TRANSPORT RESEARCH LABORATORY TITLE by The use of hot surface treatment to rehabilitate cracked asphaltic concrete surfacings in Malaysia J Emby, C Jones and M S Mustafa' Oversea's Centre Transport Research Labo'ratory ,- Crowtho~rne Berkshire U~nite'd ~Kingdom IA . X 101 - 1 - 11 A Emby,J, C R Jones and M S Mustafa, 1992. The use of hot surface treatment to rehabilitate cracked asphaltic concrete surfacings in Malaysia. In: Proceedings of the Seventh REAAAA Conference, Singapore, 22-26 June 1992. THE USE OF HOT SURFACE TREATMENT TO REHABILITATE CRACKED ASPHALTIC CONCREUE SURFACIN4GS IN MAlAYSIA, Johari Emiby Training and Research Institute, JK Malaysia Colin Jones Transport and Road Research L-aboratory UK Ir. Moharned Shaffi Mustafa Training and Research Institute, JKR Malaysia ABSTRACT In tropical climates the binder in the top few millimetres of dense bituminous surfacings suffers from rapid age hardening causing the viscosity of the bitumen to increase. The material becomes brittle and this results in premature cracking starting at the surface caused by environmental or traffic induced strains. If left untreated these cracks will propagate down through the bituminous layer decreasing its effective modulus and allowing the ingress of water resulting in more serious pavement deterioration. One meth od of arresting this type of cracking before more serious failure occurs is to heat the surfacing and then either remove or recycle the cracked layer before the cracks can reach any substantial depth. This paper describes the first two years performance of a full-scale hot surface treatment trial on a heavily trafficked urban highway in Malaysia. The performance of the trial was monitored under a joint research programme between the Training and Research Institute of the Jabatan Kerja Raya (JKR), Malaysia and the Transport and Road Research Laboratory (TRRL) of the UK- Three different hot surface techniques were used in the trial and their performance compared to that of a control section. The results showed that although these techniques have been designed to treat only shallow surface cracking, minor modfification to one of the methods may make it suitable for solving the more serious problem of reflection cracking in thin overlays. 538 1 1.0 INTRODUCTI'ON The primary cause of failure of thin asphaltic concrete overlays used to maintain flexible road pavements in Malaysia is the occurrence of reflection cracking which is initiated by cracks in the previous surfacing. These cracks can occur as soon as 2-3 months after the construction Of 40mmn thick overlays, their rate of propagation being dependent upon the severity of cracking before overlay, the pavement deflection and the commercial traffic volume. Viljoen et al, 1987, have reported that the rate of propagation of reflection cracks in asphalt surfacings is dependent on the number of load repetitions and the movement of the crack due to the deflection of the road pavement under the load. As shown in Figure 1 the movement at the crack will impose a localised strain on a layer placed above, eventually causing the material to fracture and a crack to propagate upwards through thernew overlay. Rolt et al, 1986, have shown that another common mode of -failure of both new asphaltic concrete surfacings and overlays in tropical -environments is cracking which starts at the top of the surfacing and propagates downwards. This type of cracking is induced by the oxidation and consequent hardening of the bitumen in the top 2-3mmn of the material (Smith et al, 1990) making it brittle and thus susceptible to 'top-down' cracking. This type of cracking, although exacerbated by the material properties of the top 2-3mm, can be caused by a variety of stresses. Hot surface treatment, a technique recently introduced into Malaysia, can be used to rehabilitate road surfacings which suffer from this latter type of 'top-down' cracking. This paper compares the first two years performance of three different forms of hot surface treatment with that of a control section on a heavily trafficked urban highway in Kuiala Lumpur, Malaysia. 2.0 EXPERIMENTAL DESIGN The four trial sections are located adjacent to one another in the slow lane of a three lane highway. Each section is approximately 180 metres long and all four were overlaid with a common thickness of 60mm of asphaltic concrete after the surface treatment was completed. The layout of the sections and the method of treatment prior to overlay are shown in Figure 2. 2.1 Surface Treatment The different methods of surface treatment are described below: Section 1: Heat, scarify and recompact. In this section the road surface was heated to approximately 140C, then scarified to a depth of 25mm and immediately recompacted. Section 2: Heat, remove and patch. In this section the road surface was heated to approximately 140 0C scarified and removed, and then a new 25mm layer of nominal 13mmn asphaltic concrete was layed. Section 3: Control section. The surface of this section was untreated prior to overlay. Section 4: Heat and pave. In this section the surface was heated to approximately 140 0C and immediately paved with a 20mm layer of a nominal 13mmn asphaltic concrete. 539
11i 3.0 PRECONSTRUCTION MEASUREMENT. The site was marked out by 10 metre chainages and the measurement of rutting, deflection and cracking made with reference to these chainages. 3.1 Rutting Rut depth measurements were taken using a 2 metre straight edge and calibrated wedge. The values given in Table 1 show that all four sections were structurally strong with only low values having developed, with the exception of one chainage in section 2, since its construction 15 years previously. The only exception to this was one chainage in section 2 which showed a rut depth of 19 mm. 3.2 Deflection Deflection measu rements were made with the Falling Weight Deflectometer (FWD) and the results corrected to a standard pressure of 700 N/in .The results given in Table 1 show that sections 1 and 4 had similar values to the control section (Section 3). 'Section 2 had slightly higher values. Table 1: Rutting and Deflection Values Prior to Construction 3.3 Cracking The cracking in each 10 metre length was assessed by its intensity and assessed visually using the following classification:- Intensity area. The intensity was o -No cracking 1 -Single crack 2 -Two or more cracks -not connected 3 -Two or more cracks -interconnected 4 -Crocodile cracking 5 -Crocodile cracking with spa lng The longitudinal extent and the transverse location of the cracking was also recorded and from this its area calculated. Table 2 gives the total area of pavement that was cracked prior to construction and its distribution by intensity. 540 Section No. ______Rutting (mm) Deflection (mm x 10O3) No. of tests Mean Range No. of tests Mean Range 1 12 7 4-11 12 389 283-494 2 22 7 3-19 22 526 288-866 3 18 5 3-8 18 40825-7 4 16 -S3-6 - 16 41739-2 Table 2: Cracking Prior to Construction Section No Area cracked -Crack intensity distribution (per cent) ______ (per cent) 1 2 3 4 1 35.4 0 0 75 12.5 12.5 2 27.7 0 2.3 57.3 40.4 0 3 21.7 0 11.4 88.6 0 0 4 23.1 0 2.3 84.1 6.8 6.8 The results show that the majority of cracking at the site was intensity 3 and this was common throughout all four sections. It was also noted that by the time the cracks had progressed to intensity 3 they were becoming badly spalled. 3.4 Crack Depth After recording the intensity and area of the cracking a series of 100mm cores were taken to establish the depth of the cracks. The results from these tests are shown in Figure 3. The Figure illustrates that although intensity 1 cracks were confined to the top 35mm of the surfacing, by the time they had progressed to intensity 3 they had propagated down through the surfacing to an average depth of 115mm. 3:5 Traffic The site is situated in the slow lane of a particularly heavily trafficked three lane highway in Kuala Lumpur. A classified traffic count showed that the commercial vehicle flow in the slow lane was 6100 vehicles per day. 4.0 CONSTRUCTION The heating of the road surface was by direct flame using propane burners. The average surface temperature achieved was 140'C decreasing to 650C at a depth of 25mm. The relevant construction details for each of the sections is given in Table 3. Table 3: Construction Details Section Surface Temp. at Pen. of Depth of Thickness of No temp. after rolling/ recovered recompacted/pat overlay above heating before paving bitumen chi layer road surface _________ (OC) (OC) (mm X 10,1) (mm) (mm) 1 136 110' 37' 25 60 2 - - - 25 60 3 - - - 60 4 144 78 -- 20+402. 3 I. 2 3 Recompacted layer 20mm of 13mm nominal A.C. plus 40mm of 20mm nominal A.C. No tack coat was applied prior to the 20mm layer. 541 Wheel moement . Opening of cracks doClosing of cracks O-J- Opening of cracks Fig. 1 Crack movement caused by the position of the crack in the deflection bowl (after Villoen et al. 1987) Section 1 Section 2 Section 3 Section 4 i 2SmT~~~i..j ,..2 Recampacted aye ~~~~~~~3 Patch..layer. s Qmy jj Fig.2 Schematic cross section of experimental site 180 160 Mean thickness of bituminous surfacing 140 120 .v100 f 80 CL 2 3 4 Crack intensity Fig.3 Relation between crack depth and intensity '542 5.0 PERFORMA&NCE The rutting and cracking of the sections -has now been monitored -for a period of two years. The increase in rutting has been small with the mean values for the sections ranging between 2-4mm. There has however been a differential progression in the development of cracking. 5.1 Cracking Cracking was noted in the new overlay as early as 6 months after construction and cores taken at that time showed them to be reflection cracks caused by cracks in the underlying layer. Figure 4 illustrates the increase in the area of cracking of the different sections since construction. The analysis has been restricted to those areas which had an initial crack intensity 3 because this type of cracking was the most extensive and common to all the four sections