FIELD TESTS ON BORED PILES SUBJECT TO AXIAL AND OBLIQUE PULL By Nabil F. Ismael, 1 Member, ASCE

Transcription

1 FIELD TESTS ON BORED PILES SUBJECT TO AXIAL AND OBLIQUE PULL By Nabil F. Ismael, 1 Member, ASCE ABSTRACT: The ultimate oblique pull of bored piles in sand was examined by a field testing program. Tests included axial uplift tests, lateral load tests, and oblique tests at an angle of 30 with the pile axis. The piles used were 101 mm in diameter by 1.5 m long and were installed in predominantly fine, medium-dense sand. Lateral failure occurred in the oblique tests of the present program; however, both the axial and lateral failure must be considered in practical design. The maximum oblique pull was 74% of the axial pull since the smaller lateral capacity was reached first. Both the lateral and axial capacities were predicted with reasonable accuracy. The critical angle was defined as the angle of inclination of the load from the pile axis at which the failure mode changes from axial to lateral failure. It can be determined if the soil strength parameters and the pile type and geometry are known. INTRODUCTION The primary loading on guyed structures is oblique upward pull. Several types of foundations can be employed for supporting this load, including bored or driven piles, soil anchors (such as the multihelix anchors), or earth grillages. The behavior of deep foundations under nonaxial loads is not well understood and has been examined for driven piles by laboratory model tests only (Yoshimi 1964; Broms 1965; Ramanathan and Ganapathy 1970; Awad and Petrasovits 1968; Meyerhof 1973). No field tests have been reported that examine the capacity of piles subject to oblique pull or that compare the results with the proposed analytical predictions (Broms 1965; Poulos and Davis 1980). The effect of load cycling under this type of loading has not been investigated. With improved construction techniques, bored piles are being used increasingly as foundations for guyed structures in both cohesive and cohesionless soils. Recently, a field testing program was carried out in Salmiya, Kuwait, to examine the effect of load obliquity on the pull-out capacity of vertical bored piles installed in cohesionless soils. The program included axial uplift tests, lateral load tests, and oblique pull tests, where the load was inclined 30 with the vertical. This paper presents and analyzes the test results and compares them with the analytical procedures proposed by Broms (1965). The purpose is to improve existing design methods and to gain a clear understanding of bored pile behavior under oblique pull. The nonlinear method of analysis was employed for prediction of the piles subject to lateral loading. The method is based on solving the beam column equation with the modulus of soil reaction given by a set of p-y curves along the pile (Matlock 1970; Reese et al. 1974). 'Assoc. Prof., Civ. Engrg. Dept., Kuwait Univ., P.O. Box 5969, Safat 13060, Kuwait. Note. Discussion open until April 1, To extend the closing date one month, a written request must be filed with the ASCE Manager of Journals. The manuscript for this paper was submitted for review and possible publication on September 26, This paper is part of the Journal of Geotechnical Engineering, Vol. 115, No. 11, November, ASCE, ISSN /89/ /$ $.15 per page. Paper No

2 SOIL CONDITIONS The soil profile was explored by one-auger boring, where samples were recovered and standard penetration tests were performed at 0.5-m intervals. A dynamic cone test was performed in the vicinity of the borehole. A summary of the soil conditions is given in Fig. 1, which shows the soil stratification, moisture content, SPT, dynamic cone penetration results, and the soil composition along the depth of the borehole. The soil profile consists of compact-to-dense, fine-to-coarse, windblown calcareous cohesionless sand which can be classified as SP» SM, according to the unified soil classification system. This sand, which consists predominantly of fine silica sand, has a low carbonate content of 10-20% (Ismael et al. 1986). Standard penetration test values recorded were 26 near the surface, and averaged 25 over the pile depth of 1.5 m from the ground surface. Employing the empirical correlation between the SPT values and the angle of shearing resistance (j) (Peck et al. 1974), the anglecj> is approximately 35 along the pile shafts. PROGRAM AND PROCEDURE OF FIELD TESTING The bored piles employed were 101 mm in diameter by 1.5 m long. They were first installed by drilling an auger hole, placing a 12.5-mm reinforcing steel rod in the center, and filling the hole with sand-cement grout. The steel rod was bent to form a hook that protruded above the finished pile section for connection with the pulling assembly in the case of the pull-out tests. The piles were left for 28 days prior to testing. Fig. 2 shows a layout of the piles installed and the field tests carried out. These include two uplift tests carried out vertically along the pile axis, two oblique tests carried out at an angle of 30 with the vertical axis, and two lateral loading tests where the loads were applied at an angle of 90 from the pile axis. The small size of the test piles was due to a limited reaction of the employed test setup. The setup for the uplift tests was previously described by Ismael et al. (1989). Briefly, a four-meter rigid steel beam consisting of two channels and resting on steel pads was employed in conjunction with a 196-kN calibrated hollow-cylinder jack. The pile was pulled by a central 25-mm threaded rod that passed through the jack and tied by means of a washer and nuts as shown in Fig. 3(a). The load was applied in equal increments of about 10% of the capacity. Each increment was maintained for a minimum period of 15 min and until all movements had ceased. The loads were recorded by a calibrated pressure gage. Figs. 3(b and c) show the test setup for the oblique tests and the lateral load tests, respectively. In the oblique tests, the same equipment was employed, except that the pulling rod was placed in a position inclined 30 with the vertical. The pile was pulled by a pulling rod passing through an inclined blocking, a hollow steel plate, and the hollow ram jack. Dial gages connected to inclined boards were placed to measure the vertical and horizontal displacements of the pile during the test. The lateral loading tests were carried out by placing a rigid pipe column and a 98-kN jack in line between the two piles at ground surface. Dial gages were placed next to each pile [Fig. 3(c)]. Equal increments of load were applied and maintained for a minimum period of 15 min. All tests were 1589

3 Soil Composition /. of Total r 6? -M.sand -Fine sand 5ilt.&clay/X 1 Depth (m) Soil Description Water Content SPT&CPT, Blows/30 m CD o Compact to Dense Calcareous Grey Fine to Coarse Sand with a trace of silt (SP-SM) Calcareous silty fine Med. Sand, trace of clay (SM - SO FIG. 1. Soil Conditions at Test Site

4 Q O u ifl TesU P FIG mm Pulling Rod_ jon: Pressure Gage Pump. (~\ Oblique Tests ~"^--\pi es 0.10m diamx 1.5m long Q Q Lateral Tests L 1.22 m.1 Layout of Field Load Tests Blocking^. r Steel Plate / AJack.Hollow Ram Jack _2 Channels 75mm Back to flack A 'Cm& \*3HR ' ^-Pressure Gage \_Dial Gage -101 mm Dlam pile -Dial Gage FIG. 3. Test Setup for Field Load Tests: (a) Uplift Test; (b) Oblique Test; (c) Lateral 1591

5 continued until failure had occurred and a displacement of 25 mm was recorded. ANALYSIS OF TEST RESULTS Test results are plotted in Fig. 4 in the form of load versus displacement curves. To facilitate comparison between the different curves, the displacement in the direction of the applied load is used. Thus, in the case of the oblique test, the component of the resultant displacement in the direction of the applied load is employed. The failure loads were determined by the slopetangent method. They are marked by vertical arrows in Fig. 4. The failure displacement was taken as that corresponding to the failure load. A summary of test results is given in Table 1, where the failure load and the maximum load reached are given with the corresponding displacements. The failure loads will be analyzed in the following sections, followed by an examination of the load displacement response under the different types of loading. Lateral Loading Tests For short, free-headed piles, Broms (1964) assumed that the lateral earth pressure that develops at failure is equal to three times the passive Rankine Load 32 FIG. 4. Load Displacement Curves 1592

6 Type of test (1) Uplift Oblique Lateral TABLE 1. Summary of Field Test Results Failure load" (kn) (2) Failure displacement (mm) (3) Maximum load (kn) (4) ' "Failure load was determined using the slope tang ent method. Maximum displacement (mm) (5) earth pressure. The ultimate lateral resistance can then be calculated from equilibrium requirements by the following formula: QL = 0.5y'DL 2 K p where Q L = ultimate lateral resistance; y' = effective unit weight of the soil; D = width or diameter of the pile; L = depth of the pile; and K p = coefficient of passive earth pressure as calculated by the Rankine earth pressure theory. For a cohesionless soil, the coefficient K p can be calculated from K P = 1 + sin 4>' 1 sin cj)' in which <)>' = the effective angle of shearing resistance of the soil. In a subsequent revision, Broms (1965) assumed that the lateral pressure increases linearly with depth and that this earth pressure is equal to five times the passive Rankine earth pressure. Accordingly, the lateral resistance of the pile can be computed as QL = ~ y'dl 2 K p Substituting into Eqs. 1 and 3, 1,766 kg/m 3 for 7' (measured), m for D, 1.5 m for L, and 3.69 for K p as calculated from Eq. 2 for <)> = 35, yields Q L = 7.3 kn and 12.3 kn, respectively. Comparing the preceding values with the results of the field tests obtained from Fig. 4, it is evident that good agreement exists between the maximum load of kn, which was reached at large displacement, and the calculated value of 12.3 kn from Eq. 3. The lower value of 7.3 kn calculated using Eq. 1 is close to the 8.8 kn designated herein as the failure load obtained by the slope-tangent method. Thus, Eqs. 1 and 3 give lower and upper bound values of the capacity under lateral loading. Uplift Load Tests Assuming a linear increase of the lateral earth pressure over the pile length, the uplift capacity can be obtained from the following formula: g = 0.5y'L 2 ndk u tan <j>' + W (4) where Q u = the uplift capacity; K u = the coefficient of earth pressure in uplift; and W = the effective weight of the pile. Substituting the failure load, 22.6 kn for Q u as obtained from Fig. 4, and (1) (2) (3) 1593

7 the same values for 7', L, D, and ' as in the lateral load tests, the coefficient K u is calculated to be While this value corresponds to that which has been measured from laboratory tests on piles in compacted dense sand (Broms 1965), it is significantly larger than the measured values for fullsize, large diameter piers in cohesionless medium dense sand (Ismael and Klym 1979), which ranged from The average skin friction along the pile shaft was calculated from where f s = the average shaft resistance. Substituting the preceding values for Q u, D, L, and W, f s is calculated to be kn/m 2. This is nearly half the corresponding shaft resistance for a very dense soils of similar deposit (Ismael and Al-Sanad 1986). Oblique Load Tests A close examination of Fig. 4 and Table 1 indicates that failure under oblique loading occurred at a failure load of 16.7 kn, with a corresponding displacement of 5.1 mm. Failure was caused herein by lateral failure since at this load the horizontal component was 8.3 kn, which is close to the 8.8 kn designated as the failure load from lateral load tests. The vertical component of 14.4 kn is well below the 22.6-kN failure load determined from the uplift tests. The failure load under oblique loading was only 74% of the corresponding value under axial pull. The load capacity in this case is a function of both the lateral resistance and the vertical uplift capacity of the pile (Poulos and Davis 1980). If the applied load deviates slightly from the axial direction, failure will occur in uplift. However, as the load inclination increases, a large component will be directed in the horizontal direction and lateral failure will occur. Both modes of failure must be considered in the analysis of oblique loading. If it can be assumed that the axial capacity of the pile is independent of the lateral component of the load and that the lateral capacity is independent of the axial component of the load, the inclined capacity Q can be calculated as the smaller of the following two values: Q = Q u sec 8, for axial failure (6a) Q = QL cosec 8, for lateral failure (6b) where 8 = the angle between the inclined load and the vertical axis. As previously stated, lateral failure occurred herein, and it is possible to confirm that the lateral capacity was little affected by the axial component of the load during the oblique test. The vertical and horizontal components of the oblique load are plotted against the corresponding vertical and horizontal displacements in Fig. 5. The results of the lateral load test shown by dotted lines in Fig. 5 are parallel and close to the curve giving the horizontal component of the oblique load. The difference in lateral capacity is only 0.98 kn. Unlike the preceding discussion concerning lateral capacity, the lateral component of the load appears to have some influence on the axial capacity in cohesionless soils. Assuming that the axial capacity of the pile is the same whether the pile is vertical or inclined provided that the inclination is not 1594

8 Load KN FIG. 5. Components of Oblique Load versus Corresponding Displacements large, laboratory tests by Yoshimi (1964) indicated that the ultimate pull initially increased with inclination and then decreased when the angle 8 exceeded 30. Broms (1965) analyzed the Yoshimi data and concluded that the axial capacity increased due to the change in lateral pressure caused by inclination of the applied load. This is shown in Fig. 6. Accordingly, Q u in Eq. 6 becomes H Q 5kp /gd T= 2.5 kpfg D R= 0.67 gt L FIG. 6. Earth Pressure Distribution at Oblique Pull (Broms 1965): (a) Deflection; (b) and (c) Earth Pressure Distribution; (d) Assumed Pressure Distribution 1595

9 Qu = Quo + A2 in which Q uo = the pull-out resistance when the applied load acts along the pile axis; and AQ U = the increase in pull-out resistance caused by the two lateral forces T and R in Fig. 6. The increase in pull-out resistance is expressed by Afi = (T + R) tan <)>' (8) The increase in axial capacity due to the horizontal component of the inclined load has not been confirmed by field tests. Tests should be carried out at a small inclination from the vertical of, say, 8 = 10 to induce axial failure and therefore to check whether the axial pile capacity has increased. The accuracy of the proposed method of calculation (Broms 1965), could then be established. Assuming that the aforementioned increase in axial capacity is ignored, it is possible to determine the critical inclination of load at which failure changes from axial to lateral. From Eq. 6, axial failure will occur if Q L > Q sin 8 or Q L > Q u tan 8. Conversely, lateral failure will occur if Q L < Q tan 8. Thus, the occurrence of axial or lateral failure will depend on whether the ratio Q L /Q U is larger than or less than tan 8. However, this ratio can be determined analytically from Eqs. 1 and 4 as QL = osy'dl% = K Q u 0.5y'L\DKu tan ty rrk u tan <(>' in which the effective weight of the pile W has been deleted from the denominator. It should not be ignored if it constitutes a significant component. Substituting into Eq. 9, the values for the present field tests, as an example, are tan 8 = tan 30 = 0.577, and QjQ u = [3.69/TT(4.94) tan 35] = < tan 8, therefore lateral failure controls the capacity. The critical angle 8 C = \vn~\qjq^ = tan"' = Thus, axial failure will occur for the present pile and soil conditions, if the load inclination is less than The given example demonstrates that the mode of failure controlling design and the critical angle of inclination can be determined if the parameters K p, K u, and <j> are known with reasonable accuracy. K p and 4>' can be determined readily by triaxial compression or penetration tests. The selection of K requires judgment and knowledge of available field test measurements. For design purposes under axial or oblique loads, a reasonable factor of safety should be applied to the designated failure loads. Applying a factor of safety of three to the failure loads shown in Fig. 4 yields design loads in the elastic portion of the load-displacement curves. Load-Displacement Response It may be noted by inspection of Fig. 4 that the load-displacement curve for axial loading is nearly linear over the first portion prior to failure. For oblique tests the curve becomes highly nonlinear. It appears that the nonlinearity increases with the angle of load inclination 8. For lateral loading (8 = 90 ), the response is nonlinear from the beginning. Field lateral loading tests on drilled shafts in clay (Ismael and Klym 1978) and in sand (Ismael and Klym 1981) have indicated similar nonlinear behavior and poor agree- (7) 1596

10 ment between test results and the linear analytical prediction based on the subgrade reaction theory even at low stress levels. In contrast, more accurate predictions of lateral deflections were obtained using the nonlinear method of analysis (Ismael and Klym 1978a, 1978b). The method is based on the solution of the differential equation d*y d 2 y EI + P x -{ = E s y (10) ax ax where y = the lateral deflection; x = distance along the pile; EI = bending stiffness; P x = axial load; and E s = soil modulus. For the solution of Eq. 10, it is necessary to predict a set of p-y curves along the pile. Once the p-y curves are determined, Eq. 10 can be solved numerically to yield deflection, rotation, bending moment, shear, and soil reaction for any sustainable load. The p-y curves employed herein were based on the method proposed by Matlock (1970) and the recommendations of Reese et al. (1974), with the ultimate soil resistance P taken for the sand of the test site. A comparison between the analysis prediction of lateral displacement and the results of the lateral loading test is shown in Fig. 4. Clearly, reasonable prediction of the lateral displacement was achieved with this method of analysis. RECOMMENDATIONS FOR FUTURE WORK Field tests on bored and drive piles are needed to examine both the axial and lateral failure modes. The piles should be loaded at different angles of inclination. The increase in axial capacity, if any, due to oblique loading should be determined by tests in which the angle of inclination 8 is small. The results can then be compared with the proposed analysis. The effect of load cycling should be investigated, since most guyed tower structures are subject to repeated or cyclic loading. Based on limited field tests just completed at the site of the present tests, it was found that load cycling reduces the ultimate oblique pull by 10-15%. The work should also be extended to include oblique or nonaxial compression, which occurs frequently due to lateral earth pressure and wind loads on various structures. CONCLUSIONS A program of field tests was carried out on bored piles in sand to examine the pull-out capacity under oblique loads. Tests included an uplift test, lateral loading tests, and oblique tests at an angle of 30 with the vertical. Based on test results, the following conclusions are made: 1. The pull-out capacity under oblique loads is controlled by either the lateral resistance or the axial capacity of the pile, whichever is smaller. 2. For the present field tests, the lateral resistance controlled the magnitude of the ultimate oblique pull, which was 74% of the axial pull-out capacity. 3. The magnitude of the axial and lateral capacity was calculated with reasonable accuracy from the estimated soil parameters and the coefficient of lateral earth pressure in uplift. 4. The critical angle is defined herein as the angle of inclination between the 1597

11 applied tension load and the pile axis at which failure changes from axial failure to lateral failure. It depends on the soil properties and the pile type and geometry, and it can be estimated. 5. The load displacement curves of bored piles subject to oblique pull are nonlinear, particularly as the angle of load inclination 8 increases. For lateral loading (8 = 90 ), the load displacement response is nonlinear from the beginning. Prediction of the pile response under lateral loading using the p-y method is in agreement with test results. ACKNOWLEDGMENTS The writer is grateful to M. A. Mollah and O. H. Al Khalidi of the Government Laboratories and Testing Station, Ministry of Public Works, Kuwait, for providing assistance in the field installation and load testing of the piles. APPENDIX. REFERENCES Awad, A., and Petrasovits, G. (1968). "Considerations on the bearing capacity of vertical and batter piles subjected to forces acting in different directions." Proc, 3rd Budapest Conf. on Soil Mechanics and Foundation Engineering, Budapest, Hungary, Broms, B. B. (1964). "Lateral resistance of piles in cohesionless soils." J. Soil Mech. Found. Div., ASCE, 90(3), Broms, B. B. (1965). Discussion of "Piles in cohesionless soil subject to oblique pull," by Y. Yoshimi. J. Soil Mech. Found. Div., ASCE, 91(4), Ismael, N. F., Al-Khalidi, O., and Mollah, M. A. (1989). "Upgrading footings in sand with bored piles." J. Geotech. Engrg., ASCE (in press). Ismael, N. F., and Al-Sanad, H. (1986). "Uplift capacity of bored piles in calcareous soils." J. Geotech. Engrg., ASCE, 112(10), Ismael, N. F., et al. (1986). "A study of the properties of surface soils in Kuwait." Geotech. Engrg., 17, Ismael, N. F., and Klym, T. W. (1978a). "Behavior of rigid piers in layered cohesive soils." /. Geotech. Engrg. Div., ASCE, 104(8), Ismael, N. F., and Klym, T. W. (1978b). "Uplift and lateral behavior of earth grillage foundations." Proc, IEEE Summer Meeting, Paper No. A , Los Angeles, Calif. Ismael, N. F., and Klym, T. W. (1979). "Uplift and bearing capacity of short piers in sand." J. Geotech. Engrg. Div., ASCE, 105(5), Ismael, N. F., and Klym, T, W. (1981). "Lateral capacity of augered tower foundations in sand." IEEE Trans. Power Apparatus Systems, PAS 100(6), Matlock, H. (1970). "Correlations for design of laterally loaded piles in soft clay." Proc, Second Annual Offshore Tech. Conf., Houston, Tex. Meyerhof, G. G. (1973). "The uplift capacity of foundations under oblique loads." Can. Geotech. J., 10, Peck, R. B., Hanson, W. E., and Thornburn, T. H. (1974). Foundation engineering. lohn Wiley and Sons, New York, N.Y., 310. Poulos, H. G., and Davis, E. H. (1980). Pile foundation analysis and design. John Wiley and Sons, New York, N.Y., Ramanathan, T. S., and Ganapathy Aiyer, P. (1970). "Pullout resistance of piles in sand." J. Indian Nat. Soc Soil Mech. Found. Engrg., 9(2), Reese, L. C, Cox, W. R., and Koop, R. D. (1974). "Analysis of laterally loaded piles in sand." Proc, Sixth Annual Offshore Tech. Conf., 2, Houston, Tex., Yoshimi, Y. (1964). "Piles in cohesionless soil subject to oblique pull." J. Soil Mech. Found. Div., ASCE, 90(6),

12 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具