Low temperature cracking is attributed to tensile stresses induced in an
asphalt concrete pavement that develop when the pavement is subjected to a cold
temperature. Cracking results in poor ride quality and a reduction in service life of
the pavement. Low temperature cracking has been predicted by regression
equations, mechanistic approaches and by simulation measurements. The purpose of
the study reported herein is to (1) evaluate the Thermal Stress Restrained Specimen
Test (TSRST) as an accelerated performance test to simulate low temperature
cracking of asphalt concrete mixtures and (2) develop a deterministic and
probabilistic model to predict low temperature cracking with TSRST results.
Construction histories, cracking observations and temperature data were
collected for five test roads in Alaska, Pennsylvania and Finland. A full scale and
fully controlled low temperature cracking test program was conducted at the U.S.
Army Cold Regions Research and Engineering Laboratory (USACRREL).
Specimens were fabricated in the laboratory with original asphalt cements and
aggregates from the test roads. In addition, asphalt concrete pavement specimens
were cut from the test sections. The TSRST results obtained for these samples were
correlated with the field observations. Based on a statistical analysis of the data, the
TSRST fracture temperature is associated with the field cracking temperature and
crack frequency for the test roads where mixture properties dominated low
temperature cracking. It was concluded that the TSRST can be used to simulate low
temperature cracking of asphalt concrete mixtures.
A deterministic and a probabilistic model were developed to predict crack
spacing as a function of time using the TSRST results, pavement thickness and bulk
density, pavement restraint conditions and air temperature. The affect of aging on
pavement properties was incorporated in the models by predicting the field aging
with Long Term Oven Aging (LTOA) treatment in the laboratory. The calculation
of the crack spacing is based on the theory that the pavement slab cracks when the
pavement temperature reaches the cracking temperature of the mixture and the slab
is fully restrained. The deterministic model predicts crack spacing with time
whereas the probabilistic model predicts crack spacing and its variation with time
and yields the reliability of the design with regard to a minimum acceptable crack
spacing criterion defined by road authorities.
The models were verified by comparing the predicted crack spacings for the
five test roads to the observed crack spacings. The probabilistic model is
recommended for use in predicting the low temperature cracking of asphalt concrete
mixtures. / Graduation date: 1994
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/35604 |
| Date | 21 June 1993 |
| Creators | Kanerva, Hannele K. |
| Contributors | Vinson, Ted S. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0016 seconds