This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles

for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

Designation: D8237 − 21

Standard Test Method for

Determining Fatigue Failure of Asphalt-Aggregate Mixtures
with the Four-Point Beam Fatigue Device1
This standard is issued under the fixed designation D8237; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.

1. Scope 2. Referenced Documents

1.1 This test method provides a procedure for determining a 2.1 ASTM Standards:2
fatigue curve that is developed using three or more strain D8 Terminology Relating to Materials for Roads and Pave-
levels. The resulting data can be used in the fatigue models for ments
mechanistic-empirical pavement design (that is, Pavement D75/D75M Practice for Sampling Aggregates
ME). Failure points are determined for estimating the fatigue D140/D140M Practice for Sampling Asphalt Materials
life of 380 mm long by 50 mm thick by 63 mm in breadth D979/D979M Practice for Sampling Bituminous Paving
(width) asphalt mixture beam (rectangular prism) specimens Mixtures
sawed from laboratory or field-compacted asphalt mixture, D2041/D2041M Test Method for Theoretical Maximum
which are subjected to repeated flexural bending. Specific Gravity and Density of Asphalt Mixtures
1.2 The largest nominal maximum aggregate size (NMAS) D2726/D2726M Test Method for Bulk Specific Gravity and
recommended for beams 50 mm thick is 19 mm. Beams made Density of Non-Absorptive Compacted Asphalt Mixtures
with an NMAS greater than 19 mm might significantly inter- D3203/D3203M Test Method for Percent Air Voids in Com-
fere with the material response, thereby affecting the repeat- pacted Asphalt Mixtures
ability of the test. D3549/D3549M Test Method for Thickness or Height of
Compacted Asphalt Mixture Specimens
1.3 The text of this standard references notes and footnotes D3666 Specification for Minimum Requirements for Agen-
which provide explanatory material. These notes and footnotes cies Testing and Inspecting Road and Paving Materials
(excluding those in tables and figures) shall not be considered D5361/D5361M Practice for Sampling Compacted Asphalt
as requirements of the standard. Mixtures for Laboratory Testing
1.4 Units—The values stated in SI units are to be regarded D7981 Practice for Compaction of Prismatic Asphalt Speci-
as standard. No other units of measurement are included in this mens by Means of the Shear Box Compactor
standard, with the exception of degrees (°) where angle is D8079 Practice for Preparation of Compacted Slab Asphalt
specified in accordance with IEEE/ASTM SI 10. Mix Samples Using a Segmented Rolling Compactor
1.5 This standard does not purport to address all of the E4 Practices for Force Verification of Testing Machines
safety concerns, if any, associated with its use. It is the E29 Practice for Using Significant Digits in Test Data to
responsibility of the user of this standard to establish appro- Determine Conformance with Specifications
priate safety, health, and environmental practices and deter- E2309/E2309M Practices for Verification of Displacement
mine the applicability of regulatory limitations prior to use. Measuring Systems and Devices Used in Material Testing
1.6 This international standard was developed in accor- Machines
dance with internationally recognized principles on standard- IEEE/ASTM SI 10 American National Standard for Metric
ization established in the Decision on Principles for the Practice
Development of International Standards, Guides and Recom- 2.2 AASHTO Standard:3
mendations issued by the World Trade Organization Technical R 30 Standard Practice for Mixture Conditioning of Hot-Mix
Barriers to Trade (TBT) Committee. Asphalt (HMA)

2
For referenced ASTM standards, visit the ASTM website, [Link], or
1
This test method is under the jurisdiction of ASTM Committee D04 on Road contact ASTM Customer Service at service@[Link]. For Annual Book of ASTM
and Paving Materials and is the direct responsibility of Subcommittee D04.26 on Standards volume information, refer to the standard’s Document Summary page on
Fundamental/Mechanistic Tests. the ASTM website.
3
Current edition approved July 1, 2021. Published July 2021. Originally approved Available from American Association of State Highway and Transportation
in 2018. Last previous edition approved in 2018 as D8237 – 18. DOI: 10.1520/ Officials (AASHTO), 444 N. Capitol St., NW, Suite 249, Washington, DC 20001,
D8237-21. [Link]

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

1
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
3. Terminology location. A fully executed peak-to-peak displacement (δp-p) at
3.1 Definitions of Terms Specific to This Standard: the articulating H-frame third points of the beam is induced.
3.1.1 σp-p, n—peak-to-peak stress amplitude at load cycle i. The outer third points are held in an articulating fixed position
about the neutral axis of the beam. The frequency rate has a
3.1.2 σt, n—maximum tensile stress at the fiber of the beam. default frequency of 10 Hertz (Hz) and a test temperature of
3.1.3 ɛp-p, n—peak-to-peak tensile strain at load cycle i. 20 °C. This produces a constant bending moment over the
3.1.4 ɛt, n—maximum tensile strain at the bottom fiber of center third (L/3, length between outside clamps divided by 3)
the beam. span of 119 mm 6 0.5 mm (distance may vary between
3.1.5 δp-p, n—peak-to-peak displacement as determined in manufacturers; check with manufacturers’ specifications) be-
Fig. 1. tween the H-frame contact points on the beam specimen. The
level of desired strain is pre-calculated and an input value for
3.1.6 S, n—flexural beam stiffness, which is the stress the equipment peak-to-peak deflection. The peak-to-peak de-
divided by the strain. flection at mid-length position (L/2, length between outside
3.1.7 Si, n—the initial beam stiffness determined at 50 load frames divided by 2) of a beam specimen is regulated by the
cycles. closed-loop control system measured from the mid-height
3.1.8 failure point, n—the number of cycles to failure, Nf, position (neutral axis). The peak-to-peak deflection is mea-
which corresponds to the maximum or peak normalized beam sured relative to a fixed reference point located at the outer
stiffness × normalized cycles when plotted versus number of articulating fixed position.
cycles (9.9). NOTE 1—Caution should be applied when using frequencies above
3.1.9 normalized stiffness × normalized cycles, n—see 10 Hz, Pronk (2).
Rowe and Bouldin (1).4
5. Significance and Use
3.2 For definitions of other terms used in this standard, refer
to Terminology D8. 5.1 The laboratory fatigue life determined by this standard
for beam specimens has been used to estimate the fatigue life
4. Summary of Test Method of asphalt mixture pavement layers under repeated traffic
4.1 The four-point flexural bending test method is con- loading. Although the field performance of asphalt mixtures is
ducted on compacted beam specimens to evaluate the fatigue impacted by many factors (traffic variation, loading rate, and
properties of viscoelastic asphalt mixtures using a fixed refer- wander; climate variation; rest periods between loads; aging;
ence point bending beam fixture. A cyclic sinusoidal loading etc.), it has been more accurately predicted when laboratory
pattern is initiated having no rest periods from the start properties are known along with an estimate of the strain level
induced at the layer depth by the traffic wheel load traveling
over the pavement.
4
The boldface numbers in parentheses refer to a list of references at the end of
this standard. NOTE 2—The quality of the results produced by this standard are

FIG. 1 Illustration of Actuator Response of Repeated Sinusoidal Peak-to-Peak Defection

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

2
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
dependent on the competence of the personnel performing the procedure specimen and maintain the specimen at the default test tem-
and the capability, calibration, and maintenance of the equipment used. perature of 20 °C. The temperature shall be within 60.5 °C
Agencies that meet the criteria of Specification D3666 are generally
considered capable of competent and objective testing, sampling,
throughout the conditioning and testing times.
inspection, etc. Users of this standard are cautioned that compliance with NOTE 5—Replacing an incandescent, florescent, or halogen light bulb
Specification D3666 alone does not completely ensure reliable results. with light emitting diode (LED) for your environmental chamber reduces
Reliable results depend on many factors; following the suggestions of the heat signature and improves the chamber’s ability to control within
Specification D3666 or some similar acceptable guideline provides a 60.5 °C. Globe-style bulb design improves illumination of fixture and
means of evaluating and controlling some of those factors. inside of chamber.
6. Apparatus 6.1.3 Control and Data Acquisition System—During each
load cycle, the control and data acquisition system shall be
6.1 Test System—The test system shall consist of a load
capable of measuring the peak-to-peak displacement of the
frame, an environmental chamber (temperature control
beam specimen, and adjusting the load applied by the loading
system), and a closed-loop control and data acquisition system.
device such that the specimen experiences a constant level of
The test system shall include a closed-loop, computer-
displacement on each load cycle. In addition, it shall be capable
controlled loading component which, during each load cycle in
of recording load cycles, applied loads, beam displacements,
response to commands from the data processing and control
and temperature. Minimum data capture rate and sampling
component, adjusts and applies a load such that the specimen
intervals are listed in Table 2. The minimum number of data
experiences a constant level of controlled maximum deflection
samples for each load cycle is 200.
(and resulting strain) during each load cycle. The test system
shall meet the minimum requirements specified in Table 1. 6.2 Miscellaneous Apparatus and Materials—Means or
tools for targeting the displacement sensor to the neutral axis of
NOTE 3—Test system unit calibrations are performed in mm for the specimen and proper glue (cyanoacrylate) are required for
displacement and kN for load measurements (Practices E4 and E2309/
E2309M). Unit conversions will need to be made when applying to attaching the target to the specimen. A saw suitable for cutting
calculations in Section 10. the beams with parallel faces to the proper dimensions of
6.1.1 Loading Device—The loading device shall be capable 380 mm 6 3 mm in length, 50 mm 6 2 mm in height, and
of: (1) providing repeated sinusoidal loading at a frequency 63 mm 6 2 mm in breadth (width). A clamp alignment gauge
range of 5 to 25 Hz, and (2) subjecting specimens to four-point is required for setting the proper clamp spacing between the
bending with free rotation and horizontal translation at all frames, ensuring parallelism and perpendicularity. A rigid
clamped load and reaction points as shown in Figs. 2 and 3. material beam having the dimensions specified in 6.2 and
Floating reference point bending beam fixtures are not recog- tolerance of 0.254 mm across the beam (measured using a
nized by this standard. straightedge and feeler gauge) will be the required beam gauge
for setting the proper clamping height. Yearly verification is
NOTE 4—The fundamental equations are more viable with dual con- required for the beam gauge to be in compliance.
trolling displacement sensors. The on-specimen displacement sensor
controls the peak-to-peak displacement for the waveform loading of the NOTE 6—Hard, high-strength 7075 aluminum is found to be adequate
maximum deflection value at the L/2 location, and the frame-mounted for the beam gauge. The aluminum bar off the shelf will require being cut
displacement sensor controls the H-frame point of origin location. An to a length of 380 mm (McMaster – Carr Item #9055K31).
even better approach is the use of four displacement sensors. Two dual
controlling sensors, as listed previously in this note, and two recording the 7. Hazards
L/6 and 5L/6 locations to better understand the deflections between each
of the frames. 7.1 Observe standard laboratory safety precautions when
preparing and testing asphalt mixture specimens.
6.1.2 Environmental Chamber (Temperature Control
System)—The environmental chamber shall enclose the entire
8. Sampling and Test Specimen Preparation
8.1 Laboratory-Mixed and Compacted Specimens—Sample
TABLE 1 Test System Minimum Requirements asphalt binder in accordance with Practice D140/D140M, and
Load Measurement and Control Range: ±5 kN sample aggregate in accordance with Practice D75/D75M. If a
Resolution: 0.005 N
Accuracy: 0.01 N
complete fatigue curve is desired, prepare six to nine replicate
asphalt mixture beam specimens, compacted in accordance
Displacement Measurement and Control Range: ±2.5 mm with Practice D7981 or D8079, or active AASHTO compaction
Resolution: 2.5 µm
Accuracy: 5 µm
standards for slab(s) or beam(s). Otherwise, prepare as many
specimens as desired for individual beam test results.
Frequency Measurement and Control Range: 5 to 25 Hz Laboratory-prepared mixtures are conditioned with a short-
Resolution: 0.005 Hz
Accuracy: 0.01 Hz
term oven aging (STOA) process, such as defined in Section
7.2 of AASHTO R 30 (condition loose mixture for 4 h at
Temperature Measurement and Control Range: 5 to 25 °C 135 °C). Determine the theoretical maximum specific gravity
Resolution: 0.25 °C
Accuracy: ±0.5 °C in accordance with Test Method D2041/D2041M. Determine
the bulk specific gravity in accordance with Test Method
Displacement Sensor Linear variable differential D2726/D2726M. Calculate the percent air voids in accordance
transducer (LVDT), extensometer,
or similar device with Test Method D3203/D3203M. Test at least six replicate
asphalt mixture beam specimens at different strain levels in

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

3
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21

FIG. 2 Specimen Articulation and Dimensioning

FIG. 3 Load Characteristics of Fatigue Test Apparatus Illustrated as Pure Sine Wave

order to develop a fatigue curve, as shown in Fig. 4. The extra clamp. A linear relationship on a log-log plot exists between Nf
specimens may also be tested as desired if the data appears to and the level of tensile strain (µε, microstrain = strain × 106).
include an outlier or if a beam failure occurs directly at a

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

4
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
TABLE 2 Minimum Data Capture Rate and Sampling Intervals
Intervals Cycles at each collection points
Repetitions
(Spaced equally within each range) (Included in average reported)
0 to 10 1–10 Report individual cycles
10 to 1000 10 5
1000 to 10 000 90 5
10 000 to 100 000 At least one every 1000 repetitions 5
100 000 to end of test At least one every 10 000 repetitions 5

FIG. 4 Example Fatigue Curve

NOTE 7—AASHTO R 30 also contains additional information on 8.3 Roadway Specimens—Obtain compacted asphalt mix-
long-term oven aging (LTOA) of compacted specimens for five days at ture samples from the roadway in accordance with Practice
85 °C. In addition, new research in Braham et al. (3) and NCHRP Report
871 (4) provides information on long-term aging loose mixture.
D5361/D5361M. Determine the theoretical maximum specific
NOTE 8—The type of compaction device (linear kneading, rolling gravity in accordance with Test Method D2041/D2041M.
wheel, vibratory) may influence the test results relative to representing Determine the bulk specific gravity in accordance with Test
actual construction. Check with the manufacturer recommendations on Method D2726/D2726M. Calculate the percent air voids in
compaction procedures or applicable ASTM or AASHTO standards. accordance with Test Method D3203/D3203M.
8.2 Plant-Mixed, Laboratory-Compacted Specimens—
Obtain asphalt mixture samples in accordance with Practice 8.4 Specimen Trimming—Saw at least 6 mm from all sides
D979/D979M. If a complete fatigue curve is desired, prepare of each compacted slab edge to mitigate end effects and
six to nine replicate asphalt mixture beam specimens, com- provide smooth, parallel (saw-cut) surfaces for mounting the
pacted in accordance with Practice D7981 or AASHTO active neutral axis target. The final required dimensions of the test
compaction standards for slab(s) or beam(s). Otherwise, pre- specimen, after sawing, are 380 mm 6 3 mm in length, 50 mm
pare as many specimens as desired for individual beam test 6 2 mm in height, and 63 mm 6 2 mm in breadth (width).
results. See Note 7 for long-term oven aging of specimens, if Measure the height and breadth of the specimen to the nearest
that is necessary. Determine the theoretical maximum specific 0.01 mm at three or more different points along the middle
gravity in accordance with Test Method D2041/D2041M. 100 mm of the specimen length in accordance with the
Determine the bulk specific gravity in accordance with Test applicable sections of Test Method D3549/D3549M. Deter-
Method D2726/D2726M. Calculate the percent air voids in mine the average of the measurements for each dimension and
accordance with Test Method D3203/D3203M. Test at least six record the average to the nearest 0.01 mm. The allowed
replicate asphalt mixture beam specimens at different strain difference between maximum and minimum measured values
levels in order to develop a fatigue curve, as shown in Fig. 4. of breadth and height is 1 mm. If the difference of the
The extra specimens may also be tested as desired if the data maximum and minimum values of either dimension exceeds
appears to include an outlier or if a beam failure occurs directly 1 mm, then the beam shall be recut or discarded.
at a clamp. A linear relationship on a log-log plot exists NOTE 9—Previous experience has shown that in order to minimize
between Nf and the level of tensile strain (µε, microstrain = specimen variability, it is recommended that the beams be immediately
strain × 10–6). labeled to ensure consistent orientation (top and sides) during testing,

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

5
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
relative to the compaction process. Masad et al. (5) shows that the air 9.2 Place the specimen on a stiff, flat surface in an environ-
voids at the compaction plate/compaction keys are lower than the air voids ment holding the desired test temperature for 2 h to ensure that
at the bottom of the sample.
the specimen has equilibrated to the desired test temperature
8.5 Specimen Storage—The specimens should be stored on prior to beginning the test. Temperature (T) for this test method
a 12.7-mm steel plate or similar material capable of supporting will have a default T = 20 °C. The temperature shall be within
the beams, with a flatness of 0.508 mm across the surface of 60.5 °C throughout the conditioning and testing times.
the shortest section of plate from end to end. This flat surface
NOTE 10—Two hours is sufficient to equilibrate the temperature of a
keeps the beam specimens from being pre-strained before
beam that was stored near room temperature to its testing environment.
testing. It is permissible to stack a second beam on top of the NOTE 11—Previous experience has shown that agencies have modified
first beam on storage racks. the standard to bracket 10 °C ≤ T ≤ 20 °C. The selection of temperature is
based on the climatic region for the geographical area and depth in the
9. Procedure pavement structure.

9.1 Attaching the Target to the Neutral Axis of Specimen— 9.3 Fixture Alignment—The beam gauge referenced in 6.2,
Locate the center of a specimen on one of its 50-mm high or any other alignment tools, are used for setting the proper
lengthwise sides (that is, mid-height and mid-length of the clamp height (point of origin) and correct spacing to prevent
beam). Place the beam so that the side having the target is pre-straining the beam during the clamping procedure. This
face-up before gluing to counteract gravitational effects. Apply will ensure proper alignment of the beam fixture H-frame in
cyanoacrylate (super glue) or equivalent in a circle around this reference to the articulating fixed-position outer frame clamps
point and place the target on the glue such that the top of the prior to testing. To complete the alignment and removal of the
target is at the center point and parallel to the neutral axis of the beam gauge, refer to the equipment manufacturer’s detailed
beam. Allow the glue to cure before moving the specimen. Fig. procedures.
5 illustrates the target attached to the neutral axis of the NOTE 12—If the top and bottom faces of the beam test specimen are not
specimen. parallel, it should not be an issue with the clamping. The saw cuts are

FIG. 5 Target Attached to the Beam Neutral Axis (Mid-Height, Mid-Length)

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

6
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
typically straight on all sides of the beam; even if these are not exactly to the specimen at the desired strain rate and resulting maximum
parallel to each other, the top clamps will compensate for the lack of peak-to-peak deflection that was calculated for the on-specimen displace-
parallelism, since the clamps are all independent of each other. ment sensor at mid-height and the L/2 location.
NOTE 13—It is recommended that the clamps be flat and have the
following minimum dimensions: 25 mm wide by 63 mm long, using a 9.6 Select the desired initial peak-to-peak strain (50 to
force of 600 to 1000 N to hold the beam in place while not causing any 3000 µε; typically 200 to 800 µε for conventional asphalt
indentation. mixtures; 50 to 150 µε (endurance limit determination) for
9.4 Open the clamps and slide the specimen into horizon- evaluating severely high-repetition but low-strain conditions;
tally centered position as shown in Figs. 6-8. When the 1500 to 3000 µε for some interlayer and bridge deck materials)
specimen clamps are in the proper positions, apply the clamps and loading frequency, and the load cycle intervals at which
in accordance with the manufacturer’s clamping procedures. test results are to be recorded and computed. The load cycle
Check for adequate clamping pressure by toggling (lightly intervals shall be recorded at least as frequently as listed in
shaking) each frame with the spacing fixture in place and make Table 2. Enter these values into the specific template for this
sure that all clamps are seated properly, flat against the testing program in the control and data acquisition system.
specimen. 9.7 Within the load cycles to be recorded, include an
9.5 Follow the manufacturer’s procedure to correctly set up interval near the point of 50 cycles. Determine the specimen
and position the on-specimen displacement sensor at the stiffness at the 50th load cycle; this stiffness is the recom-
mid-height and L/2 location. Once the specimen has been mended estimate of the initial beam stiffness.
mounted, allow 20 min for the temperature in the test chamber 9.8 Select a peak-to-peak displacement level (peak-to-peak
to equilibrate. Set the target peak-to-peak displacement to the strain level) near the mid-range initially for the specific
desired strain rate in the test control software. material based on trial and error or experience, such that the
NOTE 14—Switching of the stroke actuator point of origin (start specimen will undergo a minimum of 10 000 load cycles prior
location) to the on-specimen displacement sensor for fixed reference to failure. A minimum of 10 000 load cycles ensures that the
displacement sensor fixture during the 20-min equilibrium time prior to specimen does not decrease in stiffness too rapidly. Adjust the
starting the test will cause the H-frame attached to the actuator to creep peak-to-peak strain up and down on additional replicate beams
upward at approximately the same displacement rate the specimen creeps
downward between the two middle frames due to the mix’s viscous nature to evaluate performance of the material over a range of
for the specific mix design at the L/2 location. All mixtures flow at peak-to-peak strain levels in order to establish a fatigue curve.
different rates due to the materials selected for the mix design.
NOTE 15—Previous experience has shown that vertical beam fixtures
9.9 After selecting the appropriate test parameters, begin the
mitigate the creep between the clamping frames due to gravitational test. Activate the control and data acquisition system so that the
effects of the viscoelastic asphalt mixture beams in a horizontal orienta- test results at the selected load cycle intervals are monitored
tion. Beams with the dimensions given in 8.4 do not flow as much when and recorded, ensuring that the test system is operating
tested in a vertical configuration. Vertical beam fixtures have different properly. Ideally, the test should be terminated sometime after
orientation and require different fixture design to perform the test.
NOTE 16—Additional creep at the two middle frames on a horizontal the normalized beam stiffness (Ŝ) × normalized cycles (N̂) peak
beam fixture will increase due to gravitational effects from acceleration value (failure point) has been achieved on a graphical plot of
due to test frequency selection and friction induced by the energy applied normalized beam stiffness × normalized cycles versus cycles

FIG. 6 Schematic of Fixed Reference Displacement Sensor of Flexural Beam Fatigue Test Apparatus (Side View)

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

7
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21

FIG. 7 Schematic of Fixed Reference Displacement Sensor of Flexural Beam Fatigue Test Apparatus (Top View)

FIG. 8 Schematic of Flexural Beam Fatigue Test Apparatus (Side View)

(Nf), as shown in Fig. 9. To extend beyond this failure point, it crack by bending the beam on the edge of a table to show its fracture
is suggested that the test be terminated after the beam flexural location. If the beam fracture location is at one of the clamp locations, then
you should record this and report all possible premature failure reasons.
stiffness reduces by at least 15 percentage points beyond Use the clamp alignment gauge, or make a gauge to recognize the clamp
failure. With low-strain testing, it may be impractical to reach contact locations when checking for clamp edge failures. All acceptable
this desired failure point. failures should fracture in the middle 100 mm of the beam.
NOTE 17—See Tsai et al. 2002 (6) and 2003 (7) for two-stage Weibull 10. Calculation or Interpretation of Results
function to extrapolate failure point when using low-strain testing.
NOTE 18—The point of failure can also be continually evaluated by 10.1 Perform the following calculations at the specified load
monitoring the cyclic stress-versus-strain plot on the scope for the point at cycle intervals in Table 2:
which the hysteresis loop (continuous plot of stress versus strain during 10.1.1 Peak-to-peak stress on beam (σp-p) (kPa):
loading) collapses, loses its shape, appears horizontal, or combinations
thereof. 3 3a 3P
σ p2p 5 3 1000 (1)
NOTE 19—Once the beam is removed from testing, fully expose the b 3 h2

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

8
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21

FIG. 9 Normalized Beam Stiffness × Normalized Cycles versus Cycles

where: 10.1.6 Normalized stiffness × normalized cycles (MPa ×

a = distance between inside clamps centerline to centerline, cycles/MPa × cycles)—See Refs (1, 8, 9):
L/3 (mm). Distance may vary between manufacturers; Si 3 Ni
check with manufacturer’s specifications, Ŝ 3 N̂ 5 (6)
S0 3 N0
P = peak-to-peak load applied on beam (N),
b = average specimen breadth (width) (mm), and where:
h = average specimen height (mm). Ŝ × N̂ = normalized beam stiffness × normalized cycles,
10.1.2 Maximum tensile stress on beam (σt) (kPa): Si = flexural beam stiffness at cycle i (MPa),
Ni = cycle i,
σ p2p S0 = initial flexural beam stiffness (MPa), estimated at
σt 5 (2)
2 approximately 50 cycles, and
10.1.3 Peak-to-peak strain (ɛp-p) (µɛ): N0 = actual cycle number where initial flexural beam
stiffness is estimated.
12 3 δ
p2p 3h
ε p2p 5 3 1 000 000 (3) 10.1.7 Failure Point (Nf, Number of Cycles to Failure)—
~ 3 3 L 2! 2 ~ 4 3 a 2!
Occurs at the maximum or peak value of normalized stiffness
where: × normalized cycles when plotted versus number of cycles
δp-p = peak-to-peak deflection at center of beam (mm), (Fig. 9) (1, 8, 9). A portion of the data is shown in Table 3 to
a = distance between inside clamps centerline to show how the beam fatigue data is processed. The calculated
centerline, L/3 (mm). Distance may vary between normalized stiffness data can be fit to a best-fit six-order
manufacturers; check with manufacturer’s polynomial curve or Logit model (10). The derivative of that
specifications, function and finding the point where the slope equals zero can
L = length of beam span between outside clamps center- be used for easily selecting the peak of the curve and
line to centerline (mm). Verify the manufacturer’s determining Nf. In this example, the beam failure point is
dimensions, and determined to be approximately 608 000 cycles.
h = average specimen height (mm).
11. Report
10.1.4 Maximum tensile strain (ɛt) (µɛ):
11.1 Report the specimen identification number.
ε p2p
εt 5 (4) 11.2 Asphalt Mixture Description—Report the binder type,
2
binder content, aggregate gradation, and air void percentage.
10.1.5 Flexural beam stiffness (S) (MPa): 11.3 Specimen Dimensions—Report the specimen length,
σt average specimen height, and average specimen breadth
S5 3 1000 (5)
εt (width) in millimeters, within two significant digits.

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

9
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
TABLE 3 Beam Fatigue Data Processing 11.6 Report the average test temperature to the nearest
Normalized
Curve Fit 0.25 °C.
Normalized
Beam Stiffness 11.7 Report the maximum tensile stress (kPa).
Cycle Stiffness
Stiffness, ×
Number ×
MPa Normalized
Normalized 11.8 Report the maximum tensile strain.
Cycles
Cycles
50 939.41 1.00 0.67
11.9 Report the initial flexural beam stiffness at cycle 50
60 927.11 1.18 0.81 (MPa).
70 917.79 1.37 0.94
80 905.66 1.54 1.08 11.10 Report the flexural beam stiffness (MPa).
90 900.05 1.72 1.21
100 892.01 1.90 1.35
11.11 Report the measured number of cycles to failure (Nf).
200 841.51 3.58 2.69 11.12 Prepare a plot of normalized beam stiffness × normal-
300 812.37 5.19 4.04
400 790.35 6.73 5.38 ized cycles versus cycles as shown in Fig. 9.
... ... ... ...
180 000 439.71 1685.06 1644.05 12. Precision and Bias
200 000 434.43 1849.81 1791.10
220 000 432.06 2023.69 1940.22 12.1 Precision—The within-laboratory repeatability stan-
240 000 424.59 2169.46 2092.53
260 000 420.98 2330.29 2248.67
dard deviation on a log scale has been determined to be 0.278,
280 000 414.39 2470.23 2408.80 based on testing in one laboratory, three test replicates at each
300 000 408.79 2610.94 2572.59 strain level (ranging from 200 to 2000 µɛ), and eleven different
320 000 408.02 2779.75 2739.30
340 000 402.72 2915.13 2907.73
types of homogenous asphalt mixture specimens. Therefore,
360 000 398.18 3051.83 3076.29 the results of two properly conducted tests by the same
380 000 396.47 3207.51 3243.02 operator on similar replicate beam specimens in the same
400 000 390.33 3324.06 3405.62
420 000 385.52 3447.24 3561.51
equipment at the same strain level should not differ by more
440 000 381.08 3569.82 3707.87 than 0.787 on a log scale. Table 5 shows a summary of the data
460 000 377.68 3698.79 3841.70 used to generate the within-laboratory standard deviation. This
480 000 372.72 3808.87 3959.87
500 000 368.18 3919.27 4059.19 estimate of precision is a combination of both replicate
520 000 363.49 4024.10 4136.49 specimen preparation variability and testing variability. The
540 000 357.77 4113.12 4188.69 between-laboratory reproducibility of this test method is being
560 000 350.92 4183.80 4212.86
580 000 342.70 4231.77 4206.37 determined and will be available on or before December 2023.
600 000 333.40 4258.79 4166.90 Therefore, this standard should not be used for acceptance or
620 000 321.81 4247.83 4092.61 rejection of a material for purchasing purposes.
640 000 307.72 4192.91 3982.20
660 000 291.22 4091.99 3835.03 NOTE 21—The beam specimens for this precision statement were
680 000 271.57 3931.57 3651.25 compacted using a linear kneading compactor. A single compacted slab
700 000 248.78 3707.61 3431.91
will yield three individual beam specimens. The beam fatigue fixture used
a fixed referenced displacement sensor at the neutral axis at the L/2
location.
NOTE 20—See Practice E29 for information on determination of 12.2 Bias—No information can be presented on the bias of
significant figures in calculations. this method for measuring fatigue life because this is a
11.4 Minimum recordable fields are listed in Table 4. destructive test and no material having an accepted reference
Record each load cycle interval selected by the operator within value is available.
two significant figures for the units listed in Table 4.
11.5 Report all units to within two significant figures for 13. Keywords
Section 11. A conversion factor will have to be applied for Eq 13.1 asphalt mixture fatigue; asphalt mixture flexural test-
1-4 since stress and stiffness are in kPa and MPa while length ing; asphalt mixture stiffness; asphalt mixture tensile testing;
and span are in mm. fatigue life; flexural bending

TABLE 4 Example Test Results Table

Peak-to-Peak
Applied Load Beam Deflection Peak-to-Peak Strain Flexural Stiffness
Load Cycle Stress Ŝ × N̂
(N) (mm) (µ) (MPa)
(kPa)

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

10
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
TABLE 5 Single-Laboratory Precision Data
Standard
Asphalt Controlled Standard
Log Deviation
Mixture Strain Nf Deviation
Nf Log
ID (micro) Nf
Nf
1 300 85 790 4.933
300 165 560 5.219
300 54 682 57 191 4.738 0.242
2 300 39 392 4.595
300 197 160 106 172 5.295 0.431
300 241 360 5.383
200 805 800 5.906
200 1 190 120 6.076
200 408 320 390 918 5.611 0.235
3 300 192 480 5.284
300 24 278 91 001 4.385 0.458
300 48 176 4.683
200 966 160 5.985
200 1 105 600 6.044
200 306 759 426 693 5.487 0.306
4 300 75 670 4.879
300 12 290 4.090
300 43 782 31 690 4.641 0.405
200 287 000 5.458
200 456 200 5.659
200 179 833 139 339 5.255 0.202
5 800 21 310 4.329
800 8780 3.943
800 20 674 7058 4.315 0.219
400 250 960 5.400
400 36 656 4.564
400 105 440 109 418 5.023 0.418
6 350 222 920 200 180 5.348 0.235
350 555 240 5.744
350 582 440 5.765
7 2000 417 039 131 469 5.620 0.139
2000 294 039 5.468
2000 556 799 5.746
8 2000 437 999 138 582 5.641 0.202
2000 218 839 5.340
2000 181 479 5.259
9 2000 482 559 117 542 5.684 0.136
2000 487 719 5.688
2000 281 600 5.450
10 2000 186 439 83 625 5.271 0.391
2000 32 737 4.515
2000 52 479 4.720
11 2000 115 359 56 030 5.062 0.155
2000 210 839 5.324
2000 112 299 5.050
Average: 139 127 0.278

REFERENCES

(1) Rowe, G. M., and Bouldin, M. G., “Improved Techniques to Evaluate Asphalt Concrete,” Transportation Research Record, Vol 1681, 1999,
the Fatigue Resistance of Asphaltic Mixtures,” Proceedings of the 2nd pp. 179–185.
Eurasphalt and Eurobitume Congress, Barcelona, Spain, 2000. (6) Tsai, B. W., Harvey, J. T., and Monismith, C. L., “High Temperature
(2) Pronk, A. C. “Collaborative Study with 4PB Devices in Europe – Fatigue and Fatigue Damage Process of Aggregate-Asphalt Mixes,”
Round Robin Test With Three Reference Beams, Final Report,” 2016. Journal of the Association of Asphalt Paving Technologists, Vol 71,
(3) Braham, A. F., Buttlar, W. G., Clyne, T. R., Marasteanu, M. O., and 2002, pp. 345–385.
Turos, M. I., “The Effect of Long-Term Laboratory Aging on Asphalt (7) Tsai, B. W., Harvey, J. T., and Monismith, C. L., “Application of
Concrete Fracture Energy,” Journal of the Association of Asphalt Weibull Theory in Prediction of Asphalt Concrete Fatigue
Paving Technologists, Vol 78, 2009, pp. 417–454. Performance,” Transportation Research Record Vol 1832, 2003, pp.
(4) Kim, Y. R., Castorena, C., Elwardany, M., Rad, F., Underwood, S., 121–130.
Gundla, A., Gudipudi, P., Farrar, M., and Glaser, R., “Long-Term (8) Rowe, G. M., “Performance of Asphalt Mixtures in the Trapezoidal
Aging of Asphalt Aggregate Mixtures for Performance Testing and Fatigue Test,” Journal of the Association of Asphalt Paving
Prediction,” NCHRP Report 871, Washington DC, 2018. Technologists, Vol 62, 1993, pp. 344–384.
(5) Masad, E., Muhunthan, B., Shashidhar, N., and Harman, T., “Quan- (9) Hopman, P. C., Kunst, P. A., and Pronk, A. C., “A Renewed
tifying Laboratory Compaction Effects on the Internal Structure of Interpretation Method for Fatigue Measurements, Verification of

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

11
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG
 D8237 − 21
Miner’s Rule,” 4th Eurobitume Symposium, Madrid, Spain, 1989. phalt Four-Point Bending Fatigue Testing,” Transportation Research
(10) Mateos, A., Wu, R., Denneman, E., and Harvey, J., “Sine versus Record, published online 2018.
Haversine Displacement Waveform Comparison for Hot Mix As-

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@[Link] (e-mail); or through the ASTM website
([Link]). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; [Link]

&RS\ULJKWE\$670,QW O DOOULJKWVUHVHUYHG 0RQ-DQ*07

12
'RZQORDGHGSULQWHGE\
9HOORU,QVWLWXWHRI7HFKQRORJ\SXUVXDQWWR/LFHQVH$JUHHPHQW1RIXUWKHUUHSURGXFWLRQVDXWKRUL]HG