VERTICAL AND LATERAL LOAD TESTING OF DRILLED SHAFTS SOCKETED INTO ROCK AT THE POMEROY-MASON BRIDGE OVER THE OHIO RIVER

Transcription

1 VERTICAL AND LATERAL LOAD TESTING OF DRILLED SHAFTS SOCKETED INTO ROCK AT THE POMEROY-MASON BRIDGE OVER THE OHIO RIVER Jamal Nusairat 1 Project Manager Research E.L. Robinson Engineering of Ohio Co., 6000 Memorial Drive, Dublin, OH jamal@elrobinson.com Tel: (614) Fax: (614) Rick Engel Vice President E.L. Robinson Engineering of Ohio Co., 6000 Memorial Drive, Dublin, OH rengel@elrobinson.com Tel: (614) Fax: (614) Robert Y. Liang Professor Civil Engineering Department The University of Akron Akron, OH rliang@uakron.edu Tel: (330) Fax: (330) Ke Yang Graduate Research Assistant Department of Civil Engineering The University of Akron Akron, OH ky6@uakron.edu Tel: (330) Fax: (330) Number of words: 7380 Number of words in Abstract: Corresponding Author

2 2 VERTICAL AND LATERAL LOAD TESTING OF DRILLED SHAFTS SOCKETED INTO ROCK AT THE POMEROY-MASON BRIDGE OVER THE OHIO RIVER Jamal Nusairat, Rick Engel, Robert Y. Liang, and Ke Yang ABSTRACT: Drilled shafts socketed into rock have been widely used as a foundation for bridges to resist both vertical and lateral loads. Although numerous research efforts on rock socketed drilled shafts under vertical loads have been carried out, research on rock socketed shafts under lateral loads have been lacking. This paper presents the results of a testing program designed to validate the design capacity of the drilled shafts socketed into rock for both vertical and lateral loads. Testing was conducted on two 8 feet (2.44 m) diameter drilled shafts. The two shafts were instrumented with Osterberg load cells (O-Cell TM ) for vertical load testing, and fully instrumented for lateral load testing as detailed in the paper. The effect of the overburden soil around test shaft 2 was isolated by installing 11 feet diameter casing which was advanced to the top of rock. Test shaft 1 was terminated in shale while test shaft 2 was constructed 10 feet deeper to bear on hard siltstone bedrock. With a maximum applied load of 36,100 Kips (161 MN), the vertical load testing on shaft 2 broke the world record in In 2005, vertical O-Cell TM load tests at Seoul International Airport, Seoul, Korea broke the world record with the highest load in excess of 64,000 Kips (284.7 MN). The lateral load test was conducted after the vertical testing was completed, and the maximum load induced a movement of approximately 0.25 inches at the top of rock. The results of this load testing program verified the adequacy of the design assumptions, and can be used to verify the existing published analysis methods that are still in their early stage of development due to limited lateral load test data.

3 3 INTRODUCTION Drilled shafts socketed into rock have been widely used as a foundation for bridges to support both axial and lateral loads. Although numerous research efforts have been carried out on rock socketed drilled shafts under vertical loads and several documented vertical load tests on drilled shafts socketed into rock using the Osterberg Cell (O-Cell TM ) are available in the literature, the designers are still using low values for end bearing in rock compared to the allowable end bearing from the load testing, in addition, research efforts on rock socketed drilled shafts under lateral loads have been lacking. A review of literature suggests that very few existing field lateral load testing data are available on rock socketed drilled shafts. Furthermore, in many reported lateral load test data, there is a relatively thick layer of soil overlying the rock, which may limit the extent of interaction between the rock and the shaft. The few existing analysis methods for rock socketed drilled shafts under lateral loads, such as the methods by Carter and Kulhawy (1992), Reese (1997) and Zhang et al. (2000), could benefit from additional verification with actual field load test results. Carter and Kulhawy (1992) and Zhang et al. (2000) methods are either elastic or elasticperfectly plastic solution, which may not capture the actual nonlinearity of weak rocks. Reese (1997) proposed an interim p-y criterion for lateral analysis of drilled shafts in rock. However, this p-y criterion was formulated based on two field tests. It is desirable to further evaluate this criterion with additional lateral load test results. To improve the analysis methods for laterally loaded drilled shafts socketed in rock, there is a need for additional high quality lateral load test data. The main objectives of this paper are twofold: (a) to present the results of vertical load tests and a lateral load test on two fully instrumented drilled shafts (8 feet (2.44 m) in diameter, feet and feet (30.78 and 34.14) in length) socketed in shale, and (b) to evaluate the existing p-y criterion (Reese 1997) using the compiled load test data and the new lateral load test reported in this paper. TEST SITE CONDITIONS The Ohio Department of Transportation and the Federal Highway Administration supervised the design of the under construction bridge for the replacement of the U.S. Route 33 Bridge over the Ohio River between Pomeroy, Ohio and Mason, West Virginia. During the design stage, a decision was made to perform a testing program to ensure the adequacy of the design and to gather information for future designs. Two vertical load tests using the Osterberg Load cells and a lateral load test were carried out on two temporary drilled shafts constructed close to the West Virginia Tower. Two Borings S-9 and S-10 were advanced near the test drilled shafts from a floating platform. Water surface elevation of 540 feet (164.6 m) was observed. In Boring S-9, from the river bottom to a depth of 36 feet (11.86 m) (below the water surface), a layer of coarse and fine sand was encountered. This layer was described as grayish brown and orangish brown, wet and loose. Water contents varied from 20 to 24 percent. Underlying the aforementioned soil in S-9 and at the surface of S-10, gravel with sand was encountered and extended to depths of 56.9 and 57.2 feet (17.3 and 17.4 m). The gravel with sand was described as brown, wet and loose to medium dense. Water contents ranged from 13 to 20 percent. Top of bedrock was encountered at an approximate depth of 48 feet (14.6 m) below water surface. The recovered rock cores were described as shale with interbedded siltstone from

4 4 bedrock surface to depths of 75.8 feet (23.1 m) and 73.6 feet (22.4 m) below the water surface in S-9 and S-10, respectively. The average unconfined compressive strength of the shale layers was 65 TSF (6.2 MPa), and the average unconfined compressive strength of the siltstone layers was 350 TSF (33.5 MPa). Shale (mudstone) was observed in S-9 from 75.9 feet (23.1 m) to feet (31.4 m) and in S-10 from 73.6 feet (22.4 m) to feet (31.6 m) below the water surface. The average unconfined compressive strength of the mudstone layer was 2 TSF (0.2 MPa). Siltstone was recovered underlying the shale to a depth of feet (37.6 m) and feet (36.1 m) below the water surface in S-9 and S-10, respectively. The average unconfined compressive strength of the Siltstone layer was 350 TSF (MPa). Details of the two Borings are shown in Figure 1, in which q u is the unconfined compressive strength and E i is modulus of intact rock core. INSTRUMENTATION AND CONSTRUCTION OF THE TEST SHAFTS Test drilled shaft 1 was constructed approximately 120 feet (36.6 m) away from Boring S-10 and test shaft 2 was at about 95 feet (29 m) away from Boring S-9 as shown in Figure 2. Due to the variability of the site, the top elevations of bedrock at test shafts 1 and 2 were observed during construction at elevations feet (148.3 m) and feet (149.7 m), respectively. The ground surface elevation at test shaft 1 was 512 feet (156 m). The total depth of test shafts 1 and 2 were and feet (30.9 and 34.4 m), respectively. Rock socket lengths of test shafts 1 and 2 were 37.5 and 53.3 feet (11.4 and 16.2 m), respectively. Test shaft 1 was terminated in the weak shale layer to evaluate the end bearing of this type of rock, and was instrumented with one O-Cell with 6000 Tons (26.7 MN) capacity. The diameters of the bearing plate under the O-Cell was 3 feet (91.5 cm), while the top bearing plate diameter was 7-6 (2.29 m). The smaller diameter plate at the bottom of shaft 1 is for the purpose of trying to apply higher end bearing pressure. Test shaft 2 was terminated in the hard sandstone layer to evaluate the end bearing of this type of rock, and was instrumented with three O-Cells with a total capacity of Tons (80 MN). The top and bottom bearing plates for the O-Cell s were 7-6 (2.29 m). The unconfined compressive strength of concrete was 5120 psi (35.3 MPa). The diameter of the drilled shafts above bedrock was 8.5 feet (2.59 m) with a 1 inch (2.54 cm) thick casing. The drilled shafts bedrock socket was 8 feet (2.44 m) in diameter. The portion of the two drilled shafts above bedrock has a diameter of 8.5 feet (2.59 m) with a 1 inch (25.4 mm) thick casing. The test shafts were reinforced each with 28 #18 bars. In order to fully mobilize the rock-shaft interaction and isolate the lateral effect of overburden soils, an 11 feet (3.35 m) diameter casing was used to form a gap between test shaft 2 and the soils above bedrock to insure that all lateral forces were resisted only by bedrock. Both drilled shafts were instrumented with inclinometers for measuring the deflections along the length of shaft, two dial gages for measuring the deflections at the loading point, and vibrating wire strain gages shown in Figure 5. TEST SET-UP AND TEST PROCEDURE The test shafts were spaced 71-9 (21.8 m) center to center. Due to the large distance, the push apart testing setup was not feasible. A pull together using a 1.75 inch (44.5 mm) diameter grade 270 Dywidag rod and a hydraulic jack mounted at test shaft 1 was utilized as shown in Figure 3. The test shafts were constructed with an annulus 6 inch (152.4 mm) in diameter to run the rod through. A 24 x24 x1 (600 mm x 600 mm x 25 mm) bearing plates were welded to the steel

5 5 casing at each test shaft to distribute the load. The jack was mounted at test shaft 1 and nuts were used to lock the rod on both ends as shown in Figures 3 thru 5. The reference beams were supported on HP 14x73 piles driven to refusal. Four 42 inch (107 cm) diameter steel casings were driven around the H-piles to isolate the H-Piles supporting the reference beams from the river current as shown in Figure 4. The Osterberg load tests were performed on 8/27 and 8/28/2003 starting with test shaft 2 then test shaft 1 following the quick load test method (ASTM D1143), holding each successive load increment constant for four minutes by manually adjusting the O-Cell pressure. The loads applied by the O-Cells act in two directions, resisted by the side shear above the O-Cells and by base resistance below the O-Cells. After completion of the O-Cell testing, a 200 Ton (1779 KN) hollow jack was used to apply the lateral load to the drilled shafts. A 200 Ton (1779 KN) load cell was installed between shaft 1 and the jack to measure the actual applied lateral loads. The load was applied in increments of 12.5 Tons (111 KN) up to a maximum load of Tons (1223 KN). Before applying the first increment of load, initial readings from all the instruments were recorded. At each load increment, the strain readings, the dial gages readings, the tiltmeters readings, and the inclinometers readings were collected after the movement stabilizes or after a waiting period of 10 minutes. VERTICAL LOAD TEST RESULTS At the time of testing Shaft 2, the concrete unconfined compressive strength was 5,115 psi (35.3 MPa). The shaft was successfully loaded to a combined side shear and end bearing of more than 36,100 kips (161 MN). For a top loading of 5,800 kips (25.8 MN), the adjusted test data indicate this shaft would settle approximately 0.25 inches (6.4 mm) of which 0.14 inches (3.3 mm) is estimated elastic compression. For a top loading of 20,000 kips (89.0 MN), the adjusted test data indicate this shaft would settle approximately 1.0 inches (25.4 mm) of which 0.47 inches (11.9 mm) is estimated elastic compression. Figure 6 shows the load-movement curve, Figure 7 shows the equivalent top load-movement curves, and Figure 8 shows the calculated load distribution curves from strain gages along the drilled shaft rock socket. At the time of testing Shaft 1, the concrete unconfined compressive strength was 4,205 psi (29.0 MPa).The Shaft was successfully loaded to a combined side shear and end bearing of 1,850 kips (2.39 MN). For a top loading of 1,850 kips (2.39 MN), the adjusted test data indicate this shaft settled approximately 6.4 inches (162.6 mm). Figure 9 shows the load-movement curve. COMPARISON BETWEEN VERTICAL LOAD TEST RESULTS AND DESIGN VALUES The drilled shafts were designed for end bearing and side friction according to the procedures recommended in FHWA-IF Drilled Shaft: Construction Procedures and design methods. The unconfined compressive strength (q u ) in which more than half of the test results are greater for each rock horizon was selected. The side resistance for axial loading (f max ) in each bedrock horizon was estimated using a factor of safety of 2.5. Rock quality designation (RQD), spacing and thicknesses of discontinuities were evaluated to estimate the base resistance (q max ) for compressive loading. A factor of safety of 2.5 was used to determine allowable unit base resistance.

6 6 A comparison was made between the calculated and measured allowable unit base and unit side resistances based on an 8 feet (2.44 m) diameter shaft as presented in Table 1. The unit end bearing for shale (mudstone) was calculated from the results of test shaft 1, at a settlement of 0.5 inches. The unit end bearing for siltstone was calculated from the results of test shaft 2, at a settlement of 0.5 inches. LATERAL LOAD TEST RESULTS Figures 10(a) and 10(b) presents the load-deflection response measured at the loading point for test shafts 1 and 2, respectively. The deflections were averaged from the two dial gage readings for each shaft at each load increment. It can be seen that the deflections of shaft 2 at the loading point is much larger than those of shaft 1 which may be attributed to the use of casing to isolate the shaft above bedrock. Because shaft 1 mainly served as a reaction shaft and the rock-shaft interaction was not fully engaged due to the thick layer of overburden soil, only inclinometer readings of test shaft 2 are presented. The deflection versus depth of test shaft 2 at each load increment along the length of the shaft deduced from inclinometer readings are presented in Figure 11. Figure 11 indicates that the deflection at the elevation of the top of rock is approximately 0.20 inches (5.1 mm). The measured strain at the maximum applied load 275 Kips (1200 KN), the highest strain was at 57 feet (17.4 m) from the top of the shaft, which was 1350 µ-strain. The stress in the rebar at the strain gage location was approximately ksi (270 MPa), and the concrete was cracked at that stage at that depth. BACK CALCULATION OF SOIL AND ROCK PARAMETERS The results from the lateral load test were used as a basis to back-calculate the soil parameters that will result in the same movement at the top of the shaft utilizing LPILE program. The load deflection curve at the top of shaft 1 was used as the reference in the back analysis. For test shaft 1, the movement was resisted by the overburden and the rock socket. The overburden consists of gravel and sand. The parameters for the soil and rock layers were adjusted to back calculate the deflection to match the deflection from test shaft #1. Weak rock criteria proposed by Reese (1997) was used to model the rock layers. Table 2 lists the suggested parameters that were used to match the shaft top deflection at each load increment. FORWARD ANALYSIS OF LATERAL RESPONSE OF WV TOWER SHAFTS The parameters in Table 2 were used to forward predict the response of the production shafts for WV Tower. The boundary condition at the top of the shaft was assumed as free head. The soil profile from boring S-9 was used to define the layers thickness. The reinforcement in the shafts was 28 #18 bars similar to the test shafts. The nonlinear EI method was used in the analysis which accounts for the variation in the rigidity of the shaft due to the formation of cracks in the concrete. Analyses were conducted to estimate the movement of the shaft under lateral loading applied at elevation 536 (Normal pool elevation). Fig. 12 shows the movement vs. lateral load applied at elevation 536 for loading up to 275 Kips (1200 KN), and the corresponding movement measured during the lateral load test at shaft 1 where the overburden was contributing to the resistance.

7 7 SUMMARY AND CONCLUSIONS The results of the vertical (O-Cell TM ) load tests and lateral load test on 8 feet (2.44 m) diameter drilled shafts, socketed into rock were presented. A comparison was made between the allowable unit bearing pressure and unit side friction used in the design and the ones calculated from the load testing for both the sandstone and the shale layers. The data presented in this paper show that the design ultimate bearing pressure and side friction for the rock at the site is significantly less than the test results, which support the idea of increasing the ultimate design values taking into consideration the serviceability requirements of the structure. The interim p-y criterion for weak rock proposed by Reese (1997) was used to back calculate the rock parameters to match the deflection from the lateral load test. The back calculated parameters were used to predict the behavior of the production drilled shafts for the WV tower of the bridge. As clarified in LPILE technical documentation, the interim p-y criterion proposed by Reese (1997) was suggested for the cases when the unconfined compressive strength of rock ranges from 5 to 55 TSF (0.5 to 5.2 MPa). The p-y criterion predicted the behavior of the rock socketed tested drilled shafts as, in most of the layers, the unconfined compressive strength of the weak rock above the sandstone are less than 55 TSF (5.2 MPa). Therefore, the parameters used in the interim criterion by Reese can be adjusted if additional lateral load testing is done. Based on the load test and analyses results, the following conclusions can be drawn: The back-calculated p-y curves can capture the rock response under lateral loads. These curves can be used for the redesign of drilled shafts at the same site where the test was performed. The interim rock p-y criterion of Reese (1997) can provide good prediction by modifying the rock properties. It is difficult to characterize the engineering properties of a rock mass, especially highly weathered rock, in the laboratory. In-situ testing like pressuremeter is recommended for estimating design parameters. Evaluation suggest the need of further development of the analysis of laterally loaded drilled shafts in rock mass, especially for cases were the unconfined compressive strength of rock is higher than 55 TSF (5.2 MPa). ACKNOWLEDGMENT The authors gratefully acknowledge the support of Jawdat Siddiqi of the Ohio Department of Transportation (ODOT). The cooperation of Load Test Engineers, Bill Ryan, John Hayes, and Robert Simpson by sharing their experience and the data from the O-Cell testing is greatly appriciated. The contractor team of Mahan/National is gratefully acknowledged for facilitating the instrumentation and testing work during construction. NOTE For copies of the vertical and lateral load testing reports, please contact the office of Structural Engineering at the Ohio Department of Transportation.

8 8 REFERENCES 1. Carter, J. P., and Kulhawy, F. H. Analysis of laterally loaded shafts in rock, Journal of Engineering, ASCE, 118(6), 1992, FMSM Engineers, Geotechnical Engineering Report Structure Ohio 833 Bridge over the Ohio River, MEG , Pomeroy, Meigs County, Ohio Mason, Mason County, W.V Reese, L. C. Analysis of laterally loaded piles in weak rock J. Geotechnical and Geoenvironmental Engineering, ASCE, 101(7), 1997, Reese, L.C., Wang, S.T., Isenhower, W.M., and Arrellaga, J.A. LPILE, A program for the Analysis of Piles and Drilled Shafts under Lateral Loads, ENSOFT, Inc., Austin, Texas, Zhang, L., Ernst H., and Einstein H. H. Nonlinear analysis of laterally loaded rocksocketed shafts. Journal of Geotechnical and Geoenvironmental Engineering. 126(11), 2000,

9 9 LIST OF TABLES Table 1 Designed and measured unit side friction and end bearing. Table 2 Estimated parameters used to match the deflection from lateral load test. LIST OF FIGURES FIGURE 1 Elevation view of the test shafts and log of borings S-9 and S-10 FIGURE 2 Plan view, and typical sections in the test drilled shafts. FIGURE 3 Plan view of the lateral load test setup. FIGURE 4 Pictures during construction and Testing of Shafts. FIGURE 5 Instrumentation and load test set-up. FIGURE 6 Test Shaft 2, O-Cell Load Movement Curves. FIGURE 7 Test Shaft 2, Equivalent Top Load-Movement Curves. FIGURE 8 Test Shaft 2, Strain Gage Load Distribution Curves. FIGURE 9 Test shaft 1, Equivalent Top Load-Movement Curves. FIGURE 10 Load-deflection curves. FIGURE 11 Deflection profiles of shaft 2. FIGURE 12 WV tower, Predicted lateral movement at the top of the shaft compared to measured lateral movement from test shaft 1.

10 10 TABLE 1 Designed and measured unit side friction and unit end bearing. (Based on 8 Diameter Shaft) Bedrock Horizon Approximate Elevation Range Allowable Unit Base Resistance (tsf) Allowable Unit Side Resistance (tsf) (ft) Calculated * Measured Calculated * Measured Shale with Interbeded Siltstone 478 to ** 5.6 Shale (Mudstone) 463 to Siltstone 435 to ** - * Values were obtained for a Factor of Safety of 2.5. ** Assumed roughened socket sidewalls

11 11 Layer No. TABLE 2 Estimated parameters used to match the deflection from lateral load test. Top Elev. Bottom Elev. Unit Wt. Elasticity Modulus Uniaxial Compressive Strength (psi) RQD Notes (ft.) (ft.) (pcf) (psi) (%) * - * - * Sand and Gravel Shale Shale(Mudstone) 4 435? Siltstone * The parameters for sand and gravel layer were chosen as recommended by Reese for Sand. The unit weight was assumed as 125 pcf. The friction angle was assumed as 32 degrees, and the modulus of subgrade reaction as 60 pcf.

12 12 Ele. SPT q u (ft) /RQD (ksf) % 32% % % % % % 81 S-9 E i (ksi) Description (lb/ft 3 ) Top of ground Coarse and fine sand Gravel with sand Weathered shale Top of rock Shale (50%) with interbedded siltstone (50%). Shale is laminated to very thin bedded, zones silty, soft to hard. Shale (mudstone), greenish gray to brownish red, partially claylike, blocky bedding, numerous slickensided fractures Siltstone, gray to dark gray, thin bedded, zones sandy, micaceous, hard Ele. (ft) SPT q u /RQD (ksf) % 700 S-10 E i (ksi) % % % % % 664 (lb/ft 3 ) FIGURE 1 Elevation view of the test shafts and log of borings S-9 and S-10.

13 13 FIGURE 2 Plan view, and typical sections in the test drilled shafts.

14 14 FIGURE 3 Plan view of the lateral load test setup.

15 15 FIGURE 4 Pictures during construction and testing of shafts.

16 16 Ele. (ft) 8.5 ft 11 ft Ele. (ft) Dial gages 3 ft Jack Load cell ft Strain gages Top of soil 8.5 ft -OD casing 28# ft cover Top of rock Inclinometer casing O-Cell Test Shaft #1 Working Deck 11 ft-od casing Gap 8 ft Test Shaft #2 3 O-Cells FIGURE 5 Instrumentation and load test set-up.

17 Upw ard Top of O-Cell TM 2.00 Continuous unload data. Decrement not held Movement (inches) Dow nw ard Base of O-Cell TM Gross Load (kips) FIGURE 6 Test shaft 2, O-Cell load movement curves.

18 Settlement (inches) Conservative minimum creep limit Thin Line: Rigid Curve Thick Line: Rigid Curve Adjusted for Additional Elastic Equivalent Top Load (kips) FIGURE 7 Test shaft 2, equivalent top load-movement curves.

19 Mud Line Top of Shaft Elevation (ft) S. G.Level 4 O-Cell TM Points Bottom of Shaft 420 1L-2 1L-4 1L-6 1L-8 1L-10 1L-12 1L-14 1L Gross Load (kips) FIGURE 8 Test shaft 2, strain gage load distribution curves.

20 20 Osterberg Cell Load-Movement Curves Pomeroy-Mason Bridge - Meigs Conuty, OH - Test Shaft Upward Top of O-Cell TM Movement (inches) Downward Base of O-Cell TM This increment held for 14 Minutes Gross Load (kips) FIGURE 9 Test shaft 1, Equivalent Top Load-Movement Curves.

21 Lateral Loads (Kips) Deflection (in.) inclinometer Dial Gage 1 Dial Gage 2 (a) Test shaft Lateral Loads (Kips) Inclinometer Dial G age 1 Dial G age Deflection (in.) (b) Test Shaft 2 FIGURE 10 Load-deflection curves.

22 22 Deflection (in.) Top of Shaft, elev. 547ft Jacking, elev. 544ft Depth (ft) Top of Rock, elev. 491ft at 25 Kips at 50 Kips at 100 Kips at 150 Kips at 175 Kips at 225 Kips at 250 Kips at 275 Kips 110 FIGURE 11 Deflection profiles of shaft 2.

23 Lateral Loads (kips) Predicted Measured Deflection (in.) FIGURE 12 WV tower, predicted lateral movement at the top of the shaft compared to measured lateral movement from test shaft 1.