COMPARISON OF MARSHALL AND SUPERPAVE DESIGN METHODS, EVALUATION OF WHEEL TRACKING TEST OF ASPHALT MIXTURES DESIGNED BY BOTH METHODS

Transcription

1 250 COMPARISON OF MARSHALL AND SUPERPAVE DESIGN METHODS, EVALUATION OF WHEEL TRACKING TEST OF ASPHALT MIXTURES DESIGNED BY BOTH METHODS Dipl. Ing. Dr. techn. Michal Varaus Brno University of Technology, Czech Republic Abstract The research project SHRP and the system Superpave have brought a new way of asphalt mixtures design. In European countries however the Marshall design method is still preferred. In the frame of the Czech-Austrian research between the Vienna University of Technology, Institute for Road Construction and Maintenance and the Brno University of Technology, Departments of Roads, asphalt concrete (AC) mixtures were designed by both methods. In case of the design of ACmixtures the validity of the restricted zone, which was defined by Superpave in the allowed sieve curves range, was investigated. The restricted zone should be avoided according to Superpave, as not to design mixtures susceptible to permanent deformation. Three different sieve curves (above the restricted zone, in the restricted zone and below) were combined with 3 different bituminous binders (no modified, EVA-modified and SBS-modified). The resulting bituminous binder optima achieved by both methods were compared. The designed mixtures were then tested by performance related tests as wheel tracking test and some of them by fatigue test to find out their behaviour under different conditions. The article gives an overview of the test results and the comparison of both methods. 1. Introduction Asphalt mixtures have been known as an ideal material for pavement construction for a long time. The layers of flexible pavements constructed by these asphalt mixtures are durable and can be maintained in an easy way. These convenient properties can be also kept in case of increasing heavy traffic intensities by the optimalization of various technologies, the composition of the mixtures and the configuration of particular pavement construction layers. However the intensities of heavy traffic have been increasing in the recent times so dramatically, that the technologists, dealing with the asphalt mixture design, are challenged to solve the following issue: the old practice based on routine procedures is nowadays not always satisfactory. If the asphalt mixture should be used in the future even in greater extent than up to now, it is necessary to use the experience and know-how from other countries and to apply practically new findings and approaches. 2. SHRP and SUPERPAVE The above mentioned development of the heavy traffic intensities and the bad condition of the pavements in the USA initiated in the middle 80s a research project SHRP (Strategic Highway

2 Performance Testing and Evaluation of Bituminous Materials 251 Research Project), which should improve by a focused research the state of the US-pavements. The final product of the SHRP asphalt research project is a new system called Superpave (Superior Performing Asphalt Pavements). Superpave represents an improved system for specifying asphalt binders and mineral aggregates, developing asphalt mixture design and analyzing and establishing pavement performance prediction (1). The system includes an asphalt binder specification, a hot mix asphalt design and analysis system. The Superpave binder specification and mix design include various test equipments, test methods and criteria. The unique feature of the Superpave system is that it is a performance-based specification system. The test have direct relationships to the field performance. 2.1 SHRP-mix design specifications The Superpave asphalt binder tests measure physical properties that can be related directly to field performance by engineering principles. The Superpave binder tests are also conducted at temperatures that are encountered by in-service pavements. The SHRP has brought also a new approach to the design of asphalt mixtures and the prediction and testing of their long-term performance. The basis of the SHRP-mix design is to select suitable aggregates and asphalt binders. When these are selected by appropriate tests, mix design is carried out using the Superpave Gyratory compactor, which is thought to simulate the compaction of the rollers and the following heavy traffic postcompaction better than the commonly used Marshall compaction hammer, by kneading the mixture in a cylindrical mould. The aggregates are blended in several combinations, so that different trial gradations are achieved. Superpave contains control points between which the gradations have to be. In addition Superpave defines also a restricted zone in the fine part gradation. It is recommended that the designed gradation does not go through this restricted zone because of the lower resistance to permanent deformations. Each of the gradations is then tested by the gyratory compactor using calculated initial binder content (there are guidelines in Superpave how to estimate this content). Before compaction of the specimen the asphalt mixture has to run through the short term ageing procedure, where the loose mixture is stored in an oven for 2 hours at the compaction temperature. The results from the compaction of each gradation are evaluated. The Superpave volumetric design procedure contains specifications with regard to the gyratory compaction curve, which are air voids, voids in mineral aggregate (VMA), voids filled with asphalt binder (VFA) and filler to bitumen ratio. Figure 1: SHRP-sieve curve specifications for continuously graded asphalt mixtures

3 252 The gradation that shows the best properties with respect to the prescribed criteria is chosen. For definition of the optimum binder content, the chosen gradation is then tested further in the gyratory compacter using 4 various binder contents. The binder content at which the mixture achieves the air voids of 4 % is the optimal binder content. 2.2 Own research carried out according to the Marshall- and SHRP-mix design specifications The aim of our research in the mix design was to design various mixtures of asphalt concrete (with respect to the restricted zone) by the common Marshall design method and by Superpave using the gyratory compactor. The optimal asphalt binder contents achieved by both methods were determined. With these optimal binder contents asphalt slabs were compacted which were then tested in the wheel tracking tester. Some of these asphalt mixtures were also tested at fatigue test. The following questions should have been answered : - What are the optimal asphalt binder contents found by both methods for the same gradation and the same asphalt binder? - How resistant are the designed mixtures against the creation of permanent deformations? - Is it really necessary to respect the restricted zone by the design of AC mixtures regarding the resistance against permanent deformations? - How are the mixtures resistant to fatigue? Three different gradations were used for the AC design with the max. aggregate size of 11 mm - one passing above the restricted zone ( ZONE), the second going through and the third going below the restricted zone ( ZONE). These three gradations were combined with three different asphalt binders (one nonmodified binder and two modified binders), so that 9 different mixtures were designed by each method. The nonmodified binder AP 65 was obtained from the Czech refinery in Pardubice as well as the EVA modified binder MOFALTPLAST 65. The SBS modified binder STARFALT was a product of the refinery in Schwechat (Austria). To evaluate the optimal binder contents for these 9 AC-mixtures, both the Marshall method and the Superpave method were employed (see Table 1). The Marshall test specimens were compacted by 2x75 blows. The Gyratory test specimen were compacted by the PINE Gyratory compactor to max. 204 gyrations, simulating the process of postcompaction by heavy traffic. The parameters for deriving of the optimal binder content were evaluated at 126 gyrations. The number of gyrations was derived from the average seven-day maximum air temperature and the expected traffic volume of heavy lorries/24 hours in both directions. The heavy traffic was recalculated on 8 t ESAL (Equivalent Standard Axle Load) used in Superpave. Table 1: Materials and Design Methods used AC11 Binder type for Binder type for MARSHALL-DESIGN SUPERPAVE-DESIGN ZONE AP 65 Mofaltplast Starfalt AP 65 Mofaltplast Starfalt Sieve Curve ZONE AP 65 Mofaltplast Starfalt AP 65 Mofaltplast Starfalt ZONE AP 65 Mofaltplast Starfalt AP 65 Mofaltplast Starfalt The designed mixtures (altogether 18) were then tested by the wheel tracking tester according to the last draft of the pren Bituminous mixtures, Test methods for hot mix asphalt Part 22: Wheel tracking. The wheel tracking tester was of the German type modified by the use of a rubber wheel

4 Performance Testing and Evaluation of Bituminous Materials 253 instead of the steel wheel prescribed in the German standard. The test was running in a 50 C water bath and the rut depths were measured on the track of 50 mm from the middle of the slab to both sides. The number of applied cycles was Results 3. 1 Results of the mix design The results of the mix design are summarized in the following tables. In the head of the table 2 for Marshall design results requirements of both Czech and Austrian standards are mentioned together with the common requirement printed in bold. The requirements for VFA and VMA are put into brackets, as they are only recommended by the Austrian standards. In the two last columns there are optimal bitumen contents (B opt ) and the bulk specific gravities of the compacted asphalt mixtures A. Table 2: Results of the Marshall mix design MARSHALL Air Voids 3-4 % CZ = 3-5 A = 2-4 Stability min. 9 kn CZ = 9 A = 9 Flow 30-40x10-1 mm CZ = A = VFA (75-80 %) (A= 75-80) VMA (min. 17%) (A=17) B opt % A kg/m 3 AP-A 3,3 13, ,0 15,0 5, MO-A 3,0 13, ,6 14,7 5, ST-A 3,4 16, ,5 15,1 5, AP-Z 3,4 14, ,1 15,5 5, MO-Z 3,1 12, ,6 15,2 5, ST-Z 3,6 13, ,1 15,7 5, AP-B 3,7 15, ,2 17,0 5, MO-B 3,9 14, ,3 17,1 5, ST-B 4,0 14, ,7 17,2 5,7 2361

5 254 Table 3: Results of the Superpave mix design SUPERPAVE Air Voids 4 % VFA % VMA min. 14,5% B opt % A kg/m 3 AP-A 4,0 70,3 13,5 4, MO-A 4,0 70,8 13,7 4, ST-A 4,0 70,7 13,7 4, AP-Z 4,0 72,9 14,7 4, MO-Z 4,0 72,6 14,5 4, ST-Z 4,0 72,9 14,8 4, AP-B 4,0 73,7 15,2 4, MO-B 4,0 73,7 15,2 4, ST-B 4,0 74,1 15,4 4, Commentary to the tables 2,3 and 4 : Bitumen Sieve curve AP = AP 65 A = above the zone MO = Mofaltplast 65 Z = in the zone ST = Starfalt B = below the zone Commentary on the achieved results: Gyratory compactor compacts more intensively than Marshall hammer with 75 blows. The optimal binder contents from Superpave are lower than from Marshall. This is caused partly due to the higher compaction effort by Gyratory compactor and different requirements for air voids in both methods. Superpave is able to eliminate the dry mixtures with very low binder content above the restricted zone by the use of the VMA-criterion, Marshall not (this criterion is only recommended). The VMA- and VFA-criteria, which are used in both design methods can not be compared directly, as they does not mean volumetrically exactly the same (different definition). Not all sieve curves going below the restricted zone have to be accepted by Superpave. Superpave simulates the postcompaction of asphalt mixtures by heavy traffic (air voids criterion at max. gyrations: Va > 2 %) and can evaluate the compactibility of asphalt mixtures directly during compaction.

6 Performance Testing and Evaluation of Bituminous Materials Results of the wheel tracking test Table 4: Results of the wheel tracking test evaluated by the PDR criterion from the min. to the max. value MARSHALL SUPERPAVE PDR PDR AC11 AC11 [%] [%] MO-A 1,0 MO-A 0,9 AP-A 1,1 ST-A 1,0 ST-A 1,4 ST-Z 1,2 ST-B 1,5 AP-A 1,3 ST-Z 1,5 ST-B 1,3 AP-Z 1,6 MO-Z 1,6 MO-B 1,8 MO-B 1,6 MO-Z 1,9 AP-B 2,1 AP-B 1,9 AP-Z 2,7 The results of the wheel tracking test for the asphalt mixtures designed by both methods are summarized in the table 4. There are two criteria for the evaluation according to the pren : the WTS (wheel tracking slope) in mm rut depth /10 3 cycles and the proportional rut depth PDR in %. Commentary on the achieved results: Comparing the identity mixtures (same bituminous binder, same sieve curve), which were designed by different design methods (Marshall, Superpave), the Superpave mixtures achieved in 6 of 9 cases better results. The best results achieved the mixtures with the sieve curves going above the restricted zone. This can be explained by the low binder content and therefore also low temperature sensitivity. These mixtures were also tested by the fatigue test. In comparison with the mixtures with the sieve curve going below the restricted zone they achieved however much lower fatigue lives. The validity of the restricted zone was not confirmed, as the asphalt mixtures with the sieve curve going through the restricted zone have no significantly higher rut depths. The same tendency was observed also at the WesTrack-measurements (Federal Highway Administration test facility in Nevada). If we compare the asphalt mixtures with the same sieve curves but different bituminous binders, the mixtures with modified bitumens achieved as expected better results than those with no modified. 4. Summary The goal of the described research was the comparison of two different design methods for asphalt mixtures the common used Marshall design method and the Superpave design method developed in the frame of SHRP. The comparison was carried out on a limited amount of asphalt mixtures and as a following the evaluation of the susceptibility of the designed mixtures to permanent deformations was investigated.

7 256 It was found out that both design methods have as the most important criteria for optimizing the binder content the air voids characteristic of the compacted asphalt mixture. The other used volumetric characteristics as the VMA- and VFA-criteria (obligatory only in Superpave) which can further precise the design and eliminate for example the too dry mixtures are in both methods based on different definitions and can not be therefore compared directly. The compaction effort in the Superpave design method is derived from the climate conditions and the expected traffic volume, whereas the Marshall uses the fix number of blows. The gyratory compactor compacted the asphalt mixtures more intensively than the Marshall hammer with 75 blows, what has brought lower resulting optimal binder contents for Superpave mixtures. Superpave simulates also the postcompaction by heavy traffic and is able to evaluate the compactibility of the mixtures directly. The comparison of permanent deformation susceptibility of mixtures designed by both methods brought more resistant Superpave asphalt mixtures. The most resistant mixtures in both design methods were those with the sieve curve above the restricted zone and with modified bitumen. The validity of the restricted zone was not confirmed. The carried out research has shown the differences in the design methods, which have a significant influence on the resulting mixtures compositions and also on their permanent deformation resistance. The Superpave design method comes with the deriving of the compaction effort from the real in situ conditions (climate, traffic), but the binder optima seem to be relatively low for the European experience. For a more detailed comparison of asphalt mixtures characteristics a complex evaluation of asphalt mixtures would be needed based not only on the wheel tracking test, but also on other performance related tests as fatigue test (partly already carried out) and low temperature test. 5. References 1. ASPHALT INSTITUTE, Superpave Level 1, Mix Design, Superpave Series No.2 (SP-2) (Maryland, 1998). 2. THE ASSOCIATION OF ASPHALT PAVING TECHNOLOGISTS, Asphalt Paving Technology, Volumes 58-68, Sheridan Book Co., (Chelsea, Michigan ). 3. ASCHENBRENER, T. and Mac KEAN,C., Factors That Affect the Voids in the Mineral Aggregate of Hot-Mix Asphalt, Transportation Research Record 1469, TRB, National Research Council, (Washington, D.C., 1994, pp.1-8). 4. HUBER, G.A. and SHULER, T.S., Providing Sufficient Void Space for Asphalt Cement: Relationship of Mineral Aggregate Voids and Aggregate Gradation, ASTM STP 1147, Richard C. Meiniger, editor, American Society for Testing and Materials (Philadelphia, 1992). 5. LITZKA, J., STROBL, R. and VARAUS, M., Vergleich von Versuchen zum Verformungsverhalten von Asphalt, Final report on the cooperation project 20p7, (Wien, Brno, 1999). 6. VARAUS, M., Entwurf von Asphaltgemischen, Methodenvergleich und Beurteilung des Verformungsverhaltens, Thesis (Wien, 2001) 7. pren , Test Methods for Hot Mix Asphalt Wheel Tracking Test, (CEN, Brussels, 1997).