Practical Recommendations for Construction of Longitudinal Joints in Hot Mix Asphalt (HMA) Pavements

Rajib B. Mallick, Ralph White Family Distinguished Professor, Civil and Environmental Engineering, Worcester Polytechnic Institute (WPI), Massachusetts, USA, and Veeraragavan Amirthalingam, Professor, Department of Civil Engineering, IIT Madras

Hot Mix Asphalt (HMA) is made up of primarily mineral aggregates and asphalt binder that are heated and mixed in a plant at an elevated temperature (~ 150oC), placed with a paver at a lower but sufficiently high temperature (~ 140oC), and compacted with rollers before the temperature drops below a certain level, typically 90-100oC.

Compaction generally becomes very difficult to accomplish once the temperature of the mix drops below approximately 90oC because of increased stiffness. This change in stiffness is a result of the change in stiffness of the asphalt binder, which behaves as a viscous material at high temperatures, and a viscoelastic material for most other temperature regimes. A well-designed and constructed HMA pavement provides a durable, smooth and watertight surface at a relatively low cost.

HMA (also known as asphalt) pavements do not require the creation of joints as is generally the case for portland cement concrete (PCC) pavements. However, unless echelon paving is adopted, a longitudinal joint occurs in between adjacent paved lanes. In the first pass of the paver, the first lane is constructed. When the adjacent lane is placed, the first lane has cooled and is referred to as the cold side of the joint. The second lane is often referred to as the hot side of the joint. During compaction of the edge of the cold lane the pavement is not restrained, hence, a lower density (higher air voids) is anticipated close to the edge relative to the areas away from the edge. There is confinement when compacting the edge (hot side) of the adjacent lane since it is placed up against the cold joint. Because of the issues at the joint and on both sides of the joint, the density directly in the joint is almost always lower than the asphalt mixture away from the joint. A low-density joint area is an ‘Achilles heel’ for otherwise well-constructed pavements since it may allow the rapid ingress of water and debris into the pavement and may lead to subsequent disintegration and premature failure (Figure 1).


In the absence of echelon paving, three major construction processes are recommended (see Ref 1-5) to achieve adequate density at the joints: butt joint, wedge joint, and cut (and remove) joint. There is also a recommendation for compacting the first 150 mm of the existing cold lane while compacting the adjacent hot lane (5, Figure 2). In the case of a wedge joint, during the paving of the hot lane, the heated material will cover the sloped joint that can be easily compacted. It is believed that the roller can increase the density of the joint while compacting the hot side and that the overlapping hot side material on the cold wedge of the joint would create a higher density area. From the traffic safety point of view, the wedge joint is a better option compared to a butt joint with a sharp vertical drop, especially when the hot side is not paved immediately after the cold side. This is especially safer for motorcycles. Other recommendations include heating the joint and/or, and tack coating/sealing the joint before paving the hot side (5).


Field research studies conducted over the years with different configurations have yielded mixed results in terms of density and performance. For example, some studies have shown that the wedge joint is better (6) (Toepel, 2003), while some have indicated good performance with sealed joints (7). Concerns have been expressed regarding the ability to compact the sloped portion of the wedge joint. Effective design (vertical or wedge) and construction of longitudinal joints remain an ongoing topic of research because of two reasons. First, the failure of pavements at joints continues to be a major problem in HMA pavements, and second, a lack of modeling/simulation work that could provide a rational basis for joint construction methods. Most of the existing work is based on field or laboratory experiments (e.g., 8).

This article explores the temperature increase in the cold butt joint during the paving of the hot side. Is it high enough for effective compaction during the compaction of the hot side? Is the temperature increase greater for a wedge joint?

Modeling And Analysis
Thermal modeling was conducted with Mecway (9) software in the following manner:

Model joints for thermal analysis

• Apply relevant material thermal properties, boundary, and thermal loading conditions
• Simulate the transient thermal condition for a time period expected during construction
• Analyze the variation of temperature with time at different areas of the joints

In finite element analysis, a numerical technique is utilized to represent the object as a combination of relatively simple shapes (elements) that are connected at nodes (mesh). The material properties are assigned to the different elements, and the loading and constraints are applied. The different elements are related to each other through physics-based laws, such as heat transfer principles in thermal analysis, where a single temperature can be assigned to each node. The model is solved, and the mesh can be refined as desired by checking the solution through several iterations.The heat flux is calculated across the mesh from the input initial boundary condition data, and the updated temperatures are output.

Both butt and wedge joints were modeled in this analysis. The model conditions included a wind speed of 16 kmph (4.5 m/s), mix laydown temperature of 140oC, and an ambient temperature (air and existing base) of 25oC. A 450 mm section of the pavement (with a thickness of 50 mm) was modeled with the hot side paving along a cold side joint, as shown in Figure 3. In the case of the wedge joint, the right side is the cold side, whereas the left side is the hot side. The loading conditions and restraints consisted of the initial temperature of the various sections, convection at the open edges, and no heat flux along the edges of the base/existing HMA. The last condition was specified because the thermal conductivity of HMA is very low (1.6 W/m/K), and the boundary for the no heat flux was made at a distance that is sufficiently away from the model sections where no change in temperature was noted during the transient thermal analysis. The convective heat transfer coefficient (H) was calculated from the following formula (10,11):


where,Taverage = average of the surface and air temperatures, ºC, U = wind speed in m/s, Tsurface = temperature of surface, oC or oK, and Tair = temperature of air, oC or oK.


The thermal properties used in the model consisted of a thermal conductivity of 1.6 W/m/K, and a heat capacity of 1,200 J/kg/K (12). Thermal properties have been found to be insensitive to temperature within the range that has been used in this study. There is no difference in thermal properties between the different sections of the pavement.

Results
Figures 4 and 5 show the results of thermal analyses of the butt and the wedge joints, respectively. As the material of the hot lane is laid down at 140oC, the heat from the mix starts to warm up the joint. With time, the heat flows towards the cold lane, because of its relatively low temperature (25oC). Within a time of 30 minutes, the heat can be seen to penetrate to 50-60 mm of the joint (see point A in Figure 4). However, with time, due to the conduction and convection of heat (wind above the mix and the cold lane) the temperature decreases. As a result, only about 12.5 mm (see point B in Figure 4) of the joint is heated to a temperature > 90oC within the first 5 minutes of laydown. For the remaining heated area, the temperature reaches a maximum of approximately 78oC and then starts decreasing over time.


Figure 4: Butt Joint: Temperature at selected times after the placement of the hot side HMA; Hot Side HMA laydown temperature, 140oC, ambient temperature and temperature of existing base and cold side, 25oC; wind speed, 4.5 m/s (10 miles per hour); Thickness of the surface layer is 50 mm

Figure 5: Wedge Joint: Temperature at selected times after the placement of the hot side HMA; Hot Side HMA laydown temperature, 140oC, ambient temperature and temperature of existing base and cold side, 25oC; wind speed, 4.5 m/s (10 miles per hour); Thickness of the new layer is 50 mm

In the case of the wedge joint, there is a greater penetration of heat into the joint because HMA is placed over the wedge allowing higher temperatures to go farther into the cold side. Approximately 188 mm of the joint is heated up by the mix in the hot lane (see point A in Figure 5). Out of this area, a significant part of it, approximately 150 mm (see point B) is heated to a temperature > 90oC. This is significant since the relatively high temperature enables effective compaction at the interface of the sloped joint during the compaction of the hot lane. From Figures 4 and 5 it can be inferred that rolling 150 mm into the joint while compacting the hot lane is more effective in the case of a wedge joint than in a butt joint.

The following conclusions and recommendations are made from this study:

• In a butt joint, the placement of HMA in the hot lane cannot be expected to cause a rise in the temperature along the cold lane/joint area to a high enough level for effective roller compaction.
• In a wedge joint, the placement of the HMA on the hot lane side can be expected to cause a rise in the temperature of the bottom part of the wedge from the cold lane side to a level that would allow effective compaction.
• Therefore, a wedge configuration is recommended to obtain better density at the joint.
• Modeling and analysis, especially for field conditions (range of temperature) should be conducted to develop practical guidelines for joint construction.

References

1. Kandhal, P. S., and S. Rao. Evaluation of Longitudinal Joint Construction Techniques for Asphalt Pavements. Transportation Research Record 1469, TRB, National Research Council, Washington, DC, 1994.
2. Kandhal, P. S., and R. B. Mallick. A Study of Longitudinal Joint Construction Techniques in HMA Pavements. Transportation Record 1543, TRB, National Research Council, Washington, DC, 1996.
3. Brock, J. D., and T. Skinner. Longitudinal Joints: Problems & Solutions. NAPA Quality Improvement Series 121, December 1997.
4. Kandhal, P. S., T. L. Ramirez, and P. M. Ingram. Evaluation of Eight Longitudinal Joint Construction Techniques for Asphalt Pavements in Pennsylvania. Transportation Research Record 1813, TRB, National Research Council, Washington, DC, 2002.
5. Mallick, R. B., P. S. Kandhal, R. Ahlrich, and S. Parker. Improved Performance of Longitudinal Joints on Asphalt Airfield. Final report to Federal Aviation Administration (FAA), Project 04-05: Airfield Asphalt Pavement Technology Program (AAPTP), July 2007.
6. Toepel, A. Evaluation o f T e c h n i q u e s for Asphaltic Pavement Longitudinal Joint Construction. Wisconsin DOT Final Report WI-08-03, November 2003.
7. Daniel, J. S. and W. L. Real. Field Trial of an Infrared Joint Heater to Improve Longitudinal Joint Performance in New Hampshire. Presented at the Annual Meeting of the Transportation Research Board, Washington, DC. January 2006.
8. Chen, J., B. Huang, Y. Li, and X. Shu. Laboratory Evaluation of Effects of Joint Heater on Longitudinal Joint. GeoShanghai International Conference 2010.
9. Mecway Limited, Mecway, New Zealand, 2018.
10. Dempsey, B. J. A Heat Transfer Model for Evaluating Frost Action and Temperature Related Effects in Multilayered pavement Systems. In Highway Research Record 342, HRB, National Research Council, Washington, DC, 1970, pp. 39-56.
11. Solaimanian, M. and T. Kennedy. Predicting Maximum Pavement Surface Temperature Using Maximum Air Temperature and Hourly Solar Radiation. Transportation Research Record 1417, Transportation Research Board, Washington, DC, pp 1-11, 1993.
12. Chen, B., L. Rockett, and R. B. Mallick. Laboratory Investigation of Temperature Profiles and Thermal Properties of Asphalt Pavements with Different Subsurface Layers. Journal of AAPT, Volume 77, 2008.