47626(405)11

Comparisons between New and Traditional NDT Devices and Control Methods for Construction Quality of Highway Subgrade P. Xu, F.M. Wang, X.L. Li and Y.C. Cai Department of Transportation Engineering, Zhengzhou University, Zhengzhou, 450002 China; plian127@163.com ABSTRACT: Subgrade is the base of highway structures, so to evaluate its construction quality accurately is very important to ensure the required strength of the whole highway. Several subgrade segments under construction of Tai-Ao highway in China were selected as engineering examples, the point-to-point test mode was adopted, and then the deflections, dielectric values and compaction degrees were in-situ tested with the new and traditional NDT (Non-destructive testing) devices or control methods such as FWD (dynamic falling weight deflectometer), BB (static Benkenlman beam), PFWD (portable falling weight deflectometer), GPR (ground penetrating radar), special dielectric constant device, and sand cone method, and FWD, PFWD and GPR are proved to be the rapid and precise devices for quality control of the highway subgrade. Keywords: Subgrade; Quality control; Dynamic falling weight deflectometer; Portable falling weight deflectometer; Ground penetrating radar 1. INTRODUCTION There needs higher requirement for the strength and integrity of every highway subgrade, if it is not rolled or compacted enough, its bearing capacity would be lower and could not reach the design, then some diseases would occur such as subsidence damage and void under the pavement, might not only decrease the life of the subgrade and pavement, but also reduce the driving comfort and safety, and even cause traffic accidents. So it is very important to control the quality of the subgrades both completed and under construction, and by now some new and traditional NDT devices have been developed and studied in the world such as FWD, BB, PFWD and GPR. FWD is a device capable of applying dynamic loads to the pavement surface, similar in magnitude and duration to that of a single heavy moving wheel loads, and the response of the pavement system is measured in terms of vertical deformation, or deflection, over a given area using seismometers. Stiffness and modulus of the pavement layer can be back-calculated with FEM method based on the tested deflection basin, and the dynamic stiffness matrix about FWD process can be simulated with fast calculation algorithms. (Hakim et al, 1999; Kima and Mun, 2008; Picoux et al, 2009) 83Geotechnical Special Publication No. 215 © ASCE 2011 PFWD has a mass that less than FWD, and uses an accelerometers to determine the deflection. PFWD is an ideal device for quality control during compaction monitoring because it enables a rapid stiffness assessment of individual pavement layers, and one of the most important factors concerning the PFWD modulus is the size of the PFWD loading plate. (Lin et al, 2006) GPR is an accepted electromagnetic evaluation technique used for the transportation infrastructure and a variety of other applications: In the area of subgrade soil evaluation GPR techniques have been used to nondestructively identify soil type, to estimate the thickness of overburden and to evaluate the compressibility and frost susceptibility of subgrade soil; In road structure surveys, GPR has been used to measure layer thickness, to detect subsurface defects and to evaluate base course quality; In quality control surveys, GPR techniques have been used for thickness measurements, to estimate air void content of asphalt surfaces and to detect mix segregation. (Saarenketo and Scullion, 2000) But the synthesized comparisons about new and traditional NDT devices were less given by now. 2. ENGINEERING EXAMPLES IN TAI-AO HIGHWAY Tai-Ao Highway is the 7th longitudinal highway of 13 longitudinal and 7 transverse highways in China. The total length of Tai-Ao Highway is 2237 km, and starts from Taiyuan City of Shanxi Province, pass through Henan Province, Hunan Province, Guangdong Province, and ends in Macao Special Administrative Region. Several segments of Tai-Ao Highway subgrade under construction in Nanyang City of Henan Province are selected as engineering examples, which include: (1) 15th segment, which has bottom-to-top structure of natural ground, 80cm height filled soils with compaction degree of 90%, and 20cm height filled soils with compaction degree of 93%, (2) 1st zone of 16th segment, which has bottom-to-top structure of natural ground that excavated 20cm and has compaction degree of 96%, (3) 2nd zone of 16th segment, which has bottom-to-top structure of natural ground, 70cm height filled soils with compaction degree of 94%, and 20cm height filled soils with compaction degree of 96%,(4) K69+620 ～ K69+720 of 3rd road bed layers of 17th segment, which is excavated in large scale and has been strengthened and stabilized with lime content of 5%, (5) K72+473～K72+620 of 3rd road bed layers of 18th segment, which is excavated in small scale and has been strengthened and stabilized with lime content of 4%, (6) JK4+103.1～JK4+260 of 3rd road bed layers of 19th segment, which is filled and has been strengthened and stabilized with lime content of 3%. At the same place of the selected segments, GPR, PFWD, FWD, BB, dielectric constants, and compaction degrees were totally or partly in-situ tested in precedence, and then some comparisons and relationships were made and analyzed. 84Geotechnical Special Publication No. 215 © ASCE 2011 Contemporary Topics on Testing, Modeling, and Case Studies of Geomaterials, Pavements, and Tunnels D ow nl oa de d fro m a sc el ib ra ry .o rg b y Ch in a U ni ve rs ity o f M in in g & T ec hn ol og y on 0 1/ 08 /1 4. C op yr ig ht A SC E. F or p er so na l u se o nl y; a ll rig ht s r es er ve d. 3. COMPARISONS BETWEEN NEW AND TRADITIONAL NDT DEVICES AND CONTROL METHODS 3.1 COMPARISONS OF DELECTIONS TESTED BY BETWEEN FWD AND BB In China, deflection tested by BB is one of the standard index to check and accept the highway surface or subgrade, which needs a truck with standard weight, two Benkenlman beams and two dial indicators, it usually takes about 60 minutes for one complete BB test, and the result is affected by the personal error more or less, so to find a new, rapid and precise substitute device or method is imperative. Deflections were tested by BB and FWD of 1 ton, 3 ton and 5 ton individually at 23 places on the subgrade segments, and the comparisons were made between them, as shown in fig. 1. From fig. 1 we can find that the correlations are better between deflections tested by BB and FWD (5 ton and 3 ton) and the correlation coefficient of R2 is 0.90 and 0.86 individually, but worse between BB and FWD (1 ton) with R2 of 0.78, that is to say that the weight of FWD and the truck weight used for BB is more nearer, the correlation are more better between their deflection results, and FWD with weight that exceeds 3 ton is a much ideal substitute device for BB. R2 = 0.78 R2 = 0.86 R2 = 0.90 0 500 1000 1500 2000 2500 3000 1300 1600 1900 2200 2500 2800 3100 Deflection (BB) / μm D ef le ct io n (F W D ) / μ m FWD(1t) FWD(3t) FWD(5t) FIG. 1. Comparisons of deflections between BB and FWD of different weights. 85Geotechnical Special Publication No. 215 © ASCE 2011 Contemporary Topics on Testing, Modeling, and Case Studies of Geomaterials, Pavements, and Tunnels D ow nl oa de d fro m a sc el ib ra ry .o rg b y Ch in a U ni ve rs ity o f M in in g & T ec hn ol og y on 0 1/ 08 /1 4. C op yr ig ht A SC E. F or p er so na l u se o nl y; a ll rig ht s r es er ve d. 3.2 COMPARISONS OF DELECTIONS TESTED BY FWD AND PFWD The falling weight of PFWD is only 10 kg, which is less than FWD, while the FWD with higher falling weight usually has compaction action on the tested soils, and 232 deflections were tested in the sequence of PFWD, FWD 1 ton, FWD 3 ton and FWD 5 ton, and the results were compared, as shown in fig. 2. From fig. 2 we can find that the correlation coefficients of R2 decrease when the falling weights of FWD increase, which are 0.91, 0.89 and 0.81 corresponding to 1 ton, 3 ton and 5 ton. y = 0.68 x0.98 R2 = 0.91 0 200 400 600 800 1000 0 300 600 900 1200 1500 1800 Deflection (FWD 1 t) / μm D ef le ct io n (P FW D ) / μ m (a) PFWD and FWD (1 ton) 86Geotechnical Special Publication No. 215 © ASCE 2011 Contemporary Topics on Testing, Modeling, and Case Studies of Geomaterials, Pavements, and Tunnels D ow nl oa de d fro m a sc el ib ra ry .o rg b y Ch in a U ni ve rs ity o f M in in g & T ec hn ol og y on 0 1/ 08 /1 4. C op yr ig ht A SC E. F or p er so na l u se o nl y; a ll rig ht s r es er ve d. y = 0.50 x0.94 R2 = 0.89 0 200 400 600 800 1000 0 500 1000 1500 2000 2500 3000 Deflection (FWD 3 t) / μm D ef le ct io n (P FW D ) / μ m (b) PFWD and FWD (3 ton) y = 0.62 x0.86 R2 = 0.81 0 200 400 600 800 1000 0 500 1000 1500 2000 2500 3000 3500 Deflection (FWD 5 t) / μm D ef le ct io n (P FW D ) / μ m (c) PFWD and FWD (5 ton) FIG. 2. Comparisons of deflections between PFWD and FWD of different weights. 87Geotechnical Special Publication No. 215 © ASCE 2011 3.3 COMPARISONS OF DIELECTRIC CONSTANTS BACK-CALCULATED BY GPR AND TESTED BY SPECIAL DEVICE The dielectric constants are varied in two kinds of soils with different water tables and porosities, which can be reflected by the GPR images (Saarenketo, 1998), and the dielectric constants of the road structures from the ground to subgrade and surface can be back-calculated from GPR images like modulus from FWD deflection basins. The dielectric constants of every layer can be obtained by the analogy functions as follows 0 0 2 1 0 0 1 1 m r r m R A A R A A ε ε + += =− − (1) 2 0 1 1 3 2 2 2 1 0 1 1 1 1 1 m m r r r m m A A A AR R A A A A ε ε ε ⎛ ⎞ ⎛ ⎞− +⎜ ⎟ ⎜ ⎟+ ⎝ ⎠ ⎝ ⎠= =− ⎛ ⎞ ⎛ ⎞− −⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠ (2) where, εr1, εr2 and εr3 are the dielectric constants of 1st, 2nd and 3rd layer individually, A0, A1 and A2 are the GPR wave amplitudes reflected from the interfaces between the 1st, 2nd and 3rd layer in sequence, and R0 or R1 are the ratios of A0 or A1 to the total reflection amplitude of Am. 72 dielectric constants of the subgrade were back-calculated from the GPR images, and then were compared with those tested by the special device, as shown in fig. 3. From fig. 3 we can find that the correlation coefficients of R2 about back-calculated and tested dielectric constants is 0.81, that is to say that GPR can reflect the dielectric characteristics of the subgrade in large degree. y = 1.17 x - 1.22 R2 = 0.81 6 7 8 9 10 11 12 13 6 7 8 9 10 11 12 Dielectric constant (in-situ test) D ie le ct ric c on st an t ( G PR ) 88Geotechnical Special Publication No. 215 © ASCE 2011 Contemporary Topics on Testing, Modeling, and Case Studies of Geomaterials, Pavements, and Tunnels D ow nl oa de d fro m a sc el ib ra ry .o rg b y Ch in a U ni ve rs ity o f M in in g & T ec hn ol og y on 0 1/ 08 /1 4. C op yr ig ht A SC E. F or p er so na l u se o nl y; a ll rig ht s r es er ve d. FIG. 3. Comparisons of back-calculated and tested dielectric constants. 3.4 RELATIONSHIPS BETWEEN DIELECTRIC CONSTANT BACK-CALCULATED BY GPR AND COMPACTION DEGREE The dielectric constant is influenced by the soil water table and porosity, so better relationship may exist between the compaction degree and dielectric constant. After GPR image was tested, sand cone method was adopted at the same place, and water contents and natural density were tested indoors, and then the compaction degree was obtained according to the maximum dry density. The relationship between the dielectric constants back-calculated from GPR images and tested compaction degree were shown in fig. 4. From fig. 4 we can find that the correlation coefficients of R2 about back-calculated dielectric constants and compaction degrees are different in two segments, 0.86 in filled subgrade, while 0.73 in excavated subgrade, the former is much better than the latter. The soil of filled subgrade can be influenced more than that of excavated subgrade at the same rolling or compacting action, for the latter has been compacted by the above layers under long-term weight stress, on the other hand, the filled subgrade has obvious interface from the under original ground about soil properties such as porosity and water table, and so the compaction degree of filled subgrade has better relationship with the dielectric constants, and suitable for GPR rapid detection. y = 0.01 x3 - 0.87 x2 + 12.51 x + 43.52 R2 = 0.86 85 87 89 91 93 95 6 7 8 9 10 11 12 Dielectric constant (GPR) C om pa ct io n de gr ee / % (a) 16th segment of filled subgrade 89Geotechnical Special Publication No. 215 © ASCE 2011 y = 0.09 x3 - 3.40 x2 + 41.53 x - 67.74 R2 = 0.73 90 92 94 96 98 7.5 8.5 9.5 10.5 11.5 12.5 Dielectric constant (GPR) C om pa ct io n de gr ee / % (b) 17th segment of excavated subgrade FIG. 4. Relationships between back-calculated dielectric constants and compaction degrees 4. CONCLUSIONS The test results by several new NDT devices such as FWD, PFWD and GPR were compared with those of traditional devices and methods, and some important conclusions are drawn out, which are that: (1) the weight of FWD and the truck wheel weight used for BB are more nearer, the tested deflection between them have more better correlation, and FWD with weight that exceeds 3 ton can be used to replace BB for quality control; (2) PFWD and FWD have the similar correlation with FWD and BB, and PFWD is only suitable for deflection test of FWD with weight not larger than 1 ton; (3) dielectric constants can be back-calculated from the GPR images in large degree; (4) GPR is suitable for the rapid detection of compaction degree of filled subgrade, but not for the excavated subgrade that has been compacted by the above soils weight more or less; (5) FWD, PFWD and GPR are faster test services than those traditional methods such as BB and cone sand cone method, and constructor can make next decision as fast as possible based on the feedback test results and conclusions. REFERENCES Hakim, B. A., Cheung, L. W. and Armitage, R. J. (1999). "Use of FWD data for prediction of bonding between pavement layers." Int. J. Pavement Eng., Vol. 1 (1): 49-59. 90Geotechnical Special Publication No. 215 © ASCE 2011 Kima, J. M. and Mun S. (2008). "Fast spectral analysis of an axisymmetric layered structure." Mech. Res. Commun., Vol. 35(4): 222-228 Picoux, B., Ayadi, A. E. and Petit, C. (2009). "Dynamic response of a flexible pavement submitted by impulsive loading." Soil Dyn. Earthq. Engng., Vol. 29 (5): 845-854. Lin, D. F., Liau, C. C. and Lin J. D. (2006). "Factors affecting portable falling weight deflectometer measurements." J. Geotechnical & Geoenv. Engrg., Vol. 132 (6): 804-808. Saarenketo,T. and Scullion, T. (2000). "Road evaluation with ground penetrating radar." J. Appl. Geophys., Vol. 43 (2): 119-138. Saarenketo T. (1998). "Electrical properties of water in clay and silty soils." J. Appl. Geophys., Vol. 40 (1): 73-88. 91Geotechnical Special Publication No. 215 © ASCE 2011