melting
temperature of the fabric.
2.) For providing added structural
strength when placed in combination
with DGAC.
3.) In the reduction of thermal cracking
of the new AC pavement overlay.
4 – 40 Cold Recycled Asphalt
Concrete Pavement
Assembly Bill ( AB 1306) encourages
State agencies to use more recycled
materials in road construction and
repairs. Caltrans Deputy Directive DD-
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 7
17 policy statement, effective November
11, 1993, directs the Department to
recycle asphalt concrete whenever
feasible. Consideration should be given
on every project to recycle asphalt
concrete ( AC) used in highway
construction, maintenance, and
rehabilitation projects utilizing the
Department’s priority hierarchy ( see
DD- 17). Public and employee health
and safety are not to be compromised by
recycling AC on any project. To be
economical on rehabilitation projects, a
minimum of 10,000 tons ( 9070 tonnes)
of AC material should be available for
the recycle process. In the future,
calculations using the then current price
of asphalt material may change the
quantity for the minimum tons to be
economical.
Since this design method uses two
procedures ( milling and replacement), it
can be considered appropriate to smooth
a rough pavement.
Candidates for cold recycling are
pavements whose asphalt content is
uniform. The existence of heavy crack-sealant,
numerous patches, open- graded
asphalt concrete, and heavy seal coats
make the new Cold Recycled Asphalt
Concrete ( CRAC) mix design
inconsistent. Mix properties are more
difficult to control. To avoid this
problem when it occurs and still use this
recycle option, a minimum of 0.08 ft ( 25
mm) should be milled off prior to the
cold recycling operation. Light crack
sealing ( less than 5 % of the pavement)
or a uniform single seal coat will not
influence the design sufficiently to
require removal.
Caltrans has established a minimum mill
depth of 0.15 ft ( 45 mm) for cold
recycling. Since existing pavement
thicknesses will have slight variations,
the cold recycling design should leave at
least the bottom 0.15 ft ( 45 mm) of the
existing AC pavement in place. This is
to insure the milling machine does not
loosen base material and possibly
contaminate the CRAC mix design.
Traffic constraints may make CRAC
impractical since traffic is not allowed
on the lane being recycled until the
process is completed and the recycled
material is compacted.
The recycling process consists of the
following:
1. Mill the existing AC pavement to the
designed depth.
2. Mix the milled material with an oil
or rejuvenating agent and leave in a
windrow.
3. The CRAC material is then spread
with a paving machine and
compacted.
The surface of the CRAC material has a
low resistance to abrasion. Therefore,
all CRAC material must be covered with
a minimum thickness of 0.15 ft ( 45 mm)
DGAC for a wearing surface after a
short period of time after the recycling
process.
When designing the CRAC for structural
adequacy, the Tolerable Deflection at the
Surface ( TDS) is always determined
using the thickness of the existing
pavement prior to milling. The
additional Gravel Equivalence ( GE)
required to reduce the measured
deflection to the tolerable level in the
cold recycling design is a combination
of:
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 8
• The GE determined from the basic
overlay calculations, and
• The GE required to replace the
material removed by the milling
process.
The analysis must first consider milling
down to no more than what Caltrans
calls the “ analytical depth”. *
Use the following definitions for CRAC
analysis:
Mill Depth = The depth of the milling in
feet.
D80 = 80th Percentile of the deflections at
the surface in inches, for a test section.
DM = The calculated Deflection at the
Milled depth in inches.
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
TDS = Tolerable Deflection at the
Surface in inches.
PRM = Percent Reduction in deflection
required at the Milled depth.
PRM = ( ) 100  

 
 −
DM
DM TDS
The percent reduction in deflection at the
milled depth is based on a research study
that determined deflections increase by
12% for each additional 0.10 ft ( 30 mm)
of milled depth. ( 7) Since it is not known
at what milled depth the 70% PRM level
* The “ analytical depth,” as defined by Caltrans,
is the milled depth at which the required Percent
Reduction in Deflection ( PRM) reaches 70%, or
the milled depth reaches 0.50 ft ( 150 mm),
whichever comes first.
or analytical depth will be reached, a
trial and error or iterative type of
calculation is required.
Using the thickness of the existing AC
pavement and the design TI, determine
the TDS from Table 1. The deflection at
the milled depth is found from the
equation:
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
The PRM is then found:
PRM = ( ) 100  

 
 −
DM
DM TDS
Utilizing the calculated PRM value as
percent reduction in deflection, go to
Table 2, Column A, to get the total GE
required to be placed on top of the
milled pavement surface. Using the total
GE requirement and subtracting the GE
of the CRAC thickness, ( CRAC
thickness times 1.5) the thickness of the
DGAC cap is determined. The Gf for
CRAC is 1.5 and for DGAC the Gf is
1.9.
GE of DGAC = ( Total GE required) –
( CRAC thickness)( 1.5)
Thickness of DGAC = GE of DGAC/ 1.9
If the milling goes below the analytical
depth, the analysis changes. Rather than
increasing the deflections, the analysis
assigns a Gf of 1.4 to the material below
the analytical depth.
Therefore, the additional GE that is
required to replace this lower portion of
the milled pavement is:
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 9
Additional GE = [( 1.4)( milled depth
below the analytical depth)]
This additional GE is added to the total
GE determined to be placed on top of the
milled pavement surface at the analytical
depth.
Finally, a determination is made to see if
the designed thicknesses of the CRAC
and DGAC are suitable. For CRAC to be
considered, it must be cost- effective.
Items to consider are:
• The increase in the profile grade
should be at least 0.10 ft ( 30 mm)
less than the increase from the basic
overlay design; otherwise a basic
overlay would be less costly; and
• The amount of CRAC material
should be about 10,000 tons ( 9070
tonnes) or more to be cost effective.
For CRAC design, it is recommended to
round up to get the CRAC and DGAC
thicknesses.
Example 4- 2: Determine the cold-recycled
thickness and the DGAC cap
thickness for rehabilitation.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
8.0 0.030 inch 0.55 foot AC
0.50 foot AB
1.00 foot AS
Solution 4- 2:
Recommendations to be considered:
Structural Adequacy:
A 0.30- ft DGAC overlay. Refer
to Example 3- 1. ( Rubber AC
alternatives are discussed in
Section 4- 20 of this manual.)
Reflective Cracking:
A 0.30- ft DGAC overlay. ( One-half
existing AC thickness.)
Ride Quality:
A 0.25- ft DGAC overlay placed
in two layers. ( Section 3- 30).
Use Table 1 to determine that the TDS is
0.017 inch
Calculation 4- 2:
Start with a minimum milling depth of
0.15 ft and find the deflection at the
milled depth:
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
DM = ( 0.030 inch)+[( 1.2/ ft)( 0.15
ft)( 0.030 inch)] = 0.035 inch
Determine the Percent Reduction in
Deflection at the Milled Depth ( PRM):
PRM = ( ) 100  

 
 −
DM
DM TDS
PRM = [( 0.035 inch – 0.017 inch)/ 0.035
inch]( 100)
PRM = 51.0% < 70 %, the analytical
depth.
Therefore, use PRM = 51%
From Table 2, Column A, the total GE
required is 0.66 ft.
GE of CRAC = ( 0.15 ft)( 1.5) = 0.22 ft
Determine the GE that the DGAC
overlay has to provide:
GE of DGAC = Total GE required – GE
of CRAC = 0.66 ft – 0.22 ft = 0.44 ft.
Thickness of DGAC = 0.44 ft/ 1.9
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 10
= 0.23 ft. Round up to 0.25 ft.
This is not acceptable since the DGAC
thickness saved from the basic overlay is
only ( 0.30 ft – 0.25 ft ) = 0.05 ft. This
should be at least 0.10 ft.
Try again.
Trial 2: Increase the milling depth to
0.20 ft and find the deflection at the
milled depth:
DM = ( 0.030 inch) + [( 1.2/ ft)( 0.20 ft)
( 0.030 inch)] = 0.037 inch
PRM = [( 0.037 inch – 0.017 inch)/ 0.037
inch ]( 100) = 54%
54% < 70%, the analytical depth.
Therefore, use PRM = 54%
From Table 2, Column A, the total GE
required is 0.72 ft.
GE of CRAC = ( 0.20 ft)( 1.5) = 0.30 ft
GE of DGAC = Total GE required – GE
of CRAC
= 0.72 ft – 0.30 ft = 0.42 ft
Thickness of DGAC = 0.42 ft/ 1.9 = 0.22
ft. Round up to 0.25 ft.
The results did not change for the
DGAC thickness saved from the basic
overlay. This should be at least 0.10 ft.
Try again.
Trial 3: Increase the milling depth to
0.25 ft and find the deflection at the
milled depth:
DM = ( 0.030 inch) + [( 1.2/ ft)( 0.25 ft)
( 0.030 inch)] = 0.039 inch
PRM = [( 0.039 inch – 0.017 inch)/ 0.039
inch]( 100) = 56.4%
56.4% < 70%, the analytical depth.
Therefore, use PRM = 56.4%
From Table 2, Column A, the total GE
required is 0.77 ft.
GE of CRAC = ( 0.25 ft)( 1.5) = 0.38 ft
GE of DGAC = Total GE required – GE
of CRAC = 0.77 ft – 0.38 ft = 0.39 ft
Thickness of DGAC = 0.39 ft/ 1.9 = 0.21
ft. Round up to 0.25 ft.
Again the results did not change for the
DGAC thickness saved from the basic
overlay. This should be at least 0.10 ft.
Try again.
Trial 4: Increase the milling depth to
0.30 ft and find the deflection at the
milled depth:
DM = ( 0.030 inch) + [( 1.2/ ft)( 0.30 ft )
( 0.030 inch)] = 0.041 inch
PRM = [( 0.041 inch – 0.017 inch)/ 0.041
inch]( 100) = 58.5%
58.5 < 70%, the analytical depth.
Therefore, use PRM = 58.5%
From Table 2, Column A, the total GE
required is 0.82 ft.
GE of CRAC = ( 0.30 ft )( 1.5) = 0.45 ft
GE of DGAC = Total GE required – GE
of CRAC = 0.82 ft – 0.45 ft = 0.37 ft
Thickness of DGAC = 0.37 ft/ 1.9 = 0.19
ft. Round up to 0.20 ft.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 11
Discussion 4- 2:
When compared to the basic overlay
design, CRAC saves 0.10 ft of virgin
DGAC and would also decrease the final
profile grade of the shoulder thus saving
shoulder- backing material.
Now that the first consideration has been
met, consider volume. For a project 10
miles long and pavement 24 feet wide
would this produce enough CRAC
material to be cost effective? Assuming
a compacted AC density of 145 pcf, the
milling tonnage is calculated as [( 10
miles)( 5280 ft/ mile) ( 24 ft)( 0.30 ft)( 145
lbs/ cu ft)]/ 2000 lbs/ ton = 27,562 tons.
This is greater than 10,000 tons, the
minimum required, and thus is
acceptable. By reducing the overlay by
0.10 ft, a saving of 9,187 tons of new
material or natural resources would be
accomplished. [ In this example, milling
did not go below the analytical depth – it
reached 58.5% compared to the
maximum of 70%, and the depth was
less than 150 mm ( 0.50 ft) of milling.]
( See Hot Recycled Asphalt Concrete
Pavement design for an example of an
analysis with milling below the
analytical depth.)
Cold recycling is, therefore, an
acceptable recommendation because it
decreases the final overlay profile grade
thus saving virgin DGAC and shoulder
backing, and it has over 10,000 tons of
recycled material making it cost
effective for this project to bring in the
specialized equipment.
Recommendation 4- 2: Cold recycle 0.30
ft ( 90 mm) of the existing pavement and
cap with 0.20 ft ( 60mm) of DGAC.
4 – 50 Hot Recycled Asphalt Concrete
Pavement
Assembly Bill ( AB 1306) encourages
State agencies to use more recycled
materials in road construction and
repairs. Caltrans Deputy Directive DD-
17 policy statement, effective November
11, 1993, directs the department to
recycle asphalt concrete ( AC) whenever
feasible. Consideration should be given
on every project to recycle AC used in
highway construction, maintenance, and
rehabilitation projects utilizing the
Department’s priority hierarchy ( see
DD- 17). Public and employee health
and safety are not to be compromised by
recycling AC on any project. At the
present time, to be economical on
rehabilitation projects, a minimum of
10,000 tons ( 9070 tonnes) of AC
material should be available for the
recycle process. In the future,
calculations using the then- current price
of asphalt material may change the
quantity for the minimum tons to be
economical for Hot Recycled Asphalt
Concrete ( HRAC).
Since this design method uses two
procedures ( milling and replacement), it
is one that can be considered appropriate
to smooth a rough pavement.
The hot recycling operation consists of
the following:
1. Mill the existing AC to obtain the
Reclaim Asphalt Pavement ( RAP).
2. Haul the RAP to an asphalt mixing
plant. *
* This is not hot- in- place recycling ( surface
recycling) which is a maintenance procedure.
Hot- in- place recycling material does not leave
the pavement lane site.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 12
3. Add the RAP and oil or rejuvenating
agent to the new DGAC mix to
obtain the recycled mix.
4. Haul the HRAC mix back to the
project to be spread with a paving
machine and then compacted.
5. To prevent damage, traffic should be
minimized or not allowed on the
milled surface of the lane being
recycled depending on the thickness
of AC pavement remaining after
milling ( must be at least 0.25 ft left
before allowing any traffic on the
lane).
Pavements that are candidates for hot
recycling are those with uniform asphalt
content. The existence of heavy crack-sealant,
numerous patches, open- graded
asphalt concrete, and heavy seal coats
make the new Hot Recycled Asphalt
Concrete ( HRAC) mix design
inconsistent and therefore more difficult
to control the mix properties. To avoid
this problem when it occurs and still use
this recycle option on projects, a
minimum of 0.08 ft ( 25 mm) should be
milled off and stockpiled for other uses
( e. g., shoulder backing) prior to the hot
recycling operation. Light crack sealing
( less than 5 % of the pavement) or a
uniform single seal coat will not
influence the design sufficiently to
require removal.
Caltrans has established a minimum mill
depth of 0.10 ft ( 30 mm) for hot
recycling. Since existing pavement
thicknesses will have slight variations
the hot recycling design should leave at
least the bottom 0.15 ft ( 45 mm) of the
existing AC pavement in- place. This is
to insure the milling machine does not
loosen base material and possibly
contaminate the HRAC mix design.
Milling down to a depth that leaves only
0.15 ft works only when traffic is not
allowed on the pavement prior too the
HRAC material being placed and
compacted. The thin remaining surface,
if opened to traffic, would cause
degradation of the pavement and affect
the design life of the new HRAC
material.
When designing the HRAC for structural
adequacy, the tolerable deflection ( TDS)
is always determined using the thickness
of the existing pavement. In a hot
recycling design, the additional GE
required to reduce the measured
deflection to the tolerable level is a
combination of:
• The GE required from the basic
overlay calculations, and
• The GE required to replace the
material removed by the milling
machine.
The percent reduction in deflection at the
milled depth is based on a research study
that determined that deflections increase
12% for each additional 0.10 ft ( 30 mm)
of milled depth ( 7).
Since it not known at what milled depth
the 70 % PRM level or the “ analytical
depth*” will be reached, this is a trial and
error or iterative type of calculation.
Use the following definitions for HRAC
analysis:
Mill Depth = The depth of the milling in
feet.
* The analytical depth, as defined by Caltrans, is
the depth the required Percent Reduction in
Deflection at the milled depth reaches 70%, or
the milled depth reaches 0.50 ft ( 150 mm),
whichever comes first. For discussion of deeper
milling depths see Remove and Replace.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 13
D80 = 80th Percentile of the deflections at
the surface in inches, for a test section.
DM = The calculated Deflection at the
Milled depth in inches.
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
TDS = Tolerable Deflection at the
Surface in inches.
PRM = Percent Reduction in deflection
required at the Milled depth.
PRM = ( ) 100  

 
 −
DM
DM TDS
Using the thickness of the existing AC
pavement and the design TI, determine
the TDS from Table 1. Calculate the
deflection at the milled depth from the
equation:
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
The PRM is then found:
PRM = ( ) 100  

 
 −
DM
DM TDS
Utilizing the calculated PRM value go to
Table 2, Column A, to get the total GE
required to be placed on top of the
milled pavement surface. The HRAC
thickness is found by dividing the GE by
the Gf of 1.9.
If the milling goes below the analytical
depth, the analysis changes. The
existing material below the analytical
depth is considered to be of questionable
structural integrity and hence assigned
the Gf of 1.4. The additional GE that is
required to replace the portion below
analytical depth is calculated by
multiplying the Gf of 1.4 by the milled
depth below the analytical depth. This is
added to the required GE to be placed on
top of the milled surface at the analytical
depth. The total HRAC thickness
required is found by dividing the sum of
the two GE’s by the Gf of 1.9.
Finally, a determination is made to see if
the designed thickness of the HRAC is
suitable. For HRAC to be considered, it
must be cost effective. Items to consider
are:
• The increase in the profile grade
should be at least 0.10 ft ( 30 mm)
less than the increase from the basic
overlay design; otherwise a basic
overlay would be less costly; and
• The amount of RAP should be about
10,000 tons ( 9070 tonnes) or more to
be cost effective.
Unlike cold recycled material, HRAC
pavement can be used as a surface
course without a DGAC cap. The Gf of
HRAC is the same as DGAC ( i. e., Gf =
1.9 ). Therefore, this analysis can also
be used for DGAC on milled pavement
and the reclaimed asphalt pavement
could be stockpiled for future use.
Example 4- 3: Determine the milling
depth and the hot- recycled thickness for
rehabilitation.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
11.0 0.031 inch 0.75 foot AC
0.50 foot AB
1.00 foot AS
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 14
Solution 4- 3:
Recommendations to be considered:
Structural Adequacy:
A 0.50- ft DGAC overlay. Refer
to Example 3- 1. ( Rubber AC
alternatives are discussed in
Section 4- 20 of this manual.)
Reflective Cracking:
A 0.35- ft DGAC overlay. ( One-half
existing AC thickness with a
maximum of 0.35 ft.)
Ride Quality:
A 0.25- ft DGAC overlay placed
in two layers. ( Section 3- 30).
Use Table 1 to determine that the TDS is
0.011 inch.
Calculation 4- 3: Start with a milling
depth of 0.15 ft and find the deflection at
the milled depth
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
DM = ( 0.031 inch) + [( 1.2/ ft)( 0.15
ft)( 0.031 inch)] = 0.037 inch
Determine the Percent Reduction in
Deflection at the Milled Depth ( PRM):
PRM = ( ) 100  

 
 −
DM
DM TDS
PRM = [( 0.037 inch – 0.011 inch)/ 0.037
inch]( 100)
PRM = 69.9% ≈ 70%, the analytical
depth.
Therefore, use PRM = 69.9%
From Table 2, Column A, the total GE
required is 1.06 ft.
Find the HRAC thickness:
GE/ Gf = 1.06 ft/ 1.9 = 0.56 ft. Round to
0.55 ft.
The increase in grade is ( 0.55 ft – 0.15
ft) = 0.40 ft. This is acceptable since the
reduction in profile grade from the basic
overlay is 0.10 ft. This would save 0.10
ft of virgin DGAC and would also
decrease the final grade of the shoulder
thus saving shoulder- backing material.
The quantity for milling 0.15 ft in two
lanes per mile is calculated as follows:
[( 1 mile)( 5280 ft/ mile)( 24 ft)( 0.15
ft)( 145 lbs/ cu ft)]/ 2000 lbs/ ton = 1,378
tons. The project should be long enough
( and/ or wide enough) to provide at least
10,000 tons for recycling. 10,000
tons/ 1,378 tons per mile = 7.3 miles.
The percentage RAP in the mix is ( 0.15
ft milled/ 0.55 ft HRAC thickness)( 100)
= 27%. In order to get more RAP and
use less virgin material, the milling
depth should be increased.
Since the analytical depth was nearly
reached ( 69.9%) at the milled depth of
0.15 ft, all milled and removed material
below that level will be considered to be
material with a Gf of 1.4.
Trial 2: Increase the milled depth to
0.25 ft ( 0.10 ft below the analytical
depth) to save more material.
Total thickness of the HRAC = [( GE
required at 0.15 ft milled depth) + ( GE
required due to additional milling)]/ 1.9.
HRAC = [( 1.06 ft) + ( 0.10 ft)( 1.4)]/ 1.9
= 0.63 ft of HRAC. Round to 0.65 ft.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 15
The percent of RAP is ( 0.25 ft/ 0.65
ft)( 100) = 38 %.
Trial 3: Increase the milled depth to
0.30 ft ( 0.15 ft below the analytical
depth) to save more material.
HRAC = [( 1.06 ft) + ( 0.15 ft)( 1.4)]/ 1.9
= 0.67 ft. Round to 0.65 ft.
The percent of RAP is ( 0.30 ft/ 0.65
ft)( 100) = 46 %.
Discussion 4- 3:
At this milling depth the RAP content is
46% and the increase in grade is 0.35 ft.
This will save 0.15 ft of new material
compared with the 0.50 ft DGAC
overlay needed by the basic overlay and
would also decrease the final grade of
the shoulder thus saving shoulder-backing
material.
Recommendation 4- 3: Mill 0.30 ft ( 90
mm) of the existing pavement and then
replace it with a total thickness of 0.65 ft
( 195 mm) of HRAC.
4 – 60 Remove and Replace
When it is not possible to maintain the
existing profile grade using the hot
recycled hot mix ( HRAC) design, the
remove- and- replace strategy can be
used. The Remove- and- Replace ( R& R),
sometimes called Mill and Fill, operation
consists of milling the entire AC
pavement and possibly into the base
material. ( When using several milling
passes, part of the AC may be used to
reclaim asphalt pavement for HRAC.)
The entire milled depth is then replaced
with DGAC or HRAC.
This design method may be less reliable
the deeper the milling is performed. A
study has shown that deflections will
increase an average of 12% for each
0.10- ft of pavement milled off ( based on
milling depths down to about 0.50 ft).( 7)
The greater the depth of milling the less
accurate the determination may be of the
calculated deflections.
R& R design from deflections is also less
reliable if a bulldozer or a scraper is used
to remove the material under the
pavement instead of a milling machine.
This method of removing material
disturbs the integrity of the in- place
material from which the deflections were
measured.
The alternative to the use of this design
is the R- value design method ( see HDM
Chapter 600).
When using the R& R method in
designing for structural adequacy, the
tolerable deflection ( TDS) is always
determined using the thickness of the
existing pavement.
The analysis used for R& R is similar to
the HRAC analysis ( Section 4- 50). First
consider milling down to what is called
the analytical depth. This is the depth
where the required Percent Reduction in
Deflection at the Milled depth ( PRM)
reaches 70% or to 0.50 ft ( 150 mm), or
to the bottom of the pavement,
whichever comes first. As discussed
above, the 70% PRM is based on an
increase in deflection of 12% for each
0.10- ft ( 30 mm) of milled pavement.
This is an iterative type of calculation
since it not known at what milling depth
the 70% level will be reached. Use the
following definitions for the R& R
analysis:
Mill Depth = The depth of milling in
feet.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 16
D80 = 80th Percentile of the deflections at
the surface in inches, for a test section.
DM = The calculated Deflection at the
Milled depth in inches.
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
TDS = Tolerable Deflection at the
Surface in inches.
PRM = Percent Reduction in deflection
required at the Milled depth.
PRM = ( ) 100  

 
 −
DM
DM TDS
Use the thickness of the existing AC
pavement and the design Traffic Index
( TI) in Table 1 to determine the
Tolerable Deflection at the Surface
( TDS). Then find the deflection at the
milled depth.
DM = ( ) ( )  

 

  

  

80 + D80
0.10
D 12%
ft
MillDepth
The percent reduction in deflection at the
milled depth ( PRM) is then found:
PRM = ( ) 100  

 
 −
DM
DM TDS
Utilizing this calculated PRM value go
to Table 2, Column A to get the GE
required to be placed on top of the
milled surface. When the milled depth
reaches the analytical depth, the analysis
changes. The GE for the material milled
out below the analytical depth is added
to the GE required at the analytical
depth. The GE for each layer is
calculated by:
GE = ( Gf)( thickness of the layer milled)
Commonly Used Gf for Rehabilitation
Asphalt Concrete 1.9
Hot Recycled
Asphalt Concrete 1.9
Cold Recycled
Asphalt Concrete 1.5
Treated Base 1.5
AC Below the
Analytical Depth 1.4
Aggregate Base 1.1
Aggregate
Subbase 1.0
Native Soil 0
The existing base material is considered
treated if it meets all of the following
conditions:
• Its depth is equal to or greater
than 0.35 ft ( 105 mm).
• The D80 is less than 0.015 inch.*
• It was portland cement concrete
( PCC), lean concrete base
( LCB), or Class A cement
treated base ( CTB- A) when first
installed.
The replacement DGAC thickness is
found by dividing the sum of the GE’s
by the Gf of the new DGAC. For the
* See discussion in Section 3 – 10.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 17
R& R design method, use the Gf for the
new DGAC commensurate with the TI
and AC thickness found in Table 608.4
of the Highway Design Manual
( HDM).*( 8) The total DGAC thickness
can be solved for each 0.05 ft ( 15 mm)
of material milled until the desired
profile is reached. Round the
replacement thickness to the nearest 0.05
ft.
Example 4- 4: Determine the milling
depth and the DGAC thickness for
rehabilitation to maintain the existing
profile grade.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
12.0 0.030 inch 0.75 foot AC
0.50 foot AB
0.83 foot AS
Solution 4- 4:
Recommendations to be considered:
Structural Adequacy:
Solve for a basic DGAC overlay.
Use Table 1 to find that the TDS is
0.009 inches.
PRD = [( 0.030 inches – 0.009
inches)/ 0.030 inches]( 100) = 70.0 %
Use Table 2, Column A, to
determine that 1.06 ft is the increase
in GE required to reduce the D80 to
the tolerable deflection level.
* For an AC thickness greater than 0.50 ft ( 150
mm), the Gf increases as the thickness increases;
see HDM Index 608.4 ( 8).
DGAC overlay thickness = ( 1.06
ft)/( 1.9) = 0.56 ft. Round to 0.55 ft.
Reflective Cracking:
A 0.35- ft DGAC overlay. ( One- half
existing AC thickness with a
maximum of 0.35 ft.)
Ride Quality:
A 0.25- ft DGAC overlay placed in
two layers. ( Section 3- 30).
Calculation 4- 4: Now provide a
rehabilitation strategy by the R& R
method that maintains the existing
profile grade.
In this example, the analytical depth of
70% was reached at the surface, so to
obtain the GE below the surface, all the
calculations will be multiplying the Gf
times the thickness of the layer milled.
These values will then be added to the
GE required at the surface.
Find the GE removed when milling from
the analytical depth ( the surface in this
example) down to the bottom of the
pavement: GE = ( 0.75 ft)( 1.4) = 1.05 ft.
This is added to the GE at the surface
and divided by the Gf of the new DGAC
to get the thickness required: ( 1.06 ft) +
( 1.05 ft) = 2.11 ft GE.
This is a trial and error problem since the
Gf that matches the new DGAC
thickness is unknown at this time.
Assume a Gf of 1.9. ( This is usually a
good starting point since it is about the
middle of Table 608.4).
GE/ Gf = ( 2.11 ft)/ 1.9 = 1.11 ft of
DGAC. Round to 1.10 ft.
From Table 608.4 of the HDM, the Gf =
2.09 for a thickness of 1.10 ft.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 18
Calculate the replacement thickness
using the Gf of 2.09 for the 1.10- ft. The
DGAC thickness is:
GE/ Gf = ( 2.11 ft)/( 2.09) = 1.01 ft.
Round to 1.00 ft of DGAC.
To match the thickness for which the Gf
is used, the answer appears to be
between 1.00 ft and 1.10 ft. Assuming a
thickness of 1.05, the Gf is equal to 2.05
( HDM, Table 608.4) and thus the DGAC
thickness needed is:
GE/ Gf = ( 2.11 ft)/( 2.05) = 1.03 ft.
Round to 1.05 ft of DGAC. ( This
matches the thickness for which the Gf
was used.)
When the milling extends to the bottom
of the pavement ( 0.75 ft ), the removed
material is replaced with 1.05 ft of
DGAC for an increase in the profile
grade of 0.30 ft. This is 0.25 ft lower in
profile grade than the basic overlay
design method provided. This would be
an acceptable solution except the
problem was to match the existing grade.
Therefore, find to what depth the milling
has to go to have no increase in profile
grade. Below the pavement the Gf for
the existing 0.50 ft of AB material is 1.1.
The additional GE to be replaced is 1.1
times the thickness of the AB layer
milled. This will be added to the GE at
the analytical depth ( at the surface in this
example) and the GE at the bottom of
the pavement; then the total is divided by
the Gf of the new DGAC.
Instead of trying each 0.05- ft of milling,
estimate to what depth the milling might
have to go. A quick calculation of the Gf
ratio times the increase in grade, when
milling stopped at the bottom of the
pavement, is one way that sometimes
works to estimate the needed additional
depth below the pavement.
[( 1.9)/( 1.1)]( 0.30 ft) = 0.52 ft. Round
to 0.50 ft.
This would be a total depth of 1.25 ft
( 0.75 ft + 0.50 ft).
Find the GE value of the AB removed to
the estimated depth ( 0.50 ft):
GE = ( 0.50 ft)( 1.1) = 0.55 ft.
This is added to the GE’s at the
analytical depth and bottom of the
pavement, and then divided by the Gf of
the DGAC to yield the required
thickness:
GE = ( 1.06 ft) + ( 1.05 ft) + ( 0.55 ft) =
2.66 ft.
For the estimated 1.25 ft depth, the Gf is
2.18 ( HDM, Table 608.4).
GE/ Gf = ( 2.66 ft)/ 2.18 = 1.22 ft of
DGAC. Round to 1.20 ft.
This is less than the 1.25- ft thickness
that was estimated; the depth of the AB
to be removed was too much. Therefore,
reduce the estimate for the milled depth
of the AB below the pavement. Try 0.45
ft into the AB, for a total thickness of
1.20 ft ( 0.75- ft pavement and 0.45 ft
base).
GE = ( 0.45 ft)( 1.1) = 0.50 ft. This is
added to the GE’s at the analytical depth
and bottom of the pavement, then
divided by the Gf of the 1.20 ft of DGAC
obtained from Table 608.4 of the HDM
( Gf = 2.15 ):
GE/ Gf = ( 1.06 ft + 1.05 ft + 0.50 ft)/ 2.15
= 1.21 ft
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 19
Round to 1.20 ft. ( This matches the
thickness for which the Gf was used).
Discussion 4- 4: Since this is quite deep
for the milling analysis the R& R method
may not be reliable*, check the R- value
design to see if it is close to the 1.20- ft
thickness. Only 0.05 ft of AB remains
above the aggregate subbase ( AS),
therefore use the R- value of the AS with
the TI10 of 12. ( The R- value is 50 for
Class 1 and 2, and 40 for Class 3 as per
HDM, Table 608.4). Assume an R-value
of 50. The equation to determine
the GE using the R- value is as follows:
GE required=( 0.0032)( TI)( 100– R- value)
GE = ( 0.0032)( 12)( 100- 50) = 1.92
For a full depth design, add a safety
factor to the GE of 0.10 ft to allow for
construction tolerances as per the HDM.
The GE required is then 2.02 ft.
Since we expect to be close to the same
depth determined by the deflection
method ( a good place to start), use a Gf
of 2.15 for the determined 1.20- ft of AC.
( Table 608.4 of the HDM).
GE/ Gf = ( 2.02 ft)/( 2.15) = 0.94 ft.
Round to 0.95 ft.
This is less than the estimated depth of
1.2 ft. The Gf needs to be lower to
increase the depth and balance the
equation. Try a Gf of 2.02 for a 1.00- ft
thickness.
* The analysis is based on deflections measured
on material in the structural layers as well as
several feet of original ground or fill. The
deeper these layers are disturbed or removed the
more the analysis is based on material that no
longer exists and the analysis becomes less
reliable. Engineering judgment needs to be
applied.
GE/ Gf = ( 2.02 ft)/( 2.02) = 1.00 ft.
The R- value design method determined
that the existing structural section should
be removed to a depth of 1.00 ft and
replaced with new DGAC. The Remove
and Replace design method produced a
thickness of 1.20 ft of new material.
Engineering judgment is needed as to
which depth to use. In this case the
deflection measurements gives the more
conservative answer and the engineer
working on the project may have other
data to support the use of the R& R
method.
Recommendation 4- 4: Mill 0.75 ft ( 225
mm) of the existing pavement and 0.45 ft
( 135 mm) of the AB material. Then
replace those layers with 1.20 ft ( 360
mm) of DGAC. This will maintain the
profile grade.
4 – 70 Asphalt Concrete Overlay
Placed on a Cushion Course
In this option, an aggregate base ( AB)
layer ( cushion course) is placed prior to
placing the DGAC pavement. It is used
• to raise the profile grade above a
flooded area; or
• when sections of newly constructed
added- on lanes, etc., produce grade
changes for existing roadways; or
• when the basic overlay design
produces a larger than desired dense
graded asphalt concrete ( DGAC)
overlay thickness ( too costly). As
this design method uses two
procedures it can also be used to
smooth a rough pavement.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 20
As a review, the basic overlay design
method is based on reducing the 80th
percentile deflection ( D80) at the surface
back to a tolerable level ( TDS).
Knowing the TI, 80th percentile
deflection, and the existing AC
pavement thickness, the gravel
equivalence ( GE) can be determined.
Using a gravel factor ( Gf ) of 1.9, the
thickness of the new AC overlay can be
calculated.
Please note that when determining the
mean and standard deviation of a test
section, do not omit the individual
measurements on isolated failed areas,
since patching failed areas will not be
recommended when designing an asphalt
concrete overlay placed on a cushion
course.
For this option, the design is based on
the same principle as the basic overlay
with two exceptions:
• the GE required, much like new
construction, is obtained with
combinations of AB and AC
pavement to reduce the D80 for the
new AC pavement, and
• the Gf varies with the TI and
thickness, again like new
construction.*
The DGAC gravel factor ( Gf )
commensurate with the TI and new AC
thickness found in Table 608.4 of the
HDM is used. However, no safety factor
of additional thickness for new
construction as described in the HDM
for the R- value design is to be applied.
As in new construction of highway
pavement, an AB layer should never be
* For an AC thickness greater than 0.50 ft ( 150
mm), the Gf increases as the thickness increases;
see HDM Index 608.4 ( 8).
placed less than 0.35 ft ( 105 mm) thick
and the DGAC surface should never be
placed less than 0.20 ft ( 60 mm) thick on
the AB.
Example 4- 5: Determine the AC and
AB thicknesses for a Cushion Course
design.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
8.0 0.056 inch 0.55 foot AC
0.50 foot AB
0.83 foot AS
Solution 4- 5:
Recommendations to be considered:
Reflective cracking and ride quality are
inherently provided for in this type of
design.
Structural Adequacy:
Since this design is much like new
construction design, the thickness of the
existing AC pavement does not enter
into the calculations for the aggregate
base and new AC thicknesses. To find
the minimum DGAC thickness required
over the AB, use the standard design
equation from the HDM:
GE = ( 0.0032)( TI)( 100- R)
Gf for AB is 1.1. The R- value for AB is
78. The GE required over the AB is:
GE = ( 0.0032)( 8)( 100- 78) = 0.56 ft.
AC = GE/ Gf. Use the Gf obtained from
Table 608.4 of the HDM ( Estimate what
the thickness will be and use that Gf ; or
as in this example, start with a Gf of
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 21
2.01; this Gf is good for all values of AC
thickness up to 0.50 ft and a TI of 8.0).
Therefore, this is an iterative calculation
process to get the solution. Round the
thickness to the nearest 0.05 ft.
AC = GE/ Gf = 0.56 ft/ 2.01 = 0.28 ft.
Round to 0.30 ft.
Actual GE provided by the DGAC =
( 0.30 ft)( 2.01) = 0.60 ft.
Using Table 1, the TDS of the new
DGAC thickness ( 0.30- ft) with a TI of
8.0 is 0.022 inch.
Calculate the percent reduction in
deflection required at the surface ( PRD)
of the existing pavement ( D80 = 0.056
inch) to reduce the TDS for the new
pavement to 0.022 inch.
PRD = ( 100)
80
80
AverageD
AverageD − TDS
PRD = [( 0.056 inch – 0.022 inch)/ 0.056
inch]( 100) = 60.7%.
The next step is to obtain the GE
required ( combination of AB and AC) to
reduce the deflection measured on the
existing surface ( 0.056 inch) to the
tolerable deflection level of the new AC
thickness ( 0.022 inch).
Using Table 2, Column B, determine the
total increase in GE required to reduce
D80 to the TDS for the new pavement
using a PRD of 60.7%.
GE ( Total required)= 1.10 ft.
Subtract the GE of the actual DGAC
thickness from the total GE required to
obtain the GE of the AB:
GE of AB = 1.10 ft – 0.60 ft = 0.50 ft.
Finally, divide by the Gf of the AB and
round to the nearest 0.05 ft:
AB = GE/ Gf = 0.50 ft/ 1.1 = 0.45 ft.
Use 0.45 ft of AB.
Recommendation 4- 5: Use 0.30 ft ( 90
mm) of DGAC over 0.45 ft ( 135 mm) of
AB
Example 4- 6: What would be the AB
thickness if the DGAC thickness for the
previous example were increased to 0.55
ft?
The Gf varies for AC thicknesses greater
than 0.50 ft. From HDM Table 608.4,
for a TI of 8.0 and 0.55 ft of AC, the
GE is 1.12 ft.
Tolerable deflection level of the new
pavement from Table 1, is 0.017 inch,
therefore
PRD = [( 0.056 inch – 0.017 inch)/ 0.056
inch ] ( 100) = 69.6 %.
Using Table 2, Column B, determine the
total increase in GE required to reduce
D80 to the TDS for the new pavement:
GE = 1.41 ft.
Subtract the GE of the actual DGAC
thickness ( 1.12 ft ) from the total GE
required to get the GE of the AB:
GE of AB = 1.41 ft – 1.12 ft = 0.29 ft.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 22
Finally, divide the GE of AB by its Gf
and round to the nearest 0.05- ft to
calculate the AB thickness required:
GE/ Gf = 0.29 ft/ 1.1 = 0.26 ft. Round to
0.25 ft.
Use the minimum thickness, 0.35 ft for
AB.
Recommendation 4- 6: Use 0.55 ft ( 165
mm) of DGAC over 0.35 ft ( 105 mm) of
AB
4 – 80 Cushion Course Design with
Drainage Layer
In this option, an AB layer ( cushion
course) is placed prior to placing a
drainage layer and DGAC pavement. It
is similar in design to an “ Asphalt
Concrete Overlay Placed on a Cushion
Course” described in Section 4- 70. The
Gravel Equivalence ( GE) for the added
layer of the Asphalt Treated Permeable
Base ( ATPB) is subtracted from the total
GE required. [ Note that a drainage layer
requires positive outflow and is
discussed in Highway Design Manual
( HDM), Chapter 600, Topic 606]. The
thickness of the ATPB is 0.25 ft ( 75mm)
unless a unique combination of
conditions exists. See HDM, Section
606.2.
The Gf for ATPB is 1.4 as obtained from
the HDM, Table 608.4. The GE that the
0.25- ft ( 75- m) ATPB layer contributes to
the total required thickness is:
GE = ( Gf )( AB thickness) = ( 1.4)( 0.25 ft)
GE = 0.35 ft.
Since an ATPB drainage layer has an
indeterminate R- value, the minimum
thickness of the DGAC over the ATPB
is based on the equation below. The GE
of the AC is 0.4 of the total GE required
over a 50 R- value material [ see HDM
608.4 ( 4) ( b)]. The minimum thickness
of DGAC cover over the ATPB should
never be less than 0.20 ft ( 60 mm).
GE over ATPB = [( 0.4)( GE required
over a 50 R- value material)]
GE = [( 0.4)( 0.0032)( TI)( 100- R)]
Example 4- 7: Use the same data from
the example problem solved in the
“ Asphalt Concrete Overlay Placed on A
Cushion Course” design, Section 4- 70.
Determine the AC, ATPB and AB
thicknesses for a Cushion Course design
with drainage layer.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
8.0 0.056 inch 0.55 foot AC
0.50 foot AB
0.83 foot AS
Solution 4- 7:
Recommendations to be considered:
Reflective cracking and ride quality are
inherently provided for in this type of
design.
Structural Adequacy:
Find the minimum DGAC thickness
over the ATPB:
The GE over the ATPB is
[( 0.4)( 0.0032)( 8)( 100- 50)] = 0.51 ft.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 23
From the HDM, Table 608.4, the Gf for
any thickness of DGAC 0.50 ft or less is
2.01. The minimum thickness of cover
for the ATPB is:
GE/ Gf = ( 0.51 ft)/( 2.01) = 0.25 ft.
Use 0.25 ft.
Calculate the actual GE provided by the
0.25- ft of DGAC:
GE = ( 2.01)( 0.25 ft) = 0.50 ft.
Use the thickness of the new AC
pavement and the design Traffic Index
( TI) in Table 1 to determine that the
Tolerable Deflection at the Surface
( TDS) is 0.024 inch.
Calculate the percent reduction in
deflection required at the surface ( PRD)
of the existing pavement ( D80 = 0.056
inch), to reduce the TDS for the new
pavement to 0.024 inch:
PRD = ( 100)
80
80
AverageD
AverageD − TDS
PRD = [( 0.056 inch – 0.024 inch)/ 0.056
inch] 100 = 57.1 %
Utilizing the calculated PRD value, go to
Table 2, Column B to determine the
increase in GE required to reduce the
D80 to the TDS for the new pavement:
GE ( Total Required) = 0.98 ft.
Subtract the GE of the actual DGAC
thickness ( 0.50 ft) and the GE of the
ATPB ( 0.35 ft) from the total GE
required to get the GE of the AB.
GE ( Total Required) = ( GE of DGAC) +
( GE of ATPB) + ( GE of AB)
GE of AB = ( 0.98 ft)–( 0.50 ft) – ( 0.35 ft)
GE = 0.13 ft.
AB Thickness = GE/ Gf = 0.13 ft/ 1.1
AB = 0.11 ft. Round to 0.10 ft.
Use the minimum thickness, 0.35 ft for
AB.
Recommendation 4- 7 Use 0.25 ft ( 75
mm ) of DGAC, over 0.25 ft ( 75 mm ) of
ATPB, over 0.35 ft ( 105 mm ) of AB.
4 – 90 Asphalt Concrete Overlay with
Drainage Layer
Determination and discussion of the
need for a drainage layer can be found in
the California Highway Design Manual
( HDM), in Chapter 600, Topic 606.
Placement and design considerations
such as a positive outflow requirement
for a drainage layer are also found in the
HDM. This strategy can also be used to
smooth rough pavement as well as
provide the needed drainage since it
utilizes multiple layers.
The AC overlay thickness portion of this
strategy is determined using the design
method for a basic overlay, with the
Gravel Equivalence ( GE) of the Asphalt
Treated Permeable Base ( ATPB) layer
subtracted from the total GE required.
The thickness of the ATPB is 0.25 ft ( 75
mm) unless unique combinations of
conditions were to exist. [ See Highway
Design Manual ( HDM), Chapter 600,
Topic 606]. The standard layer of 0.25
ft ( 75 mm) will generally provide greater
drainage capacity than is needed under
AC pavements. Therefore, the standard
thickness generally provides sufficient
drainage and provides an allowance to
compensate for construction tolerances.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 24
Calculate the GE that the ATPB ( Gf is
1.4) contributes to the total required
thickness:
GE = ( 1.4)( 0.25 ft) = 0.35 ft.
Since an ATPB drainage layer has an
indeterminate R- value, the minimum
thickness of the DGAC over the ATPB
is based on the equation below. The GE
of the AC is 0.4 of the total GE required
over a 50 R- value material [ HDM 608.4
( 4) ( b)]. The minimum thickness of
DGAC over the ATPB should never be
less than 0.20 ft ( 60 mm).
GE over ATPB = [( 0.4)( GE required
over a 50 R- value material)].
GE = [( 0.4)( 0.0032)( TI)( 100- R)]
Example 4- 8: Determine the AC and
ATPB thicknesses for an existing AC
pavement.
Ten- Year
TI
80th Percentile
Deflection
Existing Structural
Section
10.0 0.030 inch 0.55 foot AC
0.50 foot AB
1.00 foot AS
Calculations 4- 8:
Check for overlay thickness required
for structural adequacy.
Step 1:
Obtain tolerable deflection at the
surface ( TDS).
Use Table 1:
AC = 0.55 ft and TI = 10.0
TDS = 0.012 inch
Step 2:
Compare average D80 to TDS.
0.030 > 0.012
Stem 3:
Calculate Percent Reduction in
Deflection required.
( 100) 60%
0.030
012 . 0 030 . 0 =  

 
 −
Step 4:
Determine Gravel Equivalence
( GE) required for deflection
reduction.
Use Table 2; Column A
GE = 0.85 ft
Step 5:
Find the minimum DGAC
thickness over the ATPB:
GE over ATPB drainage layer is =
( 0.4)( 0.0032)( 10)( 100 – 50) = 0.64 ft.
The DGAC thickness using a Gf = 1.9 is:
AC = GE/ Gf = 0.64 ft/ 1.9 = 0.34 ft.
Round to 0.35 ft.
Step 6:
Find the DGAC thickness
required to reduce the 80th
percentile deflection down to the
tolerable level. ( Use the standard
ATPB layer thickness of 0.25 ft.)
The GE that the 0.25- ft ATPB
layer provides is:
GE = ( ATPB thickness)( Gf )
GE = ( 0.25)( 1.4) = 0.35 ft
GE of DGAC = Total GE
required - GE of ATPB
GE = 0.85 ft – 0.35 ft = 0.50 ft
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
4- 25
Thickness of DGAC = GE/ Gf
DGAC = 0.50 ft/ 1.9 = 0.26 ft.
Round to 0.25 ft.
This is less than the minimum
DGAC thickness over the ATPB
layer.
Use 0.35 ft DGAC over 0.25 ft ATPB
for structural adequacy.
Check overlay thickness required for
reflective crack retardation.
To retard reflective cracks entering the
new overlay from the pavement below
choose a thickness for the new overlay at
least one- half the thickness of the
existing AC pavement being overlaid
( up to a maximum of 0.35 ft ( 105 mm)
for an underlying aggregate base).
Determine half of the existing pavement
thickness:
overlay =
2
0.55 = 0.275 Round
to 0.30 ft.
Check overlay thickness required for
smoothness.
The ride quality is improved by adding a
minimum 0.25- ft DGAC overlay placed
in two layers.
Discussion 4- 8:
• Reflective cracking requirement is
less than the 0.35- ft DGAC thickness
plus the 0.25- ft layer of ATPB.
• The ride quality is going to be
improved due to the two layers being
placed.
Therefore, structural adequacy governs
the overlay design thickness.
Recommendation 4- 8: Place 0.35 ft
( 105 mm ) of DGAC over 0.25 ft ( 75
mm) of ATPB.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
5- 1
CHAPTER 5
APPENDIX
5 – 10 Guidelines for Involving
Moisture and Temperature in Flexible
Pavement Rehabilitation
Moisture and temperature affect the
strength of the structural section and is
reflected in the measured deflections.
Pavement deflections increase with an
increase in the amount of moisture in the
underlying materials and an increase in
the temperature of the pavement at the
time of testing.
A saturated structural section and
subgrade along with a hot summer day
would be the extreme condition; the
structural section would be in its weakest
condition and produce the highest
deflection. Fortunately, this
environment is not the norm in
California. The hot season is normally
the dry time of the year. The most
favorable time to measure deflections
( designing for the worst case) is in the
spring of the year; the moisture content
of the basement soils will be at or near
their highest values and will affect the
deflections more than the moderate
temperature.
Presently, the magnitude of highways
needing rehabilitation makes deflection
measurements a year- round endeavor.
Judgment is required when considering
the seasonal variation of temperature,
moisture content, and test date for areas
throughout the state. With the large
variation of elevation in California,
“ spring” comes at different times of the
year. Fortunately, this allows a large
window of time to schedule deflections
throughout the state. Deflections
measured in the late summer or early fall
in the valley and desert areas, may be
influenced primarily by the temperature
and little by moisture content.
At the present time, Caltrans provides no
correction factor for temperature or
moisture content. Since higher
pavement temperatures produce higher
deflections, using the actual measured
deflection when the average pavement
temperature is 70 º F ( 21 º C) or more will
somewhat compensate for the lower
moisture content.
Deflections should not be measured
when the pavement surface temperature
is 45 º F ( 7 º C) or lower. The pavement
temperature above which deflection
measurements should not be made ( the
maximum pavement temperature at the
time of testing) will vary depending on
the weight of the deflection apparatus
being used. The California
Deflectometer with the Benkelman
Beam, and Falling Weight
Deflectometer, due to their higher
weight, should not be used to measure
deflections when the pavement surface
temperature is 130 º F ( 54 º C) or higher.
The lighter weight of the Dynaflect can
be used at any pavement temperature
above 45o F ( 7o C).
If other engineers or agencies elect to
use this Caltrans manual, the decision
and the method to correct for moisture
content and/ or temperature is left to their
discretion.
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
5- 2
5 – 20 Identifying and Recording
Distress
Examples of cracks associated with
asphalt concrete pavement and the
reason for the cracks are discussed
below. This may aid the engineer in the
design process.
1) Alligator Cracking ( Photos 7– 8):
Alligator cracking in the wheel path is a
load- associated, fatigue type of failure
for asphalt concrete. At these locations,
the evaluated pavement deflections will
almost always exceed the tolerable
values indicating that rehabilitation is
needed to restore structural adequacy.
Water should be prevented from entering
the structural section in this area
especially, due to the many cracks in a
small area that will develop into a
localized failure. As water enters the
structural section through the surface
cracks, pumping of fine material from
the roadbed and rutting often follow.
Sealing the surface cracks as soon as
they first appear will decrease the rate of
deterioration of the structural section.
2) Longitudinal Cracking in the
Wheel Path ( Photos 9 – 10):
A longitudinal crack in the wheel path is
considered a load- associated crack. The
cracking starts at the bottom of the
asphalt concrete pavement where tensile
stress and strain is highest under the
wheel load. The cracks propagate to the
surface initially as one or more
disconnected, parallel cracks. After
repeated traffic loading the cracks
connect, forming many sided, sharp-angled
pieces that develop a pattern
resembling chicken wire or the skin of
an alligator.
3) Longitudinal and Transverse
Cracking ( Photos 11 – 12):
These types of cracks are primarily
caused by shrinkage of the pavement
surface due to low temperature or
asphalt hardening, or the result of
reflective cracks from the underlying
pavement or base. If the roadway in
Photos 11 and 12 were structurally
adequate with good riding qualities,
rehabilitation may not be warranted.
Surface cracks should be sealed or a seal
coat placed to prevent water from
damaging the structural section.
4) Shrinkage and Thermal Cracking
( Photos 13 – 14):
Shrinkage and thermal cracking are not
load- associated type failures; but traffic
loads can increase the severity of the
cracks. Age hardening, overheated
mixes, insufficient asphalt content, and
normal thermal conditions are some of
the chief causes of shrinkage and
thermal cracking.
One of the prime objectives in moderate
to high rainfall areas is to seal surface
cracks and maintain the seal to prevent
water from entering the structural
section and causing accelerated roadbed
deterioration. A seal coat can be an
effective treatment in slowing the
deterioration due to moisture intrusion.
In areas of low rainfall, structural section
deterioration due to moisture entering
the roadbed may not be a major problem.
It is reasonable to delay a ten- year
rehabilitation project, or at the most just
place a seal coat, where traffic loads are
light; rainfall is low; pavement ride
quality is acceptable; and pavement
deflections indicate good structural
adequacy. A seal coat can prolong the
service life for several years without any
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
5- 3
corrective treatment for such a project.
However, a five- year design for the
Capital Preventative Maintenance
Program should not be delayed.
5) Severe Block Cracking ( See
Photo 15):
Block cracking is generally not load-associated
and usually divides the
pavement into approximately equal size
polygons or rectangular pieces. It is
mainly caused by hardening and/ or
shrinkage of the asphalt and daily
temperature cycling. However, severe
block cracking, where the size of the
polygon is approximately one or two
feet, is usually the result of a structural
failure of the pavement when the asphalt
concrete is placed over treated bases
such as cement treated base ( CTB), lime
treated base ( LTB), and lean concrete
base ( LCB). It can also occur when an
asphalt concrete ( AC) overlay has been
placed over old portland cement
concrete ( PCC) pavement. If the area is
localized, the pavement and base should
be repaired. If the area is extensive, the
rehabilitation design should be sufficient
to remedy this type of failure. Block
cracks are often greater than ¼ inch ( 5
mm) wide. Pavement deflection analysis
on treated bases, which generally
produce low deflections, may not always
provide an adequate overlay thickness
designed for structural adequacy to
minimize reflective cracking.
Experience has shown that a minimum
0.35- ft ( 105- mm) AC overlay is required
when severe block cracking exists in the
AC over treated bases or AC over PCC.
6) Settlement cracking:
These types of cracks are generally
nearly longitudinal or crescent- shaped
and are not load- associated type failures;
but traffic loads can increase the severity
of the cracks. These are due to localized
vertical displacement of the pavement
structural section due to slippage of a fill
or consolidation of the underlying
foundation material. One of the prime
objectives in moderate to high rainfall
areas is to seal these cracks and maintain
the seal to prevent water from entering
the structural section and accelerating
the displacement.
7) Localized Failures ( Photo 16):
Assuming the fatigue failure shown in
Photo 16 is not typical of the entire
project and is obviously an isolated
problem, it would be recommended that
this localized failure be replaced. Once
replaced, the rehabilitation of the entire
roadway can be based on the deflection
levels and conditions of the remaining
pavement. If a thin overlay were to be
placed over the section of roadway in
Photo 16 without performing repairs, the
surface cracks would probably recur in
the new overlay in less than a year.
Cracks should be recorded by name
width, and extent ( using percentages)
during the field survey.
Record whether crack widths are
hairline, ± 1/ 8 inch ( 3 mm), ± 1/ 4 inch ( 5
mm), or greater than ½ inch ( 10 mm).
Record the extent of cracking as follows
( showing the approximate percentages,
using continuous in both wheel paths as
being 100%):
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
5- 4
none or minimal – ( 0% to 5%);
isolated – ( 5% to 10%);
occasional – ( 10% to 15%);
intermittent – ( 15% to 50%);
nearly continuous – ( 50% to 85%);
continuous -- ( 85% to 100%).
Also, the field survey should describe
other distress such as the following:
Bleeding. Excess asphalt appears on the
surface of the pavement, usually in the
wheel paths. This should be kept from
migrating up through the new overlay by
milling to a satisfactory depth to remove
the saturated AC layer.
Minor – Surface looks slightly damp.
Major – Surface appears to be only
asphalt with little to no aggregate
showing in the surface mix.
Corrugations. Transverse undulations
appear at regular intervals due to the
unstable surface course caused by stop-and-
go traffic. Corrugations are often
associated with shoving and/ or
delamination. Note the size of the area.
To repair, mill deep enough to remove
the corrugated layer.
Light – Caused some vibration of the
vehicle, which creates no discomfort.
Medium – Causes significant vibration
of the vehicle that creates some
discomfort.
High – Causes excessive vibration of the
vehicle that creates substantial
discomfort and/ or vehicle damage
requiring a reduction in speed.
Delamination. Debonding of the surface
course from the underlying AC layer is
evidenced by shallow potholes, shoving,
or from the pavement cores where the
layers separate easily. The cause may be
insufficient tack coat at construction.
Record the number and locations. To
repair, mill deep enough to remove the
delaminating layers.
Patches. Asphalt concrete can been
added to distressed pavement in several
ways. A patch can be applied to the
surface as an overlay or placed within
the pavement after the distress had been
removed. Record the location of the
patches, whether they are in the wheel
path, half lane, or the entire lane; the
type of patches such as pothole, overlay,
inlaid, or grader patches; and the size of
patch such as spot ( hand placed), short
[ up to 100 ft ( 30 m)], and long [ greater
than 100 ft ( 30 m) ].
Potholes. Holes in the pavement
generally started when small parts of an
alligator- cracked area are dislodged by
traffic together with excessive water.
Record the depth of the pothole if it is
possible. Patch the potholes prior to
rehabilitation.
Small – Less than 1.0- ft ( 0.30- m) square.
Medium – Between 1.0- ft ( 0.30- m)
and 3.0- ft ( 0.91- m) square.
Large – Greater than 3.0- ft ( 0.91- m)
square.
Pumping. The ejection of foundation
material through cracks in the pavement
generally leaves the material as visible
residue on the surface. Record the
number of occasions or accumulated
length of the pumping. Rehab should
Flexible Pavement Rehabilitation Manual June 2001
______________________________________________________________________________________
5- 5
include limiting water intrusion into the
base.
Light – Water pumping is observed but
no fines ( or only a very small amount)
can be seen on the surface of the
pavement.
Medium – Some material can be
observed.
High – A significant amount of pumped
material exists on the surface near the
cracks.
Raveling. This is a progressive
disintegration of the asphalt concrete
surface downward by the dislodgment of
aggregate particles and binder. This
could be due to significant hardening of
the asphalt binder ( weathering) and
wo