Lateral Resistance of Short Rock Sockets in Weak Rock: a Case History

Text word count: 4140

Number of figures and tables: 13

Robert L. Parsons PhD, P.E (Corresponding Author)

Associate Professor
Department of Civil, Environmental and Architectural Engineering
1530 W 15th St., Room 2150
University of Kansas
Lawrence, KS 66045
rparsons@ku.edu
(785) 864-2946

Matthew C. Pierson
Assistant Professor, Cooperative Engineering Program
Kemper Hall, Room 226
Missouri State University
901 S. National Ave.
Springfield, MO 65897
matthewp@ku.edu
785-393-2427

Isaac Willems
Graduate Student
Department of Civil, Environmental and Architectural Engineering
1530 W 15th St., Room 2150
University of Kansas
Lawrence, KS 66045
iwillems@ku.edu
(309) 258-7743

Jie Han PhD, P.E

Associate Professor
Department of Civil, Environmental and Architectural Engineering
1530 W 15th St., Room 2150
University of Kansas
Lawrence, KS 66045
jiehan@ku.edu
(785) 393-3714

James J. Brennan
Assistant Geotechnical Engineer
Kansas Department of Transportation
2300 Van Buren
Topeka, KS 66611-1195
brennan@ksdot.org
(785) 296-3008

TRB 2011 Annual Meeting Paper revised from original submittal.

 1
2
3 Lateral Capacity of Short Rock Sockets in Weak Rock: a Case History
4
5
6 ABSTRACT
7 The results from full-scale cyclic and repeated lateral load testing of two short rock sockets in
8 weak rock and the recommendations developed for p-y analysis using those results are presented.
9 Two drilled shafts were constructed in rock sockets 42 inches in diameter to depths of
10 approximately seven feet in limestone in Wyandotte County, Kansas. The shafts were loaded
11 laterally during three separate test events. The shafts were tested under cyclic loading (load
12 reversal) for loads up to 400 kips, repeated loading in one direction up to 820 kips, and to failure
13 near 1,000 kips.
14 Test data showed that socket behavior was essentially elastic during cyclic loading for
15 loads of 400 kips (40% of nominal resistance) and lower. The shafts experienced permanent,
16 accumulating deformations during repeated loading to 610 and 820 kips.
17 Modeling of the results showed the lateral load behavior could be effectively modeled in
18 LPILE using the Reese weak rock model included with LPILE software. Recommendations
19 for use in modeling are presented.
20

1

TRB 2011 Annual Meeting Paper revised from original submittal.
 1
2
3 INTRODUCTION
4 This paper contains the results from a full-scale lateral load test of two drilled shafts constructed
5 in short rock sockets in weak limestone, and the development of recommendations for p-y
6 analysis using those results. Lateral nominal resistance of drilled shafts is of particular interest
7 with regard to bridge foundations because of the significant loading conditions they experience,
8 particularly during scour events. Lateral nominal resistance may be estimated during the design
9 process by several methods, with one of the most common being a p-y analysis. P-y curves vary
10 among soil types and rock formations, although general curves have been developed and are
11 available for use in widely available software packages such as COM624 (public domain) and
12 LPILE (proprietary software, Ensoft).
13 The purpose of this research was to test the lateral capacity and develop p-y curves for
14 short rock sockets in weak rock. Two shafts 42 inches in diameter were constructed to depths of
15 six and seven feet in limestone in Wyandotte County, Kansas. All overburden material was
16 removed prior to construction. The shafts were loaded laterally during three separate test events.
17 During the first event, the shafts were loaded in a cyclic manner (load reversal) at multiple
18 increments up to 400 kips. The cyclic loading was of interest because of the potential for lateral
19 loading in alternating directions on shafts supporting integral abutments. The shafts were then
20 loaded in one direction to 550 kips. The equipment was then reconfigured and the shafts were
21 loaded to 820 kips with repeated loading-unloading cycles at 610 and 820 kips. The loading
22 frame was then reinforced and the shafts were loaded to failure, which occurred near 1,000 kips.
23 A description of the testing, analysis, and p-y curve recommendations is presented in Parsons et
24 al. (1).
25 Analysis of the data showed that commonly used p-y curves included within the LPILE
26 software could be used to develop an accurate model of the static behavior of the shafts. Cyclic
27 loading of the shafts had little effect on shaft resistance at lower loads; however permanent
28 deformation began to accumulate at loading levels between 40 and 60 percent of nominal
29 resistance.

30 THEORETICAL BACKGROUND

31 This section contains an abridged discussion of the p-y curve method. For a more detailed
32 discussion of the p-y curve method the reader is referred to Reese et al. (2).
33 For the p-y method the pile-soil interaction is modeled as a series of nonlinear springs,
34 where p represents lateral load per unit length on a spring and y represents displacement of

2

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 the spring. The non-linear relationship is captured by the secant modulus Er, which decreases
2 according to some function, such as a hyperbolic function, as displacement increases.
3 The p-y method was extended to the analysis of single rock-socketed drilled shafts under
4 lateral loading by Reese (3). Reeses criteria, which he considered interim criteria pending the
5 availability of more test data, include consideration of the secondary structures of rock masses
6 using a rock strength reduction factor determined from the Rock Quality Designation (RQD).
7 Other criteria used for generating p-y curves have subsequently been developed (4, 5). Reeses
8 criteria have been incorporated into LPILE v 5.0 Plus (2), and were used for the analysis
9 described in this paper.
10 For this method, the ultimate reaction Pu (units of force per length) of rock is given by:
11

1 1.4 for 0 3
5.2 b for 3
12 Where:
13 qur = uniaxial compressive strength of intact rock;
14 r = strength reduction factor, used to account for fracturing of rock mass, it is assumed
15 to be 1/3 for RQD of 100% and it increases linearly to 1 at a RQD of zero;
16 b = diameter of the drilled shaft, and;
17 xr = depth below rock surface.
18
19 The slope of initial portion of p-y curves is given by:
20 Kir kir*Eir
21 Where:
22 Kir = initial tangent to p-y curve;
23 Eir = initial modulus of the rock
24 kir = dimensionless constant
25 The expressions for kir, derived by correlation with experimental data, are as follows:
26
400
100 for 0 3
3
500 for 3
27
28 The p-y curves developed from these relationships follow the shape shown in Figure 1. This
29 figure shows a p-y curve with three segments; from the origin to yA, from yA to ym, and from yrm
30 to failure.
31

3

TRB 2011 Annual Meeting Paper revised from original submittal.
 1
2
3 Kir
pur
4
5
6
7
8
9
10
11
12
yA yrm y
13
14
15 Figure 1. Sketch of p-y curve for weak rock (adapted from Reese [3])
16
17 The equations relating p and y for the curve in Figure 1 are as follows:
18
for y yA
.
for y > yA , p <pur
2
for y > 16yrm
19
20 and
21 =
22
23 where
24
25 krm = a constant between 0.0005 and 0.00005 that controls the overall stiffness of the p-y curves,
26 and;
27
.

.
2
28
29
30

4

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 FIELD TESTING
2
3 Site Investigation
4 The shafts were constructed in the fall of 2007 in the northeast quadrant of the intersection of I-
5 70 and I-435 in Wyandotte County, Kansas. The sockets were in the Plattsburg Limestone. Two
6 borings were made and cores were recovered at the site in the vicinity of the rock sockets. The
7 site geology consisted of minimal to no soil overburden, 1.5-2.5 feet of weathered to
8 unweathered sandstone over weak limestone. The overburden and sandstone were removed so
9 the sockets were entirely in limestone. More detailed information is available in Parsons et al.
10 (1). Seven unconfined compression tests of the limestone were conducted on samples from
11 elevations considered relevant to this study. The results of those tests are reported in Table 1.
12 Depths in Table 1 are with respect to the original ground level. RQD values of 81% and 64%
13 were determined for the depth of interest from the two borings. An average of 70% was used for
14 the site.
15

TABLE1RockCoreTestDataUsedforAnalysis
Unconfined Elastic Dry Moisture
SampleNo. Depth Compression Modulus Density Percent
(ft) qu(psi) E(ksi) d(pcf) w%
Upper
Layer

1412 3.33 799 292 154.4 2.7

1513 3.33 2701 448 149.5 4.3
1421 4.05 4458 958 150.6 3.7
LowerLayer

1422 7.03 7778 1333 157.5 2.1

1521 4.30 5979 1118 156.9 2.7
1522 5.30 5056 1042 156.0 3.1
1523 6.95 4778 660 152.2 4.6
Valuesusedupperlayer 1750 370
Valuesusedlowerlayer 5068 1040
E/quupperlayer 211
E/qulowerlayer 205
3
effectiveuntwt(alllayers) 159lb/ft
RQD 70%

16 Note:Ewasdeterminedfromunconfinedcompressiontesting

5

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 This rock core data was considered to represent two layers; an upper, more weathered
2 layer and a lower more competent layer. Representative values for unconfined compressive
3 strength (qu) and initial intact rock modulus (E) were adjusted from the averages somewhat to
4 account for spatial relationships among sampling points. The E/qu ratios are slightly above 200,
5 which is consistent with published values (6).
6 Concrete cylinders were taken when the shafts were constructed and the 28-day curing
7 strength was determined to be 7,300 psi. Given the additional strength gain that should have
8 occurred prior to actual testing of the shafts and based on the strength gain from cylinders from a
9 concurrent study (7), a concrete strength of 7,500 psi was used for modeling purposes.
10
11 Shaft Details
12 The test shafts were 42 inches in diameter and spaced 144 inches apart center to center. They
13 were cast in sockets approximately six feet deep for the north shaft and seven feet deep for the
14 south shaft. Shaft reinforcement consisted of 12 #11 longitudinal bars and hoops made of #5
15 bars on a one foot spacing within the socket and a spacing of approximately six inches above
16 ground at the point of load application. The load was applied approximately one foot above
17 ground level. The concrete met KDOT standard specifications for drilled shafts.
18
19 Testing
20 Lateral load testing was conducted as part of three separate tests in 2009. The first test consisted
21 of cyclic (load reversal) testing up to 400 kips for a series of primary load increments, where 400
22 kips was the maximum load that could be achieved in both directions with the equipment
23 configuration used. The equipment was configured such that two separate load systems could
24 load the shafts in opposite directions. One set of equipment with three 206 kip hydraulic
25 cylinders was used to jack the shafts apart, and a second set with two 206 kip cylinders was used
26 to pull the shafts together (Figure 2). Cycles of loading were applied to the shafts by alternating
27 loading between these sets of equipment. Five or 10 cycles were applied at each primary load
28 increment. During each cycle the load was held at the target value until deformation stabilized
29 for approximately 1 minute. Additional measurements were taken at intermediate increments.
30 Load was measured using two separate systems, load cells and hydraulic pressure. The
31 hydraulic pressure was monitored by gauge and by pressure transducer. The load cells were
32 limited to a capacity of 400 kips and served as a backup to the pressure transducer and gauge.
33 Deformations were determined from inclinometer measurements in each shaft and at two
34 locations on each shaft with string pots fixed to reference beams. Pressure transducer, string pot,
35 and load cell data was recorded automatically on a laptop computer. Photogrammetry was used
36 as a backup system. Pressure transducer and string pot information was recorded by a laptop

6

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 and data acquisition system. Inclinometer data was recorded by KDOT personnel with a data
2 logger prior to each test and after each set of load cycles.

Inclinometercasing
stringpots
stringpots
hemisphericalball

loadcells
cylinderspush
shaftstogether loadrods(4oneachside)

cylinderspush
shaftsapart
referencebeam
referencebeam
3
4 Figure 2. Test 1 setup
5
6 For the second test the equipment was reconfigured so that all five cylinders could be
7 used together to load the shafts as shown in Figure 3. Repeated loads, which consisted of
8 loading to the target value and then reducing the load to zero, were applied at 610 and 820 kip
9 load levels with 10 cycles at each load step. As loading continued for Test 2 above 820 kips, one
10 of the loading beams began to yield, forcing the test to be stopped.
11 The yielding beam was reinforced and the test was restarted as Test 3. Loading
12 proceeded to failure at approximately 1,000 kips for both shafts.
13
14 Field Data
15 Figures 4 and 5 show the deformations that occurred during each test event as measured
16 by the top string pots, which were approximately 3.8 ft above ground level. Data for individual
17 cycles are not shown in these graphs. These figures show increasing rates of deflection with
18 increasing load to failure, which occurred at approximately 1,000 kips for both shafts. Data for
19 the lower string pots on each shaft were similar.
20

7

TRB 2011 Annual Meeting Paper revised from original submittal.
 1
2 Figure 3. Loading configuration for Tests 2 and 3
3
4 These figures served as the primary physical test information used to calibrate the LPILE
5 models. The string pot deformation data was checked against inclinometer data, which was used
6 as an absolute reference when combining information from Tests 1, 2, and 3. The data plotted
7 for Tests 2 and 3 were offset based on the difference in inclinometer readings taken at the end of
8 the previous test and the beginning of the test for which the data was plotted.
9 Additional observations can be made in addition to the general trend of the data. Little
10 to no permanent accumulation of deformation was observed for cyclic loading of the shafts at
11 400 kips or lower. Accumulation of deformation was significantly greater at the 820 kip loading
12 increment than for the 610 kip loading increment. The south shaft deformed significantly more
13 than the north shaft under the same loading, reaching a deformation of nearly 0.7 inches after
14 cycling at 820 kips while the north shaft had a deformation of approximately 0.3 inches at the
15 same point. This was likely due to natural material variability; however there was a road cut that
16 was present approximately 20 feet behind the south shaft in the direction of loading which could
17 have made it possible for sliding along a weak plane to have occurred in that direction.
18

8

TRB 2011 Annual Meeting Paper revised from original submittal.
 1200
Accumulated
deformation during
1000 repeatedloading

800

600
Load (kips)

Test 1
400 Test 2
Test 3
200

0
0 0.1 0.2 0.3 0.4 0.5 0.6
Deflection (inches)
1
2 Figure 4. Deflection with loading of the north shaft 3.83 ft above ground (upper string pot)

1200
Accumulated
deformation during
1000 repeatedloading

800
Load (kips)

600
Test 1

400 Test 2
Test 3

200
Deflections arenegativebecausethedirectionof
movementisoppositeofmovementforthesouthshaft
0
-1.2 -1.1 -1 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0
Deflection (inches)
3
4 Figure 5. Deflection with loading of the south shaft 3.86 ft above ground (upper string pot)
5

9

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 For the north shaft there was no permanent deformation between Tests 1 and 2, and there
2 may have been a small rebound between testing events. However, during Test 2 the shaft
3 behaved as if it had a lower modulus in the early stages than it had during Test 1, but then
4 stiffened when loading exceeded 600 kips. For the south shaft this behavior was reversed. The
5 shaft experienced a small permanent deformation as a result of Test 1 and had a higher modulus
6 during reloading up to 610 kips. The behavior of the south shaft is consistent with the loading of
7 many geomaterials, where it would be expected that some permanent deformation would be
8 made to the material during the initial loading, and during repeated loadings the geomaterial
9 would have elastic behavior with a higher modulus in that loading range. The mechanics behind
10 the behavior of the north shaft are not well understood, but it may have gradually rebounded to
11 its original position under small lateral earth pressures; then quickly moved past its maximum
12 deformation level from Test 1 (550 kips) under loading of only 300 kips in Test 2.
13
14 Behavior during Cycling
15 Cyclic loading (load reversal) was applied at loads of 100, 200 and 400 kips for five cycles each
16 during Test 1. Ten cycles were applied for a load of 300 kips. For Test 2 the load frame was
17 reconfigured for repeated loading where loads were applied and released in the same direction
18 for 10 cycles at loads of 610 and 820 kips. This data is presented for the top string pot on the
19 south shaft in Figure 6, except for two cycles at 200 kips which were not recorded. Similar
20 accumulating deformations for high loads were also observed for the north shaft.
21 Figure 7 shows more detail for the deformations for the cyclic loading of the north shaft.
22 For this figure the deformation was reset to zero at the beginning of each set of cycles. The
23 nearly constant amplitude for each set of cycles shows there was no significant increase in
24 movement of the shaft during cycling for loads up to 400 kips.
25
26 Deformations with Depth
27 Shaft deflections were monitored with depth using inclinometer measurements. The inclinometer
28 casings were installed inline with the loading direction and show that while the shafts were
29 relatively rigid for early loading, they experienced more significant bending at higher loads. The
30 inclinometer readings from Test 2 are shown in Figure 8.
31

10

TRB 2011 Annual Meeting Paper revised from original submittal.
 900
cycle 1
800 cycle 2
cycle 3
700 cycle 4
cycle 5
600
cycle 6
500 cycle 7
Load (kips)

accumulated cycle 8
400 deformation cycle 9
duringrepeated tight grouping cycle 10
300 loading ofpoints
showselastic
200 behavior
duringcycling
100

0
-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0
1 Deflection (inches)
2 Figure 6. South shaft accumulated deformation with cyclic and repeated loading 3.83 ft above
3 ground (upper string pot)
4
0.08
100kips
DeformationfromCycleStartingPoint

0.06 200kips

0.04 300kips

400kips
0.02

0.02

0.04

0.06

0.08
0 200 400 600 800 1000
5 DataCounter
6 Figure 7. North shaft elastic behavior with cyclic loading at lower loads 3.86 ft above ground
7 (upper string pot)

11

TRB 2011 Annual Meeting Paper revised from original submittal.
 Deflection (inches) Deflection (inches)
0 0.2 0.4 0.6 0.8 -0.05 0 0.05 0.1 0.15 0.2 0.25
-3 -3

-2 -2

-1 -1

Ground Ground
Depth below ground level (ft)

0 0

1 1

2 2

3 3

4 End Test 1 - 0 load 4

End Test 1 - 0 load
Test 2 - 0 load Test 2 - 0 load
5 410 kip 5 410 kip
610 kip 610 kip

6 720 kip 6 720 kip

820 kip 820 kip
End Test 2 - 0 load End Test 2 - 0 load
7 7
1
2
3 Figure 8 Inclinometer deflection with depth for test 2 (south shaft left, north shaft right)
4
5
6
7
8

12

TRB 2011 Annual Meeting Paper revised from original submittal.
 1
2 LPILE MODELING
3
4 Modeling Parameters
5 The rock-socket test data was modeled using the commercial program LPILE for the purpose of
6 identifying the appropriate Reese p-y modeling parameters for limestone. The weak rock
7 model contained within LPILE combined with a Type 3 analysis, which considers non-linear
8 bending, was used for the analysis. Properties used in the modeling are presented in TABLE 2.

TABLE 2 LPILE Modeling Parameters

Shaft Properties

Shaft diameter 42 inches

Shaft length below ground (north shaft) 6 feet
Shaft length below ground (north shaft) 7 feet
Height of load application above ground 1 foot
Concrete strength 7500 psi
longitudinal reinforcement 12 - #11 bars
Distance from pile top (point of loading) to
12 inches
ground surface
Yield stress of steel 60,000 psi
Steel modulus 29,000,000 psi

Rock Properties
Upper Lower
Layer Layer

Intact rock strength 1750 psi 5068 psi

Intact rock modulus 370 ksi 1040 ksi
krm 0.0005 0.0005
9
10
11 Once the geometry and reinforcement of the shaft are determined, there are only two
12 remaining parameters that must be selected by the modeler. The value of krm is adjustable; and
13 the value of 0.0005 that was used is within the recommended range (2). There also may be some
14 justification for reducing the rock modulus because the interim Reese criteria may not fully
13

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 account for the lower modulus of the rock mass as compared with the modulus of the intact rock
2 samples as described in the next section.
3
4 Discussion of Modeling
5 When fitting the load-deformation curves generated within LPILE to the load test data, an
6 adjustment was required to account for the accumulated deformation that occurred during
7 repeated loading. This was addressed by shifting the LPILE curves by the amount of the
8 accumulated deformation. These LPILE curves are plotted with the field test data in Figures 9
9 and 10. These figures show that a relatively good fit is obtained between the weak rock LPILE
10 model and the field test data with regard to the nominal resistance and ultimate pile head
11 deformations, with the nominal model resistance being within about 10% of the observed
12 ultimate load for both shafts. A selection of actual p-y curves generated within LPILE is
13 presented in Figure 11. These p-y curves apply to both shafts. The model results predict the
14 shafts reaching maximum moment and failing near 900 kips with nearly all bending above a
15 depth of three feet. Lengthening the shafts has minimal impact on model resistance or pile head
16 deflections.

1200
Adjustments ofmodelforpermanent
deformationsfromrepeatedloading
1000

800
LPILE with
permanent
600
load (kips)

deformations

north shaft
400

200

0
0 0.2 0.4 0.6 0.8 1
deflection (inches)
17
18 Figure 9. LPILE model and load test data for the north shaft 3.83 ft above ground (upper string
19 pot)

14

TRB 2011 Annual Meeting Paper revised from original submittal.
 1200
Adjustments ofmodelforpermanent
deformationsfromrepeatedloading
1000

800

600
LPILE with
load (kips)

permanent
400 deformations

south shaft
200

0
-1.2 -1 -0.8 -0.6 -0.4 -0.2 0
deflection (inches)
1
2 Figure 10. LPILE model and load test data for the south shaft 3.86 ft above ground (upper string
3 pot)
4

Depth

5
6 Figure 11. p-y curves using the intact rock modulus, produced using LPILE (2)

15

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 The model did predict somewhat less deformation than was observed in the working load
2 range, particularly for the south shaft (Figure 9). It also predicted that nearly all rock movement
3 and shaft bending would occur in the top three feet of the socket, while inclinometer
4 measurements show some movement all the way to the bottom of the sockets. This may have
5 occurred because the model used does not fully account for a reduction in the modulus of the
6 rock mass compared with the intact rock modulus; and to a degrading of the modulus and
7 accumulation of deformation from the cyclic and repeated loading. Reducing the rock modulus
8 used in the model by a factor between one and two orders of magnitude for the north shaft and
9 slightly more for the south shaft improved the match in the working range, did not affect the
10 nominal model resistance significantly, and eliminated the need to include offsets for the
11 accumulating deformations from cyclic and repeated loading.
12
13 CONCLUSIONS AND RECOMMENDATIONS
14
15 Two 42-inch diameter drilled shafts constructed in short rock sockets in weak rock were laterally
16 loaded to failure. Cyclic and repeated loading steps were conducted for a series of load steps
17 prior to failure. The following conclusions were drawn from the field data.
18 The nominal resistance of both rock sockets was approximately 1,000 kips.
19 The nominal resistance was reached at approximately 0.45 inches of lateral movement for
20 the north shaft and 0.95 inches for the south shaft. Both of these deformation values
21 include deformation that accumulated during periods of repeated loading. Maximum
22 deformations for static load test conditions would likely have been less.
23 Deformations for the south shaft were likely due to material variability, but may have
24 been affected (increased) by the presence of a road cut approximately 20 feet behind the
25 shaft.
26 The shafts behaved in an elastic manner for five cycles of loading at 100, 200 and 400
27 kips (40% of ultimate load) and 10 cycles at 300 kips.
28 The shafts experienced permanent, accumulating deformations for repeated loading at
29 610 kips (approximately 60% of nominal capacity), and even greater deformations at 820
30 kips.
31
32 The resulting field data was modeled using the interim Reese weak rock model included
33 within the commercial software LPILE. The following conclusions were developed based on the
34 modeling.
35 The nominal resistance and ground line deformations of the rock sockets could be
36 modeled reasonably well using the weak rock model contained within LPILE. Predicted

16

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 nominal resistance was within 10 percent of field measurements and the slope of the
2 load-deformation curve (modulus) was consistent with field data when accumulated
3 deformations were accounted for.
4 For this model, most of the data to be entered is driven by the material properties and
5 geometry, which makes construction of the model very straightforward.
6 The authors used a value of 0.0005 for krm, which is the upper end of the recommended
7 range.
8
9 Based on these conclusions, the following preliminary recommendations are made for
10 modeling of limestones. They are considered preliminary because they are based on a single test
11 program and should be updated as more data becomes available.
12 Use of the weak rock model included within LPILE for modeling short rock sockets is
13 supported by the observations from this research.
14 Within this model it is recommended that a value of 0.0005 be used for krm if no other
15 information is available.
16 It is also recommended that for cyclic or repeated loading design where the number of
17 cycles is expected to be relatively small (i.e. extreme events), the limestone can be
18 considered elastic for loads of less than 40% of the nominal resistance.
19 If the intact rock modulus is the basis for selecting the rock modulus value used in
20 LPILE, use of a reduced value may be warranted to more accurately model shaft bending
21 and deformations in the working range.
22
23 ACKNOWLEDGEMENTS
24
25 The authors wish to thank the people of the Kansas Department of Transportation for their
26 financial and logistical support that made this research possible. We particularly want to thank
27 the people of the KDOT Geotechnical Unit and KDOT Maintenance for their help in bringing
28 this project to fruition. We also wish to thank Mr. Jim Weaver of the University of Kansas (KU)
29 for his help in designing and fabricating the equipment and Mr. Justin Clay of KU for his help in
30 fabrication of the equipment for Test 1 and with some of the theoretical background presented in
31 this paper. We also wish to thank Mr. Paul Axtell and Dan Brown of Dan Brown and
32 Associates, who helped with the testing and interpretation of data. The help of all who
33 participated is greatly appreciated.
34
35
36

17

TRB 2011 Annual Meeting Paper revised from original submittal.
 1 REFERENCES
2
3 1. Parsons, R.L., I. Willems, M.C. Pierson, and J. Han. (2010). Lateral Capacity of Rock
4 Sockets in Limestone under Cyclic and Repeated Loading. Kansas Department of
5 Transportation. 86p.
6 2. Reese, L.C., S.T. Wang, W.M. Isenhower, and J.A. Arrellaga (2004). LPILE Plus 5.0 for
7 Windows, A Program for the Analysis of Piles and Shafts Under Lateral Loads.
8 Technical Manual. Ensoft, Inc. Austin TX.
9 3. Reese, L.C. (1997). Analysis of Laterally Loaded Piles in Weak Rock. Journal of
10 Geotechnical and Geoenvironmental Engineering. ASCE. Reston, Virginia. v123 n11.
11 1010-1017.
12 4. Cho, K.H., S.C. Clark, B.D. Keany, M.A. Gabr, and R.H. Borden. (2001).Laterally
13 Loaded Drilled Shafts Embedded in Soft Rock. Soil Mechanics. Transportation
14 Research Record. Journal of the Transportation Research Board. n1772. Transportation
15 Research Board. Washington D.C. 3-11.
16 5. Gabr, M.A., R.H. Borden, K.H. Cho, S. Clark, and J.B. Nixon. (2002). P-y Curves for
17 Laterally Loaded Drilled Shafts Embedded in Weathered Rock. North Carolina
18 Department of Transportation. FHWA/NC/2002-008. 289p.
19 6. Goodman, R.E. (1989). Introduction to Rock Mechanics, 2nd ed. John Wiley and Sons.
20 New York. 562p.
21 7. Pierson, M.C., R.L. Parsons, J. Han, D. A. Brown, and W.R. Thompson. (2008).
22 Capacity of Laterally Loaded Shafts Constructed Behind the Face of a Mechanically
23 Stabilized Earth Block Wall. Kansas Department of Transportation. Report KU-07-6.
24 237p.
25
26

18

TRB 2011 Annual Meeting Paper revised from original submittal.