Dr. Habtamu Melese (habtamu.zelelew@lucyconsulting.

com)
Please use only for class purposes.

ASPHALT MIXTURE PERFORMANCE TESTER (AMPT)

The Asphalt Mixture Performance Tester (AMPT) was utilized to conduct the performance
testing of the asphalt mixtures included in this project (Figure 14). The AMPT testing simulates
the true field conditions such as traffic loading, rate of loading, temperature, and confinement
that the pavement experiences during its design life. The AMPT equipment is capable of
conducting several performance tests such as dynamic modulus (|E*|), S-VECD fatigue, flow
number (FN), and overlay.

The accuracy of the dimensions of the AMPT test specimens is vital to obtain valid
performance test specimens suitable for testing. The AASHTO PP 601 provisional standard
practice tolerance limits were followed for specimen diameter, height, end flatness, and
perpendicularity (Table 11). The asphalt mixtures for performance testing were compacted to 8.5
% air voids in the gyratory compactor in order to achieve the 7.0 + 0.5 % target air void content
of the AMPT test specimen. The test specimens were cored from the center 100 mm of a 150
mm diameter SGC specimen. The sample ends were trimmed from a height of more than 180
mm down to 150 mm. The AMPT fatigue test specimen dimension included a diameter of 100
mm and a height of 130 mm. Water was used during coring and sawing to avoid disintegration of
core samples.

1
AASHTO PP60 “Preparation of Cylindrical Performance Test Specimen Using the SGC”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Figure 14. IPC Global AMPT

Table 11. Specimen Fabrication Criteria

Dynamic Modulus
Property Fatigue**
and Flow Number*

Average Diameter (mm) 98 to 104 100 to 104

Standard Deviation of Diameter (mm) < 0.5 < 0.5
Height (mm) 147.5 to 152.5 127.5 to 132.5
End Flatness (mm) < 0.5 < 0.5
End Perpendicularity (mm) < 1.0 < 1.0
*
AASHTO PP 60
**
AASHTO TP 107
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

DYNAMIC MODULUS, |E*| TEST

Dynamic modulus is the modulus of a viscoelastic material (i.e. asphalt mixture). In linear
viscoelastic theory, the absolute value of the complex modulus |E*|, by definition, is the dynamic
(complex) modulus. The dynamic modulus of an asphalt mixture test is a response developed
under sinusoidal loading conditions and is normally identified by E*. In the general literature, the
term dynamic modulus is often used to denote any type of modulus that has been determined
under non-static load conditions. The dynamic modulus test is a stress-controlled test where a
sinusoidal axial compressive load is applied to the AMPT specimen and the resulting applied
stress and recoverable axial strain response are measured. It is calculated by dividing the
maximum peak-to-peak stress by the recoverable peak-to-peak strain. The time lag in degrees
between the applied stress and resulting strain defines phase angle (ϕ).
Temperature and load frequency influence the dynamic modulus response of asphalt
mixtures. The AASHTOWare Pavement ME Design guide recommends dynamic modulus
testing performed using five test temperatures (-10, 4.4, 21.1, 37.8, and 54.4 C) and six
frequencies (0.1, 0.5, 1, 5, 10, and 25 Hz). Due to equipment cost considerations, the AMPT test
equipment was not developed to achieve and control a test temperature of -10 °C. The dynamic
modulus test can be performed using the remaining four test temperatures (4.4, 21.1, 37.8, and
54.4 C) and the -10 °C data is determined through the calculation of the dynamic modulus
master curve.
Since the dynamic modulus test is non-destructive at low temperatures, the same set of four
replicates of the asphalt mixtures were tested at the three lower temperatures (4.4, 21.1, and 37.8
°C) while another set of four replicates were fabricated and tested at the high temperature (54.4
°C) (Table 12). In accordance with AASHTO TP 792 test method, the unconfined dynamic
modulus tests were performed from the lowest temperature to the highest temperature and from
the highest frequency to the lowest frequency.

2
AASHTO TP79 “Determining the Dynamic Modulus and Flow Number for HMA using AMPT”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Table 12. Dynamic Modulus Test Matrix

Temperature (°C)
Mix ID Remark
4.4 21.1 37.8 54.4
PMLC1-GTR20-RAP10 4 Replicates 4 Replicates
Plant
PMLC2-GTR20-RAP10 4 Replicates 4 Replicates
Produced
PMLC3-GTR20 4 Replicates 4 Replicates

Evaluation of Dynamic Modulus Test Data

The dynamic modulus test results are summarized in Appendix E. The data quality indicators
shown in Table 13 were verified at each loading frequency and test temperature during the
dynamic modulus test. The indicators were used to assess the quality or acceptability of the
measured |E*| data. If one or more of the indicators exceed the limit, the measured |E*| values
will be rejected. In this project, the reported |E*| values for each test temperature and frequency
comply with this criteria.
The dynamic modulus test variability was assessed using the coefficient of variation (COV)
statistics obtained by dividing the standard deviation by the mean of dynamic modulus values. A
decrease in test frequency and an increase in test temperature resulted in higher dynamic
modulus test variability. The black space diagram was used to evaluate the inconsistencies in the
data trends and identify errors in the dynamic modulus test dataset. The black space diagram is
represented by using the measured |E*| and phase angle in semi-log space (Figure 15). Slight
dynamic modulus test variability is observed at higher temperature ranges.
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Table 13. Data Quality Statistics Requirement3

Data Quality Statistic Limit

Deformation drift In the direction of applied load

Peak-to-peak strain 75 to 125 microstrain (Unconfined)

Load standard error 10 %

Deformation standard error 10 %

Deformation uniformity 30 %

Phase uniformity 3°

1.E+05
PMLC1-GTR20-RAP 10

PMLC2-GTR20-RAP 10
Dynamic Modulus, |E*| (MPa)

1.E+04 PMLC3-GTR20

1.E+03

1.E+02

1.E+01
0 10 20 30 40
Phase Angle (deg.)

Figure 15. Mixture Black Space Diagram

3
AASHTO TP79 “Determining the Dynamic Modulus and Flow Number for HMA using AMPT”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Dynamic Modulus Test Results

The dynamic modulus test results (dynamic modulus and phase angle) of the PMLC
asphalt mixtures tested at four temperatures and six frequencies are presented in Figures 16
and 17. Generally, at a constant test temperature, the dynamic modulus increases with
increasing loading frequency; at a constant loading frequency, the dynamic modulus
decreases with increasing test temperature. Interpretation of the data provided in these
figures can provide a fundamental characterization of asphalt mixture behavior under
various loading and temperature conditions through the use of master curves.
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

15000

Dynamic Modulus, |E*| (MPa)

PMLC1-GTR20-RAP10
12000 PMLC2-GTR20-RAP 10
PMLC3-GTR20
9000

6000

3000

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(a)
8000
Dynamic Modulus, |E*| (MPa)

PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
6000
PMLC3-GTR20

4000

2000

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(b)
3000
Dynamic Modulus, |E*| (MPa)

PMLC1-GTR20-RAP10
2500
PMLC2-GTR20-RAP10
2000 PMLC3-GTR20

1500

1000

500

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(c)
1000
Dynamic Modulus, |E*| (MPa)

PMLC1-GTR20-RAP 10
800 PMLC2-GTR20-RAP10
PMLC3-GTR20
600

400

200

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(d)

Figure 16. Dynamic Modulus; (a) 4.4 °C, (b) 21.1 °C, (c) 37.8 °C, and (d) 54.4 °C
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

60
PMLC1-GTR20-RAP10

Phase Angle, f (deg.)

50
PMLC2-GTR20-RAP 10
40 PMLC3-GTR20

30

20

10

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(a)
60
PMLC1-GTR20-RAP10
Phase Angle, f (deg.)

50
PMLC2-GTR20-RAP 10
40 PMLC3-GTR20

30

20

10

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(b)
60
PMLC1-GTR20-RAP10
Phase Angle, f (deg.)

50 PMLC2-GTR20-RAP10
40 PMLC3-GTR20

30

20

10

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(c)
60
PMLC1-GTR20-RAP 10
Phase Angle, f (deg.)

50 PMLC2-GTR20-RAP10
40 PMLC3-GTR20

30

20

10

0
25 10 5 1 0.5 0.1
Frequency (Hz)
(d)

Figure 17. Phase Angle; (a) 4.4 °C, (b) 21.1 °C, (c) 37.8 °C, and (d) 54.4 °C
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Mixture Master Curve

The dynamic modulus test results of an asphalt mixture at various temperatures can be
shifted along the temperature or frequency axis to form a single dynamic modulus master
curve at a desired reference temperature or frequency. This procedure takes advantage of
the time-temperature superposition principle of viscoelastic materials and allows the user to
look at the mixture response at multiple loading frequencies and temperatures on a single
master curve. The dynamic modulus test data was used to construct master curves using the
standard sigmoidal function given in Equation 7. Similarly, the phase angle master curves
were developed using the modified sigmoidal function (Equation 8). The Williams-Landel-
Ferry (WLF) relation was used to describe the shift factors (Equation 9).

(7)

(8)

(9)

where,
|E*| = dynamic modulus
= minimum value of |E*|
= maximum value of |E*|
ωr = reduced frequency
and = parameters describing the shape of the master curve
= phase angle
= shift factor
T = temperature
Tref = reference temperature
b, c, d, C1, and C2 are model constants
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

The dynamic modulus and phase angle master curves, in log-log and semi-log scales,
for the asphalt mixtures at a reference temperature of 21.1 °C are shown in Figures 18 and
19, respectively. Visually the stiffness and viscoelastic properties of the mixes are
comparably similar. Presence of 10 % RAP did not appreciably increase the results of
dynamic modulus test.

1.E+05
PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
PMLC3-GTR20
1.E+04
Dynamic Moduus, |E*| (MPa)

1.E+03

1.E+02

1.E+01

10000000
0.00001

100000000
100000

1000000
0.0001

1000

10000
0.000001

0.01

10
0.001

100
0.1

Reduced Frequency (Hz) (T Ref = 21.1 °C)

(a)

25,000
PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
20,000 PMLC3-GTR20
Dynamic Moduus, |E*| (MPa)

15,000

10,000

5,000

0
10000000
0.00001

100000000
100000

1000000
0.0001

1000

10000
0.000001

0.01

10
0.001

100
0.1

Reduced Frequency (Hz) (T Ref = 21.1 °C)

(b)
Figure 18. Dynamic Modulus Master Curves; (a) log-log scale and (b) semi-log scale
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

100
PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
PMLC3-GTR20

Phase Angle, f (deg.)

10

10000000
0.00001

100000000
100000

1000000
0.0001

1000

10000
0.000001

0.01

10
0.001

100
0.1

1
Reduced Frequency (Hz) (T Ref = 21.1 °C)
(a)

40
PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
PMLC3-GTR20
30
Phase Angle, f (deg.)

20

10

0
10000000

100000000
0.00001

100000

1000000
0.0001

1000
0.000001

10000
0.01
0.001

10

100
0.1

Reduced Frequency (Hz) (T Ref = 21.1 °C)

(b)
Figure 19. Phase Angle Master Curves; (a) log-log scale and (b) semi-log scale
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

FATIGUE TEST
Fatigue cracking is one of the major types of distress in asphalt pavements. Fatigue
testing of asphalt mixtures involves subjecting specimens to repeated loading either using a
controlled stress mode or a controlled strain mode. Over the years, several laboratory test
methods have developed to evaluate and characterize the fatigue performance of asphalt
mixtures. The commonly used fatigue tests include the beam fatigue test, semi-circular
bending fatigue test, and direct tension or indirect tension fatigue test. These fatigue test
methods are empirical in nature. Fatigue cracking in asphalt pavements is a complicated
phenomenon and requires mechanistic approaches with rigorous theoretical considerations.
Recently, the researchers at the North Carolina State University (NCSU), led by Professor
Richard Kim, have developed a Simplified Viscoelastic Continuum Damage (S-VECD)
model for characterizing the fatigue properties of asphalt mixtures.

A Simplified Viscoelastic Continuum Damage (S-VECD) Model

The S-VECD model is mathematically a rigorous approach where fundamental material
properties are incorporated. The key function is the damage characteristic curve (C versus
S) that relates the amount of damage (S) in a specimen to the material integrity or pseudo
stiffness (C). A detailed description of the derivation of S-VECD model is referenced in the
footnote below4. Overall, a complete characterization of damage in asphalt mixes can be
obtained using the S-VECD approach under wide range of temperature, strain, frequencies
as well as prediction of endurance limit and number of cycles at failure. The S-VECD
model requires the linear viscoelastic (LVE) material properties to determine viscoelastic
damage characteristic relationship. The ALPHA-Fatigue proprietary software was used to
analyze results of finger print dynamic modulus and fatigue tests.

4
Underwood, B. S., Baek C., and Y. R. Kim (2012) “Simplified Viscoelastic Continuum Damage Model as
Platform for Asphalt Concrete Fatigue Analysis”. Journal of the Transportation Research Board, No. 2296,
Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 36-45.
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Linear Viscoelastic (LVE) Characterization

The S-VECD model requires the LVE testing at multiple temperatures and frequencies.
The primary LVE material responses are the dynamic modulus and phase angle. Three
AMPT dynamic modulus tests were performed for each mix to obtain the LVE properties.
Dynamic modulus tests were performed using four temperatures (4.4, 21.1, 37.8, and 54.4
°C) and six frequencies (25, 10, 5, 1, 0.5, and 0.1 Hz). The testing order is from low to high
temperatures and from high to low frequencies in order to minimize damage to the
specimens. The LVE tests were performed using specimen dimensions of 100 mm x 150
mm (diameter x height). The target on-specimen peak-to-peak strain level within 50 to 75
microstrains was used.
The fingerprint dynamic modulus test under tension-compression (“pull-push”) mode
was performed before fatigue testing wherein the sample is subjected to a very low non-
damaging strain amplitude. The modulus measured in fingerprint test is denoted by .
The specimen-to-specimen variability can be evaluated using the Dynamic Modulus Ratio,

. A DMR of unity denotes no variability among specimens. The average

DMR values of the PMLC mixes range between 0.76 and 0.94 indicating acceptable
specimen-to-specimen variability (Figure 20). The error bars shown in the figure indicate
the standard deviation of DMR resulted from three test specimen replicates.

2.0

1.5
DMR

1.0

0.5

0.0

Figure 20. Dynamic Modulus Ratio (DMR)

Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Viscoelastic Damage Characterization

The viscoelastic damage characterization is determined by performing controlled strain

cyclic tension (“pull-pull”) test. The fatigue tests were performed using specimen
dimensions of 100 mm x 130 mm (diameter x height). Three extensometers were mounted
on the sides of each specimen at 120° apart. All fatigue tests were performed at a constant
frequency of 10 Hz using a target temperature of 21 °C. Selection of this test temperature is
consistent AASHTO TP 1075. This intermediate temperature is suitable for the material’s
viscoelastic damage characterization without the effect of viscoplasticity, which leads to
specimen damage and macro cracks. Three on-specimen strain levels (450, 550, and 600
microstrain) were used to establish the linear viscoelastic damage envelope of the asphalt
mixes. The number of cycles at which a sharp decrease in phase angle defines the failure
cycle, Nf (Figure 21). A typical mid-specimen failure which is an indication of good fatigue
test is shown in Figure 22. The failure occurs within the measurement zone where the
extensometers are capable of capturing the evolution of damage throughout the entire test.

60

50
Phase Angle, f (deg.)

40

30

20

10
Nf

0
0 2000 4000 6000 8000
Number of Cycle

Figure 21. Phase Angle versus Number of Cycle

5
AASHTO TP 107 “Determining the Damage Characteristic Curve of Asphalt Concrete from Direct
Tension Cyclic Fatigue Tests”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Figure 22. Fatigue Test Mid-Specimen Failure

The ALPHA-Fatigue version 3.0.5 software developed by InstroTek Inc was used to
analyze the fatigue test data. The damage characteristic curves (C versus S curves) for the
asphalt mixtures are shown in Figure 23, which depicts the mixture’s resistance to damage.
Comparatively, the PMLC mixes resulted in similar damage characteristic curve. The
curves shown in these figures may not be sufficient to rank mixtures’ fatigue performance
as it ignores the significant influence of the materials resistance to deformation 6. Such a
comparison can be achieved by using predicted endurance limits over wide range of test
temperatures (5 °C to 25 °C). Figure 24 presents the predicted endurance limits of the
asphalt mixtures. The PMLC mixtures demonstrated similar endurance limits. Presence of
10 % RAP did not appreciably increase resistance to fatigue cracking.

6
B. S. Underwood, Y. R. Kim, and M. N. Guddati. (2006). “Characterization and Performance Prediction of
ALF Mixtures Using a Viscoelastoplastic Continuum Damage Model.” Journal of Association of Asphalt
Paving Technologists, Vol. 75, pp. 577-636.
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

1.0
PMLC1-GTR20-RAP10
PMLC2-GTR20-RAP10
0.8 PMLC3-GTR20

Pseudo stiffness, C
0.6

0.4

0.2

0.0
0.E+00 2.E+04 4.E+04 6.E+04 8.E+04 1.E+05 1.E+05 1.E+05

Damage, S

Figure 23. Damage Characteristics Curve

250
Endurance Limit (Microstrain)

PMLC1-GTR20-RAP 10

200 PMLC2-GTR20-RAP 10

PMLC3-GTR20
150

100

50

0
0 5 10 15 20 25 30

Temperature ( C)

Figure 24. Endurance Limit

Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

FLOW NUMBER (FN) TEST

The permanent deformation properties of the asphalt mixtures were evaluated using the
Flow Number Test (FN). The axial deviator stress in FN test is repeatedly applied for 0.1 s
followed by a 0.9 s rest period. In this test, the number of load cycles and cumulative axial
strains are continuously recorded. The number of load cycles corresponding to the
minimum rate of change of permanent axial strain during a repeated load test defines FN.
The Francken method calculates FN iteratively at every load cycle until the rate of change
of accumulated strain reaches a minimum global value. This method assigns a FN value of
10,000 for the asphalt mixtures with no tertiary flow. The test terminates at either 10,000
load cycles or the accumulation of 50,000 microstrain (5 % strain), whichever occurs first.
The FN test temperature was determined using LTPPBind version 3.1 software. The
LTPP report showing the closest weather stations (NJ 6843) to the pavement project
location is shown in Figure 25. The temperature corresponding to the 50 % reliability
pavement temperature 20 mm below the surface and unadjusted to traffic was determined
as 54.1 °C and was chosen as the FN test temperature. The FN test conditions including
deviator stress levels and number of specimens are presented in Table 14. The axial
deviator stress levels included 600 and 800 kPa. A confinement stress of 69 kPa was used
for the confined FN tests. The FN tests include a contact stress (5 % of deviator stress) to
establish uniform stress state within the specimen.
 Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Figure 25. LTPPBind Pavement Location Temperature7

Table 14. Flow Number Test Matrix

Unconfined FN Confined FN
Mix ID (Confinement = 0 kPa) (Confinement = 69 kPa) Remark
600 kPa 800 kPa 600 kPa 800 kPa
PMLC1-GTR20-RAP10 X X X X
Plant
PMLC2-GTR20-RAP10 X X X X
Produced
PMLC3-GTR20 X X X X

7
LTPPBind version 3.1 software
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Flow Number Test Results

Figure 26 presents the average flow number of the asphalt mixtures tested under
unconfined and confined FN test conditions. The error bars shown in these figures indicate
the standard error of FN measurements using four test specimen replicates. An increase in
FN may yield higher deformation resistance and vice-versa. The PMLC mixes measured
very low flow number values (within 50 cycles) when tested at 600 kPa unconfined FN
(Figures 26a). The PMLC1 and PMLC2 mixes both containing 10 % RAP slightly
measured higher flow number. The stress dependency of the asphalt mixtures during FN
test is also shown in this figure. An increase in deviator stress resulted in increased damage
to the specimen and therefore resulted in lower FN values. The confined FN test resulted in
flow number of 10,000 cycles when tested at 600 kPa (Figures 26b). Detailed flow number
test results for each test replicate and their overall statistics are presented in Appendix F.
AASHTO TP 79 minimum average flow number requirements for unconfined flow
number tested using 600 kPa is shown in Table 15. For the asphalt mixes included in this
project, the minimum average flow number is 105 cycles. Based on the finding presented in
Figure 26a the asphalt mixes failed TP 79 requirements.

Table 15. Minimum Average Flow Number Requirements8

HMA WMA
Traffic Level,
Minimum Average Minimum Average
million ESALs
Flow Number Flow Number
<3 - -
3 to < 10 50 30
10 to < 30 190 105
≥ 30 740 415

8
AASHTO TP79 “Determining the Dynamic Modulus and Flow Number for HMA using AMPT”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

600 kPa 800 kPa

80

Flow Number, FN (Cycle) 60

40

20

0
PMLC1-GTR20-RAP 10 PMLC2-GTR20-RAP 10 PMLC3-GTR20
(a)

600 kPa 800 kPa

12000
Flow Number, FN (Cycle)

9000

6000

3000

0
PMLC1-GTR20-RAP 10 PMLC2-GTR20-RAP 10 PMLC3-GTR20
(b)

Figure 26. Flow Number; (a) Unconfined and (b) Confined

Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

HAMBURG TEST
The permanent deformation and moisture damage (stripping) properties of the asphalt
mixtures were evaluated using the Hamburg test in accordance with AASHTO T3249. The
Hamburg wheel-track device is shown in Figure 27. It is an electrically powered test device
capable of moving a steel wheel (203.2 mm diameter and 47 mm wide) back and forth
across a test specimen. The load on the wheel is 705 ± 4.5 N. The wheel moves
approximately 50 passes across the specimen per minute. The maximum speed of the wheel
is approximately 0.305 m/s. It has a water bath capable of controlling the temperature
within ± 1.0 °C over a range of 25 to 70 °C. A gauge capable of measuring the depth of the
impression of the wheel within 0.01 mm is mounted on this device to measure the depth at
the midpoint of the wheel’s path on the test specimens.
A typical Hamburg test results corresponding to non-stripping and stripping mixes is
shown in Figure 28. The Hamburg test parameters include, creep slope, stripping slope, and
stripping inflection point. The creep slope is the inverse of the deformation rate within the
linear region of the deformation curve. It indicates rutting susceptibility of asphalt mixtures
due to traffic loading. The stripping slope is the inverse of the deformation rate after
stripping occurs. It measures the accumulation of permanent deformation due to moisture
damage. It is used to estimate the relative resistance of the asphalt mixtures to moisture-
induced damage. The stripping inflection point is the number of wheel passes
corresponding to the intersection of the creep slope and the stripping slope. In other words,
it is the number of wheel passes at which moisture damage starts.
The Hamburg test requires two test specimens for each of the wheel-tracks. The asphalt
mixtures prepared using 7 ± 0.5 % target air voids were compacted in accordance with
AASHTO T 31210. The resulting specimen dimension was 61 mm height and 150 mm
diameter. Table 16 presents the number of specimen replicates used in Hamburg test. The
Hamburg test was conducted at 50 °C until 20,000 passes or 20 mm rut depth, whichever
occurred first.

9
AASHTO T 324 “Standard Method of Test for Hamburg Wheel-Track Testing of Compacted HMA”
10
AASHTO T 312 “Standard Method of Test for Preparing and Determining the Density of Hot Mix Asphalt
(HMA) Specimens by Means of the Superpave Gyratory Compactor”
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Figure 27. Hamburg Test Experimental Setup

Non-stripping
Creep Slope

Stripping
Inflection Point

Stripping
Slope
Stripping

Figure 28. Typical Hamburg Test Result

Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Table 16. Hamburg Test Matrix

Mix ID No. of Replicate Remark

PMLC1-GTR20-RAP10 5 Replicates
Plant
PMLC1-GTR20-RAP10 4 Replicates
Produced
PMLC1-GTR20 4 Replicates

Hamburg Test Results

The Hamburg test results of the asphalt mixes are presented in Figure 29. These figures
include the pictorial representations (top view and front view) of the specimens after
Hamburg test. Summary of the Hamburg test results (i.e., creep slope) for the studied
asphalt mixtures are shown in Figure 30. Higher creep slope relates to higher rate of
deformation. Visually, the PMLC mixes measured comparable deformation and moisture
damage resistance.
The Texas Department of Transportation (TxDOT) specifies the minimum number of
wheel passes in Hamburg test to reach a rut depth of 12.5 mm when tested at a temperature
determined by the performance grade of the asphalt binder. As presented in Table 17, these
values are >10000 for mixes produced with PG 64 binder, >15000 for mixes produced with
PG 70 binder, and >20000 for mixes produced with PG 76 binder. The asphalt mixes tested
in Hamburg were evaluated as per the TxDOT specification and none of the mixes failed
the minimum required number of passes (Table 18).
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

Number of Passes
0 5000 10000 15000 20000
0

-5

Rut Depth ( mm)

-10

-15

-20
PMLC 1 SP 2&8 LWP PMLC 1 SP 3&9 RWP PMLC 1 SP 5&10 LWP

PMLC 1 SP 6&7 RWP PMLC 1 SP 1&4 LWP

(a)
Number of Passes
0 5000 10000 15000 20000
0

-5
Rut Depth ( mm)

-10

-15

-20
PMLC 2 SP 7&10 RWP PMLC 2 SP 5&8 LWP

PMLC 2 SP 4&9 RWP PMLC 2 SP 6&11 LWP

(b)
Number of Passes
0 5000 10000 15000 20000
0

-5
Rut Depth ( mm)

-10

-15

-20
PMLC 3 SP 7&12 LWP PMLC 3 SP 4&5 RWP

PMLC 3 SP 10&11 LWP PMLC 3 SP 8&9 RWP

(c)

Figure 29. Hamburg Test Results; (a) LMLC1, (b) LMLC2, and (c) LMLC3
Dr. Habtamu Melese ([email protected])
Please use only for class purposes.

0.000

Creep Slope (%) -0.010

-0.020

-0.030

Figure 30. Creep Slope

Table 17. TxDOT Hamburg Test Requirement11

High Temperature Minimum Passes to

Binder Grade 12.5 mm Rut Depth
PG 64 or lower 10,000
PG 70 15,000
PG 76 or higher 20,000

Table 18. Evaluation of Rutting using TxDOT Specification

Binder Passes to 12.5

Mix ID Pass/Fail Remark
Grade mm Rut Depth
PMLC1-GTR20-
PG 64-22 > 20,000 Pass
RAP10
PMLC2-GTR20- Plant
PG 64-22 > 20,000 Pass Produced
RAP10
PMLC3-GTR20 PG 64-22 > 20,000 Pass

11
A Manual for Design of Hot Mix Asphalt with Commentary, NCHRP Report 673 (2011).