Field Evaluation of the System for Calibration of the Marshall Compaction Hammer.
Field Evaluation of the System for Calibration of the
Marshall Compaction Hammer.
(5005 K)
Shenton, H. W., III; Cassidy, M. M.
NISTIR 5553; FHWA-RD-95-063; 76 p. February 1995.
Sponsor:
Federal Highway Administration, Washington, DC
Available from:
National Technical Information Service
Order number: PB95-190674
Keywords:
compaction hammer; Marshall method; asphalt; bituminous;
building technology; calibration; pavement design;
standards; tests
Abstract:
A system for calibrating the Marshall compaction hammer
has recently been developed at the National Institute of
Standards and Technology, in collaboration with the
American Association of State Highway Transportation
Officials (AASHTO), Materials Reference Laboratory
(AMRL). The calibration system consists of a
spring-mass device with an integral force transducer and
a high-speed data acquisition system. The force
delivered by the hammer to the calibration device is
recorded as a function of time and analyzed to determine
the peak force and impulse. Time histories from a
series of hammer blows are analyzed to determine the
average peak force, average impulse, and cumulative
impuse. The proposed calibration procedure is based on
adjusting the number of hammer blows delivered to a
specimen, such that a standard compactive effort is
supplied during the compaction process, regardless of
slight variations in the Marshall hammer. In an earlier
laboratory evaluation program, the calibration system
and procedure proved to effective in reducing the
variability of Marshall test results. Presented in the
report is a summary of a field evaluation program of the
calibration system and procedure. In this study,
Marshall specimens were prepared in bituminous
laboratories using "production" Marshall hammers:
twelve laboratories, or field "sites", participated in
the study. Sixteen Marshall specimens were prepared at
each site. Four specimens were prepared using a
standard 50-blow Marshall procedure and four using a
standard 75-blow Marshall procedure; these are referred
to as the uncalibrated specimens. Four specimens were
prepared using a calibrated blow count corresponding to
a standard 50-blow cumulative impulse, and four were
prepared using a calibrated blow count corresponding to
a standard 75-blow cumulative impulse; these are
referred to as the calibrated specimens. Height, air
voids, flow and stability were determined for each of
the specimens. Results were compiled and analyzed to
determine the between-laboratory variability of the data
for the uncalibrated and calibrated specimens. The
system was, in general, ineffective in reducing the
between-laboratory variability of the test results in
the full data set. The variability of the calibrated
test results increased or decreased relative to the
uncalibrated results for different specimen properties.
The system was effective, however, in reducing the
variability of the test results when evaluated in the
reduced data set, that included results from nine sites;
data from three sites were eliminated in the statistical
analysis because these results were believed to be
flawed, in a way that may have compromised the Marshall
test results. In the reduced data set, the variability
of the calibrated test results decreased by as much as
thirty percent, relative to the uncalibrated results.
One possible explanation for the marginal reduction in
variability with calibration is that the study sample of
Marshall hammers was typical of the total population:
nine of the twelve machines were from the same
manufacturer, and six of those were less than three
years old. The machines provided reasonably consistent
results, thus, there was little room for improvement.
This is supported by comparison of the uncalibrated test
results to data from the AASHTO Materials Reference
Laboratory, Proficiency Sample Program.
Building and Fire Research Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899