90

TRANSPORTATION RESEARCH RECORD 1353

NCHRP Asphalt-Aggregate Mixture

Analysis System
HAROLD
}AMES

A.

L.

VON QUINTUS, CHUCK

s.

HUGHES, AND

ScHEROCMAN

A summary of the procedures for an asphalt-aggregate mixture

analysis system (AAMAS) developed under NCHRP Project 96(1) is provided . Most of the information is documented and
reported in NCHRP Report 338. The implementation process of
the AAMAS procedure is reviewed, and some of the problems
that can be incurred regarding the field control of asphalt concrete
mixtures being designed by the AAMAS process are discussed .

The asphalt-aggregate mixture analysis system (AAMAS) research project, NCHRP Project 9-6(1), was initiated because
AASHTO realized the importance of tying mixture design to
structural design and pavement performance variables (J).
Project 9-6(1) was completed in three phases. The first phase
(completed in October 1986) was concerned with evaluating
the feasibility for the development of an AAMAS. Phase I
identified the primary forms of pavement distress (associated
with both load and environment), evaluated current testing
and mixture design procedures, and identified new or modified laboratory procedures to be considered in the development of the AAMAS. Items that the Phase I concept emphasized included mixture preparation, conditioning, testing,
and analyzing asphalt concrete specimens to duplicate field
conditions. Tests to measure the engineering properties of
asphalt concrete mixtures for estimating pavement performance were also included and discussed .
Phases II and III were concerned with developing procedures for the AAMAS concepts and tying structural design
to mixture design. This project emphasized compatibility between mixture design and structural design, including the
AASHTO design manual. Phase II (completed in February
1989) included the initial development work, and Phase III
included follow-up field studies and conversion of the
AAMAS into a mixture design procedure . Phase III was completed in M:iy 1990, and the final report for this project was
published in March 1991. In summary, Project 9-6(1) resulted
in the development of an AAMAS for evaluating dense-graded
asphalt concrete mixtures proposed for use primarily on highvolume roadways and in a mixture design procedure based
on performance-related criteria. These criteria are compatible
with the recommendations from NCHRP Project 1-26 (2) .

The fourth phase of this evolutionary process was initiated

through a series of four 2-day workshops sponsored by FHW A.
The purpose of these workshops was to review the current
procedures and obtain input and information from state highway agencies (SHAs) on problems with the use of AAMAS
to design and evaluate asphalt concrete mixtures in their specific laboratories (3). These workshops have been completed.

AAMAS OVERVIEW
Mixture Design and Evaluation
AAMAS consists of three basic laboratory steps . The first step
is simply the initial mixture design phase, \Vhich is accomplished with current mixture design procedures or with the
procedure based on the AAMAS concept (i .e ., performancerelated criteria) . The mixture design procedure using the
AAMAS concept is included in Part I of NCHRP Report 338
(J) . An agency can use either the AAMAS approach or its
own current procedure to determine the design asphalt content and job mix formula . The performance-related mixture
design procedure using the AAMAS approach is a scaleddown version of AAMAS; it was formulated considering implementation and production factors in SHA laboratories.
Once an initial mixture design has been completed, these
materials are mixed, compacted, and conditioned in the second step. This step includes age-hardening simulations (both
for production and for the environment) , moisture conditioning, and traffic densification. This second step is the mixture compaction and conditioning phase.
After the materials have been mixed, compacted, and conditioned, the specimens are tested in the third step to measure
critical mixture properties. This third step provides the dat:i
that can be integrated into pavement design and analysis models
to predict pavement performance . This third step is the mixture evaluation phase; it is compatible with results from NCHRP
Project 1-26 (2). The mixture evaluation phase includes the
laboratory testing and performance evaluations.

Procedural Manual
H .. L. Von Quintus, Brent Rau.hut Engineering, Inc., 8240 MoPac,
Suite 220, Austin , Tex . 78759. . S. Hughe , Virginia Transportation
Research Counci l, P . 0. Box 3817, University Station. Charlotte.~
ville, Va. 22903. J. A. Scherocman, 11205 Brookbridge Drive Cincinnati , Ohio 45249.
'

These three steps are integrated in the procedural manual

for mixture design and evaluation . The procedural manual
(1, Part I) is divided into four sections. Figure 1 shows the
current AAMAS procedure in flow-chart form, identifying

Von Quintus et al.

91

Obtin 119terlal ~t
for llixture 0.1.,..

SECTIONS 1, 2

Concb:t Initial llixture

DeBilJl in Accordlnce With
119rshall, Hwm or other P~
Detenifne the Asphalt Cont9rlt and
Additive, if Required, Gr.ilrv Selected

Prepare Bit..ll'IOU9 lllxture

SECTION 3

AASllTO T 167

Initial OVen Heatirv of Bulk

llixture - Three Hours a 275 F

~t 6 ~iaxlal ~ion
Test Speci.,.. to Air Void Level
After Construction (5 to BX>

CcJlil*:t 18 Di-tral Test

Specl.,.. to .,. Air Void Level
After Construction (5 to llX>
Gro14> Test Speci.,..
Based on Air Voids

Traffic D-iflcation of
6 Speci.,.., D-ify to
Refusal at 140 F.
lh:ondi tioned
9 Specimns

T~ature

lloisture Condition
3 Speci-

Condition
(Accelerated Agirv>
6 Speci-

Gnq> Test Speci.,..

Based on Air Voids

Initial Auirv
Test Speci_,,.
at 1D4 F

2 Days a 140 F

llodified
MSHTO T28J

Test Speci_..
at 41, 77, 1D4 F

Final Asllrv
5 Days a 225 F

Test Specimns
at 41 F
lleasure Resil io:nt llcdllus,

Test Speci.ns
at 41 F

lleesure Res il i o:nt

llcdllua- and Static
Creep llcdllus

Indirect T-fle Strength

and Failure Strains

~onfined
~esaive

lleasure

Fatigue and Tlletwml Cr.:klrv Predictions

lloisture D-ae I. Disintegration Evah.mtions

Strervth

lleasUre Resili ent

llcdllus and Static
Creep lloWlus

Ruttlrv Predictions

to Structural
Design Re!JJl,._,ts

~re

SECTION 4

FIGURE 1 Flow Chart for AAMAS.

the four sections; Table 1 summarizes the approximate time

required for the laboratory compaction, conditioning, and
testing of asphalt concrete mixtures. Section I provides criteria
and values recommended for selecting the mixture components, and Section II presents the procedures used to design
dense-graded asphalt concrete mixtures. Figure 2 shows the
mixture design procedure (including Sections I and II) in flowchart form. Section III, the mixture analysis section, includes
procedures for preparing, conditioning, and testing specimens
for measuring properties required for structural design and
evaluation. Section IV, the mixture performance evaluation,

discusses mechanistic-empirical models used to evaluate asphalt concrete pavements.

PARAMETERS AND TESTS INCLUDED IN

AAMAS
Pavement Distress
Distresses selected for incorporation into AAMAS include
rutting, fatigue cracking, low-temperature cracking, and

TRANSPORTATION RESEARCH RECORD 1353

92

TABLE I Summary or Approximate Time Required for Laboratory Compaction, Conditioning, and Testing or Asphalt Concrete Mixtures
Using AAMAS

Time In Days
Laboratory Steps

10

11

1. Prepare & Mix Materials

2. Initial Heat Conditioning

of Loose Mix

3. Specimen Compaction -

u- - ....

Moisture Conditioned
Temperature Conditioned
Traffic Densified
4. Measure Air Voids & Sort
Into Subsets

ll
3

6
6

?4

5. Moisture Condition Samples

6. Heat Conditioning

7. Traffic Denslfication
8. Test Unconditioned Specimens

6
3 41F

3 77F

3 ra> 104F

9. Test Heat Conditioned

Specimens

10. Test Moisture Conditioned

Specimens
11. Test Traffic Denslfied
Specified

~
0

umt>-cua In b1ock1 rapresen t Iha numbor or .apoclmons and/or Inst rerope returo. Tho touu l mo uamo to complo te tho entlro AAMA~ process It
are In relallon to the time needed 10 run the Marsh all and Hveem mix design methods.

moisture damage. Secondary consideration is given to raveling or disintegration and loss of skid resistance.

Mixture Tests

To evaluate in the laboratory how asphalt concrete mixtures

will perform under traffic and the environment, a review and
a study of various test procedures were conducted. To design
mixes for preventing the aforementioned distresses, it is necessary to use a test that measures the engineering properties
and characteristics of the asphalt concrete mixture that are
related to the distress or performance measure. Tests were
selected on the basis of simplicity, efficiency, reliability, reproducibility, and sensitivity of mixture variables. Special consideration was given to sample size. For AAMAS to be useful
and applicable for a range of mixtures, the test procedures
and equipment must be capable of preparing and testing different size specimens that are compatible with the aggregates
used in the mix.
Five tests are used as tools for mixture evaluation in
AAMAS. These tests are the static cylindrical (unconfined
compression) creep and recovery test, the diametral resilient
modulus test, the indirect tensile strength test, the indirect
tensile creep and recovery test, and the gyratory shear test.

6 ra> 104F

los.t th an 2 woekt. 1no umet 1nown

The compressive strength of the mix is also measured in accordance with the creep and recovery compressive test.
The AAMAS program requires a combination of laboratory tests and conditioning procedures to evaluate the behavior and performance characteristics of asphalt concrete
mixtures. All factors considered, tensile strain at failure, gyratory shear strength, and creep are the properties most useful
for evaluating and comparing different mixtures. Resilient
modulus is required, but only because of its incorporation
into the AASHTO design guide. Thus, tensile strain at failure,
creep, resilient modulus, and gyratory shear strength are used
to ensure that the mixture, as placed, will satisfy the structural
design requirements.

Initial Mixture Design Optimization Guidelines

Guidelines are provided for selecting an aggregate blend and

selection of an initial asphalt content for optimizing the mixture's performance on the basis of predictions of fatigue cracking , rutting, and thermal cracking. The program ASPHALT
( 4) was found to be a good tool for selecting the "seed" asphalt
content in mixture design and theoretically determining the
relationship between asphalt content and air void. In addition,
correlations were performed between the engineering prop-

Von Quintus et al.

93

Models are included to predict fatigue cracking, rutting , moisture damage, and low-temperature cracking.

Sair.ple Ii Obtain Katerlal Components far

Mixture Design
Asphalt, Aggregate, H1neral Ftller,
Additives

""' Characteristics
l ur ' Or r 1'1111 t.ti. rh.P lu 1
.:it lht

Heuure g, Derlne t he Ph'f'altI

O...rttlrtutct a( t l'lc
Aggreg.ites, as Requ i red by
t he Spec if ica tio ns

Aspha l t, as Required by th e
Speclr1ca t 1ons

SELECT INITIAL AGGREGAT E

BLEND & RANGE OF
ASPHALT CONTENTS

LABORATORY PREPARATION AND

CONDITIONING PROCEDURES
Plant-Hardening Simulation

Pn:p11r1: 9H11111lnou l'lla tura

a t Each Selected Asphalt Content

INITIAL AGING OF LOOSE MIXTURE

Ccmpactlon of Three To Nine
SpectmnsatSelec:ted
Asphalt Content

r - - - - - i Measure Ai r Voids,

Untt W1:tght ,
YHA, and Ot her Propert les

Traf f1cDens l flc a t lonof

Specimens to Refusal

RESISTANCE TD
FRACTURE

RESISTANCE TD
SHEAR DISPLACEMENT

lndirt1:t Ttmolle Tests

9 11 F

Gyr.story Testing Machine

8 uo f

Resilient Kodulus, Strength

and ra llure St r1 h1s

Gy ratory Shear St ress

and St rains

RESISTANCE TD
UNIAXIAL DEFORMATION
Unconfined Uniaxial
Cmpress lon Tests 8 UM f

Res 11 lent Mcdu lus , Coopreu he

Strength .ind Failure Stra ins,
and Cree p Modulus

Heuure Atr Voids,

Unit Wetght, VHA, and Other

Proper t ies

SELECT DESIGN ASPHALT CONTENT

FIGURE 2 Flow chart for design of dense-graded asphalt

concrete mixtures.

erties and factors normally considered during mixture design.

From these analyses , it was found that the product of voids
filled with asphalt (VFA) and aggregate diameter were related
to work, VFA was related to tensile strain at failure, and
voids in mineral aggregate (VMA) was related to indirect
tensile strength. Through these correlations, an initial asphalt
content and aggregate gradation are selected before any mechanical tests are performed. Using the distress functions suggested in NCHRP Project 1-26, criteria for mixture optimization and adequacy have been presented for a range of traffic
and environmental conditions in the procedural manual.

Mixture Performance Evaluation

No consensus exists on the proper mathematical models to

use for predicting the behavior and performance of asphalt
concrete mixes. Such models are being developed by other
researchers under additional NCHRP contracts. Further, the
Strategic Highway Research Program (SHRP) is to conduct
research work in this same area under Contracts A-005 (Performance Models and Validation of Test Results) and A-001
(Development of Performance-Based Specifications for
Asphalt-Aggregate Mixtures) . These research efforts , however, will not be completed until 1993. Thus, the types of
performance relationship recommended by NCHRP Project
1-26 were used to evaluate the mixture's response to loads.

To determine the plant-hardening simulation, the penetration

and viscosity values of plant-produced material were compared to those conditioned (or aged) in the laboratory. The
thin-film oven test (TFOT) at 285F appeared to do a reasonable job of matching the asphalt cement characteristics
(penetration and viscosity value) after mix production. Thus,
the TFOT is used to predict the physical characteristics of the
asphalt after mix production, when there is no historical data
on binder aging through a mix plant.
The virgin asphalt cement is mixed with the aggregate blend
and the loose mix placed in a forced-draft oven set at 275F
(or the expected mix discharge temperature from the plant)
for 3 hr. After the first 1.5 hr, the mixture is removed from
the oven and remixed by hand and replaced in the oven for
the final 1.5 hr. The exact time the mixture is in the forceddraft oven can be determined from extraction tests of different
samples aged at different times (2 , 4, 8, 12, and 24 hr) . The
time selected is that which will age or harden the liquid asphalt
to the penetration and viscosity values measured from the
TFOT at 285F or to the actual binder properties after mix
production.
Mixture Compaction

Compaction was one of the critical factors studied in preparing

samples for laboratory evaluation. From an evaluation and
comparison of field cores and laboratory-compacted specimens, it was found that specimens compacted with the gyratory shear compactor more consistently matched the engineering properties measured on field cores. Thus, the gyratory
shear compactor was the device included in the procedural
manual.
Two methods of compaction are written into the procedure.
One uses the U.S. Army Corps of Engineers' gyratory testing
machine (GTM; ASTM D3387), and the other method uses
the Texas gyratory shear compactor (ASTM D4013). The
GTM is the preferred device , because angle of gyration , specimen height (an estimate of the decrease in air voids), and
the mixture's resistance to compaction can be monitored with
each revolution.
Moisture Conditioning

Another critical item was moisture conditioning or moisture

damage evaluation. Two procedures were used to evaluate
the moisture susceptibility of asphalt concrete mixtures . These
were the modified Lottman procedure , or AASHTO T283 ,
and the procedure documented in NCH RP Report 246, or the
Lottman procedure. The Lottman procedure recommended
in NCHRP Report 246 consistently showed a more severe

94

TRANSPORTA TION RESEARCH RECORD 1353

conditioning and testing technique. However, a version of the

modified Lottman procedure was used in AAMAS, because
of the concern that the procedure in NCHRP Report 246 is
too severe and unduly damages the specimens before testing.
In summary, a vacuum of 26 in. of mercury is applied to the
specimen for 15 min, the sample is frozen for 16 hr and placed
in a 140F water bath for 24 hr, and then it is tested. The test
temperature of the procedure is 77F, and a loading rate of
2.0 in./min is used, which are consistent with the other tests
used in AAMAS . The moisture-conditioned test specimens
are compacted to the air void level immediately after construction.

of its tie to the AASHTO design guide. ASTM D4123 (Indirect Tension Test for Resilient Modulus of Bituminous Mixtures) was the primary test procedure used . Three test temperatures are used: 41, 77, and 104F (5 , 25, and 40C) . The
secondary test method for resilient modulus is a modification
of ASTM D3497 (Dynamic Modulus of Asphalt Mixtures); it
is used as the conditioning procedure for all uniaxial compressive type tests. The differences or modifications are that
a rest period of 0.9 sec was used with a 0.1-sec load pulse and
that both the instantaneous and total resilient moduli were
calculated.

Environmental Aging Simulation

Indirect Tensile Strength and Failure Strain Test

A long-term age-hardening simulation procedure was also

developed. However, the change in physical properties of the
asphalt and mixture were available for the AAMAS test sections only over a short time period. The pavements were cored
twice: immediately after placement and 2 years after placement. The recommended procedure is to place compacted
specimens in a forced-draft oven set at 140F for 2 days. The
specimens are then rotated, the oven's temperature is increased to 225F, and the specimens are left in the oven for
additional 5 days.
These heat-conditioned specimens are then used for measuring the resilient modulus, indirect tensile strength, strain at
failure, and indirect tensile creep at 41F. These test specimens are also compacted to the air void level immediately
after construction.

The indirect tensile strength and failure strain are determined

using a test method derived from ASTM D4123 (Indirect
Tension Test for Resilient Modulus of Bituminous Mixtures).
The actual test procedure used is published in the AAMAS
final report (1). The test is conducted at the same three temperatures as the resilient modulus test: 41, 77, and 104F (5,
25, and 40C). The loading rate used at 41F is 0.05 and 2.0
in./min (1.27 and 50.8 mm/min); a rate of 2 in./min (50.8 mm/
min) is used only at 77 and 104F (25 and 40C). These data
are used in the fatigue cracking evaluation.

Traffic Densiflcation
Asphalt concrete mixtures density under traffic. To simulate
that densification process and its effect on the mixture's properties , specimens that had been compacted in the laboratory
to an air void content similar to that of the field specimens
were further compacted to a refusal density. This additional
densification was accomplished using the Corps of Engineers'
GTM. The initially compacted specimens are cooled to 140F
(60C) and then compacted further in the GTM gyratory device . Initial sample height readings were obtained before the
refusal densification and again after each 25 to 50 revolutions
of the machine. The compaction process is stopped when the
mixture's resistance reduces excessively or when there is an
excessive increase in the angle of gyration.
If the Corps of Engineers' GTM is unavailable, the procedure manual suggests that the Texas gyratory shear compactor be used. With the use of this device, however, the
traffic densification process continues immediately after initial
compaction (i.e., the temperature is not reduced and the mold
stays in the compaction machine) .
LABORATORY TEST PROCEDURES
Resilient Modulus Test
One of the primary test methods considered in the AAMAS
study was the repeated-load resilient modulus test, because

Indirect Tensile Creep and Recovery Test

The indirect tensile creep and recovery test is conducted on
field cores and on laboratory-compacted specimens . This test
is conducted in accordance with a procedure published in the
AAMAS final report. In summary, a static load of fixed magnitude is applied along the diametral axis and the horizontal
displacement measured over a 60-min loading duration. After
the fixed load is removed, the resilient strain is also measured
over 60 min . The indirect tensile creep and recovery test is
performed at a test temperature of 41F for use in the lowtemperature cracking evaluation.

Uniaxial Unconfined Compression Creep and

Recovery Test
Laboratory-compacted test specimens, which are a minimum
of 4 in. in diameter and 4 in. high, are tested in uniaxial
compression. A static load of fixed magnitude is applied along
the cylindrical axis of an asphalt concrete specimen for a set
amount of time. The total axial (compressive) deformation
response of the specimen is measured and used to calculate
the creep compliance at particular durations of time. After
the fixed load is released, the resilient deformation is also
measured over a set amount of time. Although the test can
be conducted with confining pressure on the specimens, all
testing for the AAMAS project was completed without the
use of confining pressures. For the AAMAS testing, the loading and unloading times are both 60 min and the data are
used in the rutting evaluation.
The unconfined compressive strength test is also performed
on a few specimens at 104F to calculate the compressive strain

Von Quintus

el

95

al.

at failure. This failure strain is used to set a limiting value for

the allowable permanent deformation in the rutting evaluation for the mix being tested.

KOT MIX ASPHALT CONCRETE DESIGN

GRAPHIC ANALYSES
MIXTURE DESIGN IDENTIFICATION No:
MIXTURE DESIGNATION: _ _ _ _ _ _ _ _ __
COMPACTION METHOD

AAMAS MIXTURE DESIGN PROCEDURE

INDIRECT
TENSILE
STRENGTK, PSI

DATE : _ _ _ __
_ _ _ __

PRO~ECT :

& DEVICE: _ _ _ _ _ _ _ _ _ _ _ _ _ __
UNCONFINED
COMPRESSIVE
STRENGTK, PSI

RESILIENT MODULUS, KSI

INDIRECT
TENSILE

UNCONFINED
COMPRESSION

Selection of Mixture Components

The program ASPHALT ( 4) is used as a starting point to
select a gradation and the seed asphalt content for that gradation . ASPHALT provides a theoretical relationship between asphalt content, VMA , air voids, and film thickness
for a specific aggregate blend or gradation. Mixture is prepared in the laboratory at the seed asphalt content, as well
as two asphalt contents above and two below this seed value.
ASPHALT CONTENT BY VOLUME,

Initial Heat Conditioning of Loose Mix

STRAIN AT
FAILURE,
MILS, INCH

ASPHALT CONTENT BY VOLUME,

CREEP
MODULUS,
KSI

GYRATORY SHEAR STRESS, PSI

(REVOLUTIONS=
)

Mixture at each of the asphalt contents is aged in a forceddraft oven using the procedure previously discussed (i.e. , placing a mix in the oven at 275F for 3 hr) .

Compaction
Specimens are compacted at each asphalt content using ASTM
D3387, if the Corps of Engineers' GTM is available, or ASTM
D4013 at a specified compactive effort . Three indirect tensile
specimens per asphalt content are compacted at an air void
level anticipated after construction (i .e. , 6 to 8 percent), and
a minimum of three uniaxial compression test specimens per
asphalt content are used in the traffic densification procedure
or compacted to the refusal density (i.e ., no increase in density
with additional compactive effort) . The design air void level
for the refusal density is 3 percent or greater.

Mixture Testing
Three specimens at each asphalt content are initially tested
for resistance to fracture (Figure 2). Indirect tensile resilient
moduli and strength tests are performed on the same sample
to define the initial allowable range of asphalt contents to
meet the design criteria for resistance to fracture.
Gyratory shear tests are run with the GTM during the traffic
densification procedure to ensure that minimum design requirements for shear are met (i.e., resistance to shear). For
mix design, a minimum shear value of 54 is used in the procedure.
Uniaxial compression creep and recovery tests are performed on specimens compacted to the refusal density. The
uniaxial compression creep and recovery test is used to ensure
that the design value will satisfy the deformation criteria (i.e. ,
resistance to deformation) . The minimum creep modulus value
used for design is dependent on the pavement structure. Figures 3 and 4 are included in the procedural manual for presenting the mixture design data.

ASPHALT CONTENT BY VOLUME,

ASPHALT CONTENT BY VOLUM!.,

FIGURE 3 AAMAS graphical presentation of mixture design

data for engineering properties.
Allowable Range of Asphalt Contents
The allowable range of asphalt contents for the specific aggregate gradation is defined as being those values that are
within the minimum and maximum limits as established by
the fatigue, shear, and deformation criteria. Figure 5 is a
graphical summary of the mixture design test for selecting the
design asphalt content and an allowable tolerance. These criteria include a minimum creep modulus for different structures, minimum gyratory shear strength, and minimum tensile
strain at failure for fatigue . Figure 6 illustrates the minimum
fatigue criteria.

AAMAS MIXTURE EVALUATION

Initial Heat Conditioning of Loose Mix
The mixture is aged in a forced-draft oven using the procedure
previously discussed (i.e., placing the mixture in the oven at
275F for 3 hr).

Compaction
After initial heat conditioning, eight sets of three specimens
are compacted. Eighteen specimens are compacted to be used

TRANSPORTATION RESEARCH RECORD 1353

96
HOT MIX ASPHALT CONCRETE DESIGN
GRAPHIC ANALYSES
MIXTURE DESIGN IDENTIFICATION

No : _ _ _ __ _ _

MIXTURE DESIGNATION :
COMPACTION METHOD

DATE: _ _ _ __

PROJECT: _ _ _ __

& DEVICE : _ _ _ __ _ _ _ _ _ _ _ _ _ __

AIR VOIDS, %
INITIAL

VOIDS FILLED WITH ASPHALT,

(AT REFUSAL)

REFUSAL

for indirect tensile testing and six specimens to be used for

uniaxial compression type testing. All specimens are compacted using the GTM because of its ability to monitor changes
in mixture behavior with a reduction in air voids. If the GTM
is unavailable, the specimens can be compacted using the
Texas gyratory shear compactor.
All specimens are initially compacted to the air void level
specified immediately after construction. Of the 18 specimens
compacted for indirect tensile testing, 9 are unconditioned , 6
are conditioned by the accelerated aging technique, and 3 are
moisture-conditioned. The six uniaxial compression specimens are all unconditioned and used in the traffic densification
procedure.

Grouping by Air Voids

ASPHALT CONTENT BY VOLUME,

ASPHALT CONTENT BY VOLUME,

MIX UNIT WEIGHT,

Before conditioning and testing, all specimens are grouped

in sets of three by air voids. The intent is to have both the
average air voids and standard deviation of air voids between
the different sample sets as close as possible.

VOIDS IN MINERAL AGGREGATE,

PCF

(AT REFUSAL)

INITIAL

Moisture Conditioning

ASPHALT CONTENT BY VOLUME,

ASPHALT CONTENT BY VOLUME,

FIGURE 4 AAMAS graphical presentation of mixture design

data for compaction properties.

Three specimens are moisture-conditioned in accordance with

a modification of AASHTO T283, with the exception that the
test procedure is at 77F and a loading rate of 2 in./min is
used.

Accelerated Aging

Six indirect tensile specimens are used for accelerated aging

for the low-temperature cracking evaluation . These specimens
ASPHALT CONTENT BY TOTAL VOLUME, %
ENGINEERING PROPERTIES
Total Resilient Modulus
(Layer Coefficients)
Tensile Strain at Failure and
Total Resilient Modulus
Gyratory Shear Stress and
Shear Index
Creep Modulus
COMPACTION PROPERTIES
Aggregate Unit Weight
Final Air Voids, %
VMA (Porosity),

VFA (Oeoree of Saturation!

Allowable Ranqe of the Design

Asphalt Content

ASPHALT CONTENT BY TOTAL WEIGHT, %

FIGURE 5

Worksheet for summarizing test results and selecting design asphalt content and tolerance.

Von Quintus et al.

97

40
30

"

:::'.

20
.'.'i
.<
;.:

o.i

"'
;:>

"";:::
H

...<

10

;:l

...
"'
"'""
H

"'z

..."'
"
"'

l .._.......

................

_..~~~~~'------"--'"-'-...;.._'--"'--...;.._

.6

.8

~~~~"'--~__;,__;,_..

20

10

LOG TOTAL RESILIENT MODULUS, psi x 10

__

40

"'--~

60

FIGURE 6 Minimum tensile strains at failure required for mixture as function of

total resilient modulus as measured by indirect tensile testing techniques.

are aged using the procedure previously discussed (i.e., placing the specimens in a forced-draft oven at 140F for 2 days,
then elevating the oven 's temperature to 225F for an additional 5 days of aging).

Traffic Densification
The six uniaxial compression specimens previously compacted
are all used in the traffic densification process. The temperature of these specimens is reduced to 140F. The specimens
are then placed in the GTM, and additional compactive effort
is applied to achieve the refusal density of the mix. The gyratory shear stress and sample height are monitored with a
number of gyrations during this densification process.
If the GTM is unavailable, specimens are compacted using
the Texas gyratory shear compactor to the refusal density or
air void content during initial compaction. In other words,
the six specimens are compacted to the refusal density for
uniaxial compression testing at the initial compaction temperature.

Mixture Testing
The repeated-load indirect tensile resilient modulus test is
performed on all unconditioned and conditioned specimens
(18 specimens). The indirect tensile strength and failure strains
are measured on all unconditioned and moisture-conditioned

specimens, and one set of accelerated age specimens (15 specimens). The indirect tensile creep and recovery test is performed on the second set of accelerated age specimens (three
specimens). These tests are performed on the specimens as
previously discussed and used to predict fatigue and lowtemperature cracking.
The uniaxial compression specimens are used to predict
rutting and distortion type failures from the uniaxial compressive resilient modulus, unconfined compressive strength
and failure strain, and compression creep and recovery tests.
The repeated-load uniaxial compression resilient modulus is
measured on all traffic-densified specimens at 104F (six specimens). The unconfined compressive strength is measured on
one set of traffic-densified specimens and the compressive
creep and recovery measured on the other set of specimens.

Mixture Performance Evaluation

For the fatigue cracking evaluation, the total resilient modulus
and tensile strain at failure are used for mixture evaluation.
These two values are measured on the unconditioned specimens are 41, 77, and 104F. These values can be used to
generate a fatigue cracking curve or can be compared to the
results for a "standard" mixture, as graphically illustrated in
Figure 6.
For thermal cracking, the test results from the indirect tensile strength and creep and recovery tests on the accelerated
age specimens are used to define the critical temperature

98

TRANSPORTATION RESEARCH RECORD 1353

change at which cracking can be expected to occur. The creep

and recovery tests results are used to evaluate and determine
the creep modulus at the lower testing temperature.
For the moisture damage evaluation of the mixture, the
tensile strength ratio, the resilient modulus ratio, and the
tensile strain at failure ratio are used to evaluate the mixes'
susceptibility to moisture damage.
For rutting evaluation, three tests are performed on the
traffic-den ified specimens. These are the repeated-load resilient modulus , the unconfined compressive strength , and the
compre sive creep and recovery te ts. The uniaxial compression creep and recovery test is used to eva luate the rutlin
potential of each mixture , a hown by an example in Figure
7. It can also be used to estimate the expected level of rutting
to occur in the pavement for each of the asphalt concrete
layers. The unconfined compressive strength and resilient
moduli are used to set limiting criteria for the mix under
specific loading conditions for the pavement and environment.

IMPLEMENTATION
It hould be recognized and understood that implementing
the AAMA concepts and methodology will not be a quick
proces becau e m t of these tests and evaluation procedure
are unfamiliar to some SHA personnel. Therefore, it becomes
very important that each agency take a sy tematic approach
in reviewing the AAMAS concept when considering its implementation.
It is obvious that many of those tests previously discussed
are not adaptable or practical for the use of field control of
mixtures. However, other fundamental properties of the mixture are adaptable to field control. These are the indirect
tensile strength and unconfined compressive strength. There
are relationships, which are mixture-dependent, between
strength properties and those used in AAMAS to evaluate
mixture performance.
The evaluation and implementation of the AAMAS concepts for the design and control of asphalt concrete mixtures

Area ol Low

Ruttl.na Polenllal

"fil,

104 +-- - - 1 -.....::,..__,i.-_

,;

__;;~..,,..-1----1-

~"0

J..rea of U:odente
Hulling Potenlle..I
0'-l'llnal MUlt\l'ra
Cho,raclerJllu)

;a

""fil

10 3 :r-----t--~-t---~i-----1~---1Area of

Hlah

Rulll.ng PolenUal

102 .._,_,_,,........_,_..._,...,
10

102

Loading Time, sec.

FIGURE 7 Asphalt conaete mixture rutting potential for

surface layers of asphalt concrete pavements.

still requires that many questions be answered about the policies o'f different agencies. A few of these are listed as follows:
Are the AAMAS tests practical or adaptable for field
control variables? For example, what length of time is required to run the test and analyze the results? Many contractors can place 2,000 to 3,000 tons of asphalt concrete a
day . It would be highly advantageous that the results of quality
control and acceptance tests be obtained within a short time
period. Additionally, some of the more sophisticated tests
may require that the expertise of field and laboratory personnel be upgraded from present levels, both for the contractor and for SHA personnel.
Can the AAMAS concepts be readily implemented where
different organizations are responsible for mix design? For
example, some SHAs require that contractors or consultants
conduct the mixture designs; others are responsible for mix
design themselves. In some cases, this may prevent smaller
contractors from competing on smaller projects, becau e they
do not have the financial backing to purchase the equipment.
I the AAMAS compaction, conditioning , and testing
equipment praclical for the field control of mixtures? If one
device is used in the laboratory for mixture design and another
device u ed in the fie ld , equivalency factors become extremely
important. Mo t eq uivalency factors are mixture-dependent
which can result in confu ion between the field and laboratory, similar to what exists to date with the empirical-based
methods.
Will AAMAS be cost-effective for those SHAs that control and accept mixtures based on specifications geared toward
method as opposed to end product?
With these few questions, implementation and acceptance
of the AAMA concepts will not be simple. There should be
at lea l four teps in the implementation process: (a) famil iarization with AAMAS , (b) training (c) education and (d)
fie ld pilot studies. The famil iarization with AAMA is simply
an under landing of the concept and methodology mployed
by AAMA . This i. a relat ively hol'l part of the implementation process.
The second step of the implementation process is training.
It is the more detailed in terms of how to run the test and
interpret the test results. Training is important to ensure that
the tests are performed in accordance with the procedure and
that the output of the tests is being interpreted properly .
The third part of the implementation is education. This is
prohahly the most important step toward full cale implementation of AAMAS. Basically, the education part i to
evaluate , on a trial basis, mixes for high-volume roadways.
1'he objective i to allow the user to become confident in
using AAMAS, understanding the properties measured and
sensitivity of those properties to pavement performance, and
establishing typical properties for their local materials. This
part of the implementation process is also the more timeintensive, because it involves most of the learning curve.
The final step of the implementation is conducting mix
designs and analyzing those mixes for actual project . This
tep is the o ne that leads to defining the time requirements
that are required to perform the tests on a routine basis and
to establish day-to-day operational procedures in a working
laboratory.

99

Von Quintus et al.

SUMMARY
In conclusion, the development of an AAMAS, as initiated
through NCHRP Project 9-6(1), is a very important element
of a multimillion-dollar research effort involving SHRP,
FHWA, and the asphalt pavement industry, an effort that
will ultimately result in improved performance of asphalt concrete pavements. Premature and costly pavement failures can
be drastically reduced by (a) designing structures that more
realistically consider traffic loadings, climate, and material
conditions; (b) selecting asphalt, aggregates, and additives or
modifiers consistent with the structural design; (c) producing
new or modified asphalt binders that provide the desired characteristics for minimizing distress; and (d) developing and
using performance-related specifications for control of construction.
ACKNOWLEDGMENTS
The research work summarized in this paper was performed
under NCHRP Project 9-6(1) by Brent Rauhut Engineering
and is reported in NCHRP Report 338. This project was spon-

sored by NCHRP. The primary consultants involved in the

work were Dallas N. Little of Texas A&M University, Robert
P . Lottman of University of Idaho, Byron E. Ruth of University of Florida, James L. Burati of Clemson University,
and Thomas W. Kennedy of University of Texas. Harry Smith
and Daniel Dearasaugh were the NCHRP contract technical
managers for this project.

REFERENCES
1. H . L. Von Quintus, J . A . Scherocman, C. S. Hughes, and T. W.
Kennedy. NCHRP Report 338: Asphalt-Aggregate Mixture Analysis System : AAMAS. TRB, National Research Council, Washington, D.C., March 1991.
2. M. R. Thompson, and E. Barenberg. Calibrated Mechanistic Structural Analysis Procedures for Pavements. Preliminary Draft Final
Report, NCHRP Project 1-26. TRB, National Research Council,
Washington, D.C., March 1990.
3. A Two-Day Workshop on an Asphalt-Aggregate Mixture Analysis
System: AA MAS. Workshop outline and notes. FHWA, U.S. Department of Transportation, March 1991.
4. R . A. Jimenez and D. A. Dadeppo. Asphall Concrele Mix Design.
Report FHWA/AZ-86/189. Arizona Department of Transportation, Phoenix, June 1986.