1、外文原文 where: D1, m = Fracture coefficients. t = Loading time in seconds. The correlation used for the D1 fracture parameter is: )l o g (923.1)l o g (0 1 0 3.2)l o g (7 9 5 7.00 1 3 0 6.05 2 4 1.8)l o g ( 1 R T F Oa AVFAVTD (3.3.56) where: T = Test temperature ( C ) (i.e., 0, -10, and 20 C ) Va= Air v
2、oids (%) VFA = Void filled with asphalt (%) = 100 Vbeff = effective binder content, % ARTFO = Intercept of binder Viscosity-Temperature relationship for the RTFOcondition For the m parameter, the correlation used is: TP e nVTm P e nV F AA 4 6 0 5.0770 0 2 4 7.00 1 1 2 6.0 0 0 1 6 8 3.004596.000185.0
3、1628.1 77 (3.3.57) where: T = Test temperature ( C ) (i.e., 0, -10, and 20 C ) Va= Air voids (%) VFA = Void filled with asphalt (%) Pen77 = Penetration at 77 F = )72 97 3.2(100694.257288.8 1 1 7 75013.29010 V T SA A = Intercept of binder Viscosity-Temperature relationship VTS = Slope of binder Visco
4、sity-Temperature relationship The outcome of the m value has been set to a lower limit of 0.01. For all three levels of data input, the Design Guide procedure requires tensile strength at10oC. Level 1 and 2 requires actual test data for tensile strength, whereas, level 3 hasbuilt in typical values b
5、ased upon the asphalt mix properties, similar to creep compliancevalues. For Levels 1 and 2, the strength test is conducted on the same test specimen that is usedto establish the 100 second creep test. After the creep test is completed, failure isevaluated using a load rate of 51 mm per minute. The
6、reported tensile strengths are theaverage of the three replicates used. In the recommended test protocol, a special procedure is utilized to determine the failureload achieved during the indirect tensile test. This is again an important modificationbecause the failure load has been found to be less
7、than the maximum load that thespecimen can undergo. Thus, once the instant of failure is found (approach uses thedeflection measurement difference), the failure load can be defined and the tensilestrength computed from: iSXft DCPS 2 (3.3.58) where: Pf = Failure load. Csx= Correction factor (previous
8、ly defined). t = Specimen thickness. D = Specimen diameter. For Level 3 analysis, the tensile strength at 10 oC was also correlated with mixtureproperties as with the creep compliance fracture parameters. The best indicators were theair voids, the void filled with asphalt content, the Penetration at
9、 77 oF, and the A interceptof the binder temperature-viscosity relationship for the RTFO condition. The followingcorrelation is used in the analysis: (3.3.59) where, the tensile strength (St) is given in psi. The outcome of the equation was set to alower limit of 100 psi. Based on the above correlat
10、ions for Level 3 analysis, default values for creep complianceand tensile strength for several binders used in this analysis were calculated. )l o g (2 9 6.2 0 3 9)l o g (71.4 0 57 0 4.05 9 2.1 2 23 0 4.00 1 6.1 1 47 1 2.7 4 1 67722R T F OaatAP e nVFAVFAVVSA summaryof the results is presented in App
11、endix HH. Other inputs required are test duration, thickness of the asphalt layer, and coefficient ofthermal contraction. For the coefficient of thermal contraction, the program has twooptions. In one instance, mixture VMA, aggregate thermal contraction and asphaltthermal contraction values are requ
12、ired for estimating the mix coefficient of thermalcontraction or the user can directly enter the value for the mix thermal coefficient. Step 2: Development of the master creep compliance curve Enhanced data analysis techniques (through the new program MASTER are claimed toprovide accurate evaluation
13、s of the time-temperature shift factor (aT) and creepcompliance model statistical fitting techniques through Prony and Power Model forms, aswell as the development of the creep compliance master curve (CCMC). In this analysis, a special data trimming technique is used to provide the best estimateof
14、the mix response parameters. If three specimen replicates are used, each havingmeasurements on both faces, a total of six strain (compliance) versus time curves can bedeveloped. The trimming approach recommends elimination of the extreme high and lowreadings and then utilization of the averaging of
15、the remaining four measurement straintimeresponses. Finally, the new procedure no longer requires the binder stiffness to extend the mixturecreep compliance results to longer loading time for the construction of the CCMC. Theresults of the CCMC analysis are fit to a Prony series defined by: vNi iieDDD )1()0()(1（ 3.3.60） Tat（ 3.3.61） Where: = reduced time. t = real time. aT= temperature shift factor.
