By: Jenny Liu, Ph.D.
Asphalt concrete (AC) is widely used as the surface layer for roadway pavement structure.
In cold regions, pavement structures covered with an asphalt concrete surface layer are usually vulnerable to thermal cracks. As asphalt pavement is cooled, the pavement contracts at a rate governed by the coefficient of thermal expansion leading to thermal stress. The top layers of pavement are usually exposed to larger temperature fluctuations than the layer below it. This difference in temperature between the top and bottom of AC layers creates a temperature gradient, with the thermally induced stress being largest at the top surface and decreasing towards the bottom of the asphalt layer. If the thermally induced stress is greater than the tensile strength of the asphalt, an initial crack develops on the surface and propagates downwards. Thermal cracks that develop perpendicular to traffic (major transverse thermal cracks) are thought to form in a top-down manner. This form of crack commonly penetrates a foot to several feet into the sub-pavement layers. This is one of the most prevalent pavement distresses found in Alaska and cold areas of other northern states, and requires significant repair efforts to maintain an acceptable pavement condition. The low temperature cracks are extensive enough that a significant portion of the department of transportation maintenance and operations budget is allocated to sealing and the associated work required to repair low-temperature cracking. Until new technology may someday eliminate cracking, significant funds will continue to be spent on crack sealing and repair. Innovative and cost-effective approaches and techniques to preserve and maintain existing highway systems (other than “worst-first”) are needed.
It’s been done before
The practice of introducing precuts on pavements to reduce maintenance and extend pavement life can be traced back to the 1950s. The concept of precutting pavement was first recommended as a method of controlling the location and severity of reflective cracks in an AC overlay on an old portland cement concrete (PCC) pavement. Many states, particularly in the northeastern U.S., have developed procedures for the design and construction of the saw-and-seal technique for asphalt overlays over PCC pavements. The first documented project evaluating a precut technique on new asphalt pavement was the Minnesota study completed in 1974. The purpose of the study was to determine if precutting joints would reduce or eliminate uncontrolled thermal cracking of flexible pavements. According to the study, precut depth and precut spacing are two significant factors for the success of this technique applied to new AC pavements. Another type of pavement structure for which the precut technique was considered was AC overlays over AC pavements to mitigate both thermal and reflective cracking. Therefore, precutting could be a very promising preservation technique for pavements in cold climates.
The Alaska Department of Transportation and Public Facilities’ (ADOT&PF) first attempt to investigate precutting as a method for potentially reducing thermal crack maintenance in Alaska was included in an experimental feature project in 1984. Precuts on Phillips Field Road remained in a very satisfactory condition during the long life of the pavement with minimal spalling and little to no settlement. After 32 years of service, its general driving condition was found to be good. Many longitudinal cracks and block cracks were observed on the precut sections. Only seven natural transverse cracks were found on the entire 1,300-ft precut section, including two low-severity cracks. Unfortunately, no formal study or continuous monitoring was conducted on Phillips Field Road. Though there had been no formal evaluation of this test section since its 1984 construction, this section provided a promising look at the longevity and potential success of the precut technique. The test section that remained, prior to 2016 rehabilitation, provided excellent informal positive reinforcement to the ideas tested on two new projects in 2012 and 2014.
A 10-mile reconstruction project near Healy included precuts in four separate roadway sections for almost 12,000 centerline ft of experimentally precut roadway.
Taking another look
In 2012 ADOT&PF built another 1-mile precut section in a repaving construction project at Richardson Highway. Different than the Phillips Field Road project, which involved new embankment and pavement, Richardson Highway construction involved only about the top 6 in. of the existing pavement structure—thus leaving thermal cracks in the underlying unbound materials. In 2014 a 10-mile reconstruction project near Healy, Alaska, included precuts in four separate roadway sections for a total of almost 12,000 centerline ft of experimentally precut roadway. Four sections represented major variations in new embankment thicknesses and pavement structure designs used within the project. To further understand important variables in the thermal cracking process and to develop a systematic approach for implementing application of precutting in AC pavements, the research team at University of Alaska-Fairbanks (UAF) has conducted continued field monitoring of these three sites in interior Alaska since 2012.
The Moose Creek test sections were constructed in the summer of 2012, and visited in summer/fall in 2013, 2014, 2016 and 2017, representing field surveys conducted one year, two years, four years and five years after the construction. As all the sections were not built in the same length, a parameter named “natural crack spacing” was used in this study instead of the exact amount of cracks, and was statistically determined by dividing the total length of the section by measured number of transverse cracks in the section. Generally, higher natural crack spacing indicates better control of the transverse crack locations and higher effectiveness of the precut technique. As expected, the natural crack spacing decreased with time, which means the number of cracks increased with time. It was found that all the precut sections demonstrated higher natural crack spacing compared with the control section, regardless of precut depth, precut spacing, and pavement age. This indicates that precut treatment may generally increase thermal cracking resistance of the pavement. Sixteen experimental sections, including four control sections at the Healy project, were established in 2014 and surveyed in 2015, 2016, and 2017. As expected, and consistent with the observation in the Moose Creek project, natural crack spacing decreased or remained the same with time, which means the number of cracks increased with time. It was found that the precut sections demonstrated higher natural crack spacing compared with the control section regardless of precut interval and pavement age for pavement structures II, III, and V. No natural transverse crack was found on the control section of pavement structure IV in 2015 and 2016, and only one crack was observed in 2017. Field investigation also showed that the precut technique behaved differently. Both active and non-active precut cracks were identified, and a precut crack could completely or partially capture a natural thermal crack.
In this study, the UAF team also revealed that the performance of the precut technique could be influenced by several factors such as precut spacing, precut depth and pavement structure. Most of these preliminary observations indicate that shorter precut spacing is more promising in controlling the location of natural cracks. In this study, various depths of cuts were used in pavement structures I (in the Moose Creek project) and IV (in the Healy project) only. Using 2017 survey results as an example, observations tenuously indicated that there may be an optimum precut depth (depth ratio – precut depth/AC thickness ratio = ½ in this study) that is the most effective in controlling the locations of the natural cracks. In terms of the effect of pavement structure, among five different types of pavement structures applied, the natural crack spacing results were consistent with the robustness of pavement structures. A stronger and/or perhaps thicker pavement structure is likely to be more effective in controlling the locations of natural transverse cracks when the precut technique is used.
The research team also conducted a preliminary cost analysis based on the necessary information collected from ADOT&PF Northern Region maintenance engineers and local contractors, and some reasonable assumptions as well. Three combinations of pavement design life and sealant life were used in this study in order to cover different situations, including 30-year pavement life and three-year sealant life, 30-year pavement life and five-year sealant life, and 20-year pavement life and five-year sealant life. The concept of net present value (NPV) was used to calculate the cost that happens until the pavement life is reached. An overall average savings for all scenarios combined of 21% was found for the Moose Creek sections compared to the control section. Among the three combinations, the one with 30-year pavement life and three-year sealant life is most advantageous to the precut technique, as sealing may be done on natural cracks most of the time and, using this combination, the precut installation cost carries the smallest weight during calculation. Thus, the combination of 20-year pavement life and five-year sealant life is the least advantageous to the precut technique. The precut technique is generally cost-effective even with varying design parameters. It is expected that the number of natural transverse thermal cracks will increase with time and, therefore, it is anticipated that the precut sections will show their cost-effectiveness in the future.
Moose Creek test sections were constructed in summer 2012, and revisited for field surveys in summer/fall of 2013, 2014, 2016 and 2017.
A crack through
The results from the Phillips Field Road precut test section were based on 32 years of casual observations plus a final 2016 careful inspection and mapping of thermal cracks. Those from the Richardson Highway, Moose Creek and the Parks Highway precut test sections were from relatively short time periods of observation. However, all three case studies indicated precutting to be an economically promising way of controlling natural thermal cracks in AC pavements in interior Alaska. Even short-term economic benefits, evaluated after three to five years of service, appear to range between 2% and 21%. Shorter precut spacing along with stronger and/or thicker pavement structures look promising with respect to crack control, according to preliminary results. Also, there may be an optimum precut depth that produces the best results. Continuing evaluation and monitoring of these test sections are needed to recommend an effective design methodology and construction practice for Alaska and other northern states.
About The Author: Liu is a professor of civil and environmental engineering at the University of Alaska-Fairbanks.