assessing adhesion of driven pipe piles in clay using...
TRANSCRIPT
Assessing Adhesion of Driven Pipe Piles in Clay UsingAdaptation of Stress History and Normalized Soil
Engineering Parameter ConceptSteven R. Saye, M.ASCE1; Dan A. Brown, M.ASCE2; and Alan J. Lutenegger, M.ASCE3
Abstract: This paper presents a method to estimate the side resistance of a driven displacement-pipe pile in clay using the stress history andnormalized soil engineering parameter (SHANSEP) concept. The side resistance is treated as an adhesion, and this adhesion is normalized to theeffective overburden stress. This normalized adhesion is then related to the soil overconsolidation ratio using an adaptation of the SHANSEPconcept to separate the normally consolidated behavior from the overconsolidated behavior. This approach provides a rational means of screen-ing representative measurements of undrained shear strength from measurements that may have been affected by sample disturbance. Assess-ments of the soil overconsolidation ratio are developedwith laboratory odometer test data and an empirical approach using laboratory undrainedstrength data. A database of pile load tests screened for the effects of incomplete set-up and soil-property data screened for the effects of sampledisturbance is used to evaluate the normalized side resistance and the relationship of normalized side resistance to overconsolidation ratio.DOI: 10.1061/(ASCE)GT.1943-5606.0000842. © 2013 American Society of Civil Engineers.
CE Database subject headings: Cohesive soils; Pipe piles; Clays; Driven piles; Adhesives; Stress history; Soil consolidation.
Author keywords: Cohesive soil; Pipe piles; Side resistance; Side adhesion.
Introduction
The axial resistance of piles in cohesive soil is most often estimatedusing empirical correlations of side resistance with total stress-strength parameters, i.e., undrained shear strength su. Such an ap-proach can provide a reliable indication of pile resistance if themethod is developed through extensive local experience in a specificgeologic formation with a consistent method of determining su andapplied in a manner that is faithful to the original empirical cali-bration to piles of similar type, size, and embedded length. However,such empirical methods suffer from variability and inconsistencyand often show poor correlations with long-term axial resistancewhen applied broadly to a wide range of diverse geologic con-ditions. Factors affecting the use of empirical correlations with suinclude variability related to sample disturbance, inconsistent testmethods used to determine undrained shear strength, the effectsof in situ state of stress due to overconsolidation, pile installa-tion methods, and other factors. Attempts to develop a practicalapproach using effective stress-strength parameters have notachieved broad acceptance due to the relatively complex natureof the problem of predicting the effects of pile installation on thestate of stress at the pile-soil interface as well as the difficulty in
determining effective stress-strength parameters of the remoldedsoil at the interface.
The stress history and normalized soil engineering parameter(SHANSEP)–based approach (Ladd and Foott 1974)described inthis paper provides the means to address many of the limitations ofconventional empirical correlations of axial resistancewith su. In thisapproach, the average long-term side resistance of driven pipe pilesqs is treated as an undrained adhesion that is normalized to theaverage effective overburden stress s9vo along the length of the pileand related to the average overconsolidation ratio (OCR) using theSHANSEP concept (Ladd and Foott 1974).Model pile load-test dataobtained by Steenfeld et al. (1981) are presented to illustrate theapplication of the SHANSEP concept to pile side-adhesion calcu-lations, and an empirical correlation is developed to relate the averagenormalized side adhesion qs=s9vo developed from the pile load tests tothe averageOCRdeveloped from laboratory odometer tests. Pile load-test data from sites considered normally consolidated are used toassess thenormally consolidatednormalized side adhesion ðqs=s9voÞNC.
Assessment of the design OCR profile is important, and a com-bination of approaches is used to select the design OCR profile forpile capacity calculations. From initial assessments of OCR usinglaboratory odometer tests, this paper presents an approach to assessthe OCR with undrained strength data su commonly available fromunconsolidated undrained (UU) triaxial compression tests (su UUC)and unconfined compression tests (su UC). The effects of sampledisturbance on the laboratory UUC and UC strength data aredescribed, and an approach is presented to screen the data for dis-turbance. Cone-penetration test data are used to make furtherassessments of the OCR profile rather than directly assessing qs. TheOCR is related to measured normalized side adhesion qs=s9vo froma database of driven piles bearing in clay using both OCR calculatedfrom laboratory su UC and UUC data and OCR values measuredusing odometer and in situ tests. This paper addresses the long-termcapacity of piles reflecting full set-up. The record of load tests wasscreened to remove data with short delays between driving and
1Senior Geotechnical Engineer, Kiewit Engineering Co., Kiewit Plaza,Omaha, NE 68131 (corresponding author). E-mail: [email protected]
2President, Dan Brown & Associates, LLC, 300 Woodland Road,Sequatchie, TN 37374. E-mail: [email protected]
3Professor, Dept. of Civil and Environmental Engineering, Univ. of Mas-sachusetts, Amherst, MA 01003-5205. E-mail: [email protected]
Note. This manuscript was submitted on April 26, 2011; approved onSeptember 20, 2012; published online on December 14, 2012. Discussionperiod open until December 1, 2013; separate discussions must be submittedfor individual papers. This paper is part of the Journal of Geotechnical andGeoenvironmental Engineering, Vol. 139, No. 7, July 1, 2013. ©ASCE,ISSN 1090-0241/2013/7-1062–1074/$25.00.
1062 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
testing where incomplete set-up was thought to significantly un-derstate the pile capacity.
In clays, the normalized side adhesion for long piles installedwithout significant interruption described by Semple and Rigden(1984) is shown to be similar to that for short piles, and closed-endpiles are similar to open-ended piles. This paper presents two ex-amples of displacement-pile performance to illustrate application ofthe SHANSEP-based approach.
Calculation Methods to Assess Pile Side Adhesion
The total-stress design methods that relate su to pile adhesion arebroadly identified as the alpha method based on work by Tomlinson(1957, 1970), Randolph (1983), and Semple andRigden (1984) usingthe relationship:
qs=su ¼ a (1)
where qs 5 measured unit-pile side adhesion, su 5 soil undrainedstrength measured as UC or UUC, and a 5 empirical correlationfactor developed with pile load-test data.
Tomlinson (1957) considered qs to be an adhesion along the shaftof the pile that approximates the remolded soil strength immediatelyafter driving and increases to the initial soil strength after consoli-dation.Vesic (1977)describes this time-related increase inpile capacityrelated to pore pressure dissipation and other factors. Subsequently,Randolph (1983) developed empirical correlations relating a to thenormalized undrained shear strength su=s9vo. Semple and Rigden(1984) later developed additional empirical relationships between aand su=s9vo presented on Table 1. The writers adapt the databasereported by Semple and Rigden (1984) with calculated OCR valuesand assigned OCR values in Table 1 with designations of qs datathought to reflect incomplete set-up and low qs values.
The effective-stress beta method described by Burland (1973),Esrig et al. (1977), and Esrig and Kirby (1979a, b) assumes that thefailure occurs in the soil near the pile-soil interface and relates qs tos9vo as
qs=s9vo ¼ KS tanw9OCRm ¼ bs9vo (2)
where qs 5 pile side adhesion, s9vo 5 vertical effective stress, KS 5lateral stress ratio,s95 drained-soil friction angle adjoining the pile,OCR 5 soil overconsolidation ratio, m 5 exponent relating theincrease in normalized side adhesion to the increase in OCR, andb 5 empirical correlation coefficient.
As noted by Sladen (1992), the a and b approaches are relatedthrough the normalized undrained strength and OCR. For valuesof su=s9vo , 0:7, these two methods diverge and match again atsu=s9vo 5 0:15.
Stress History and Normalized Soil EngineeringParameter–Based Approach
In the SHANSEP-based approach, qs is normalized to s9vo and thenrelated to the soil OCR in the same format as the concept fornormalized undrained shear strength presented by Ladd and Foott(1974), as presented in Eq. (3):
qs=s9vo ¼�qs=s9vo
�NCðOCRÞm (3)
where qs=s9vo 5 normalized pile side adhesion, ðqs=s9voÞNC 5normally consolidated normalized pile side adhesion, OCR 5 soilOCR, and m 5 exponent representing the increase in qs=s9vo withincreasing OCR.
The relationship of OCR and pile shaft adhesion is demonstratedin model tests of displacement piles penetrating kaolin clay reportedby Steenfeld et al. (1981) that were used to develop the EffectiveStress Model 4 (ESM4) method described by Kraft (1982). The pileshaft adhesion for the model steel piles was measured for threedifferent OCR conditions. Fig. 1 presents the plot of qs=s9vo versusOCR for these controlled conditions and illustrates the relationshipof qs to the soil OCR through this adaptation of the SHANSEPconcept. The ordinate intercept is the normally consolidated nor-malized side adhesion ðqs=s9voÞNC for the clay. The increase inqs=s9vo with increasing OCR is measured as the slope of the log-logplot m. Although some variation in ðqs=s9voÞNC occurs for these testdata, the data indicate a relationship between qs=s9vo andOCR that issimilar to the SHANSEP concept with values of ðqs=s9voÞNC 5 0:23and m5 0:71. Use of the OCR provides a method to address themode of failure in the correlation in that the normalized su for eachmode of failure is expected to be proportional to ðOCRÞm.
The relationship of OCR to pile shaft adhesion is demonstratedfrom the results of the field load tests from the sites summarized inTable 2 and presented in Fig. 2. These data represent observationsof the average qs=s9vo developed from published pile load tests ofpipe piles that are thought to exhibit complete dissipation of excesspore-water pressure with full set-up prior to testing. High-qualitymeasurements of the average OCR in the supporting clay wereobtained using laboratory odometer tests.Most of the tests in Table 2and Fig. 2 represent Norwegian Geotechnical Institute test sites thatwere reported by Almeida et al. (1996). These pile load tests ref-erencing the averageOCRdeveloped from laboratory odometer testsshow that ðqs=s9voÞNC 5 0:20 and m5 0:7. These tests suggest thatthe initial OCR profile in the supporting soils has a primary effect onthe load-carrying capacity of the pile. Other factors affecting stressconditions around the pile, set-up, and remolding of the soil at thepile interface are thought to be secondary factors affecting the re-lationship between the pile qs and the OCR.
The data from Fig. 2 suggest that the value of qs=s9vo for piles atsoft soil sites would be expected to approach 0.20, although trulynormally consolidated stress conditions are relatively rare in nature,and many soft soils exhibit an OCR. 1. For example, Burland(1973) presented data for qs versus the average pile length in soft soilshowing qs=s9vo values ranging from0.2 to 0.4. To evaluate values ofðqs=s9voÞNC using additional pile load-test data for straight-sidedpiles, the data in Table 3 were assembled from sites where thegeologic history shows nearly normally consolidated stress con-ditions in the bearing soil. The relatively small contribution fromend-bearing capacity in these compression tests was calculated as9suAp, where Ap 5 pile end area. The calculated qs=s9vo valuesreported by Bjerrum et al. (1969), Dawson (1970), Bozozuk (1972),and Fellenius (2006) for downdrag loads on piles in soft clay are alsopresented in Table 3 with OCR5 1 assigned to these tests showingvalues consistent with the normally consolidated values.
The ðqs=s9voÞNC data summarized in Fig. 3 show a relativelynarrow range of valueswith a general trend of decreasing normalizedadhesion with increasing plasticity index Ip. The data suggest anaverage ðqs=s9voÞNC of 0.20, consistent with the interpretation fromthe data inFig. 2. Further evaluation of soil type and the correspondingðqs=s9voÞNC value appears to be merited for assessment of downdragloads on piles, but variations in ðqs=s9voÞNC with soil type appear to bea secondary factor affecting the capacity of piles in clay soils.
Overconsolidation Ratio Assessment
In the SHANSEP-based approach, the qs=s9vo calculations requirean estimate of the OCR profile. Laboratory odometer tests are con-sidered an important part of this assessment, especially for low-OCR
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1063
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Tab
le1.
Sum
maryof
Pile
Load-TestR
esults,L
aboratoryData,andSou
rces
from
Sem
pleandRigden(198
4)
Loadtestno
.L(m
)D(m
m)
L=D.
50Openor
closed
ended
s u=s9 vo
q s=s9 vo
Calc.OCR
Assigned
OCR
Shear-strength
test
a
Plasticity
index
Com
ments
820
.476
2No
Closed
0.21
,1
U0.92
60?
6,20
21.6
457
No
Closed
0.21
,1
U0.93
60?
3,7,
1719
.261
0No
Closed
0.22
,1
U0.99
60?
478,
489
15.2
356
No
Open
0.23
0.24
,1
1.3
U1.05
60LaboratoryOCR
491,
493
12.2
356
No
Open
0.23
0.23
,1
1.15
U1.00
60LaboratoryOCR
854,
855
43.9
305
Yes
0.23
0.18
,1
1V
0.79
868
96.0
610
Yes
0.23
0.13
,1
1U
0.55
a
869
73.8
610
Yes
0.25
0.18
,1
1U
0.71
a
873
22.6
767
No
Open
0.26
0.30
,1
1.6
U1.13
Disturbed
451
66.4
325
Yes
Closed
0.27
0.14
,1
1U
0.52
a
4230
.532
5Yes
Closed
0.29
0.19
,1
1U
0.65
444,
450
45.7
325
Yes
Closed
0.35
0.15
1.12
1.12
U0.42
a
507,
508
29.0
330
Yes
Closed
0.37
0.38
1.20
U1.02
Disturbed
3013
.732
5No
Closed
0.40
0.38
1.32
U0.94
Disturbed
150
18.3
325
Yes
Closed
0.43
0.42
1.45
2.7
U0.97
Disturbed
4548
.261
0Yes
Open
0.43
1.45
U0.59
20Disturbed
844,
846,
848,
851
11.6
114
Yes
Closed
0.47
0.30
1.62
1.9
FV
0.64
60LaboratoryOCR
856
12.2
168
Yes
0.50
0.30
1.75
1.75
M0.62
325
14.0
351
No
0.51
0.40
2.3
FV
0.78
30Vane
6739
.627
4Yes
Closed
0.56
0.27
2.01
2.01
U0.49
4330
.561
0Yes
Open
0.57
2.06
2.06
U0.59
3044
3,44
922
.932
5Yes
Closed
0.57
0.30
2.06
2.06
U0.52
368,
369
25.9
325
Yes
Closed
0.62
0.35
2.29
2.29
U0.56
435,
436,
437,
438
25.3
274
Yes
Closed
0.76
0.36
2.95
2.95
U0.48
7014
.952
8No
0.79
0.41
3.09
3.1
U0.52
998
32.0
274
Yes
Closed
0.82
0.43
3.24
3.25
U0.52
4210
612
.832
5No
Closed
0.87
0.50
3.49
3.5
M0.57
10Step-tapered
547,
549
16.8
610
No
1.15
0.63
4.95
5.0
U0.55
31,32
13.7
325
No
Closed
1.22
0.58
5.33
5.3
U0.47
829,
830
13.1
274
No
Closed
1.37
0.67
6.16
6.1
Q0.49
4049
5,49
720
.461
0No
Open
1.98
0.87
9.76
9.75
M0.44
*9.1
450
No
Open
2.65
1.35
14.05
14.0
M0.51
23,24
18.3
762
No
Open
2.90
15.72
15.7
U0.46
35
Note:U5
unconfi
ned;
M5
othertests;FV
5fieldvane;Q
5qu
icktriaxial;V
5labo
ratory
vane.
a Oversized
closureplate.
1064 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
cohesive soils. In situ tests play an increasingly important part ingeotechnical studies, but the historical record of pile performance isdominated by su data developed from laboratory strength tests. Toincorporate these historical data into the SHANSEP-based approach,the following empirical approach is used to relate su=s9vo to OCRbased on the approach described by Ladd et al. (1977):
OCR ¼"�
su=s9vo�OC�
su=s9vo�NC
#1=m(4)
where ðsu=s9voÞOC 5 normalized undrained strength for the over-consolidated soil, ðsu=s9voÞNC 5 normalized undrained strength for
Fig. 1. qs=s9vo versus OCR for the Cambridge model pile test in Kaolin Clay (data from Steenfeld et al. 1981 and Kraft 1982)
Table 2. Case Study Sites qs=s9vo versus OCR from Odometer Tests
Loadtest no. L (m) D (m) L=D
Open orclosedended qs=s9vo
LaboratoryOCR
CPTOCR
Liquidlimit
Plasticityindex Source reference Symbol
C1j 4.5 0.17 27 C 1.52 24 13.4 75 50 Canons Park, England CrossCP5f 3.8 0.10 37 C 1.83 25 13.7 75 50 Almeida et al. (1996) CrossCP6d 3.7 0.10 36 C 1.87 17 13.7 75 50 CrossCP7d 3.6 0.10 35 O 1.76 17 13.7 75 50 CrossC 9 0.305 29.5 O 0.56 5.0 5.5 40 20 Cowden, England Long lineD 9 0.305 29.5 C 0.57 5.0 5.5 40 20 Almeida et al. (1996) Long lineE 9 0.305 29.5 C 0.56 5.0 5.5 40 20 Long lineF 9 0.305 29.5 O 0.56 5.0 5.5 40 20 Long lineG 9 0.203 44.3 O 0.65 5.0 5.5 40 20 Long lineH 9 0.203 44.3 O 0.74 5.0 5.5 40 20 Long lineI 9 0.203 44.3 C 0.61 5.0 5.5 40 20 Long lineJ 9 0.203 44.3 C 0.59 5.0 5.5 40 20 Long lineK 9 0.203 44.3 C 0.74 5.0 5.5 40 20 Long lineJacked 12.2 0.114 107 C 0.34 2.0 2.0 95 55 Hamilton AFB, USA
Almeida et al. (1996)Opendiamond
A1 5.0–15.0 0.219 46 C 0.265 1.55 1.5 70 40 Onsoy, NorwayB1 5.0–15.0 0.812 12 O 0.23 1.5 1.5 70 40 Almeida et al. (1996) Open squareA 9.9 0.219 45 C 1.57 22.5 22.5 55 35 Tilbrook Grange,
England NGIShort line
C 14.5 0.219 66 C 1.43 17.5 17.5 55 35 Almeida et al. (1996) Short lineC 0.225 1.25 41.5 18 Seed and Reese (1955) Solid square
1 14.0 0.324 43.2 O 0.21 1.0 1.0 100 50 Annacis Robertson et al.(1985)
Solidtriangle
15.2 0.356 42 O 0.24 1.3 Empire, LA Azzouz andLutz (1986)
Solid circle
15.1 0.457 33 C 0.22 1.2 1.2 38 18 Northwestern NGESFinno et al. (1989) Open
triangle13.1 0.274 48 C 0.67 6.1 6 70 40 Univ. of Houston NGES Asterisk
O’Neill et al. (1982)24.4 0.219 111.4 C 0.225 1.2 75 50 Blanchet et al. (1980) Open circle
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1065
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
the normally consolidated soil, OCR5 overconsolidation ratio, andm 5 exponent relating the increase in the normalized undrainedstrength to the OCR.
Ladd (1991) showed that the laboratory undrained strengthmeasurement of soil varies with the type of test, soil type, and OCR.The data presented in Fig. 4(a) (Koutsoftas and Fischer 1976;Koutsoftas and Ladd 1985) show the linear relationship of thenormalized su for triaxial compression (TC) using both Ko con-solidation (CKo UTC) tests and UUC for tests of alluvium in theAtlantic Ocean offshore of Atlantic City, New Jersey. These datashow that when plotted against OCR obtained from laboratoryodometer tests, the normalized su UUCundrained strength reasonablymatches the strength from reconsolidated CKo UTC tests at varyingOCR values with the normally consolidated value of su=s9vo 5 0:33.Ladd (1991) presented laboratory strength data for the Ko consoli-dated, normally consolidated, normalized strength, ðsu=s9voÞNC, fortriaxial compression, triaxial extension, and direct simple sheartests with respect to the soil plasticity index Ip. For the triaxialcompression mode of failure, ðsu=s9voÞNC is reasonably uniform withan average value of 0.32. Summarized in Fig. 4(b) are the su UUCand odometer test data from the sites listed in Table 4 that are usedto develop the following empirical relationship between the OCRand normalized su UUC strength tests on undisturbed samples:
OCR ¼�su=s9vo0:32
�1=0:8
(5)
Chen and Kulhawy (1993) present laboratory undrained strength andOCR data for a wide range of soils. The UUC and OCR data arepresented inFig. 4(c),with the trend line for theundisturbed strengthdataassembledby thewriters inFig. 4(b). Thewriters have indicatedbyusinga different symbol the data that are considered likely to be affected bysample disturbance, consistent with the data in Fig. 4(b) and Eq. (5).
All these results suggest that the use of disturbed strength datawould erroneously result in a lower calculated OCR and qs=s9vovalues than undisturbed data. Odometer tests (considered less sus-ceptible to sample disturbance than su UC and UUC tests) and in situtests are valuable additions to the site-characterization programs andtools to improve the OCR and qs=s9vo assessments. A discussion ofsample disturbance follows.
Sample Disturbance and Low su Values
Disturbance of the samples for UUC andUC tests leads to unreliableand potentially highly variable assessments of su and a. Errorsrelated to sample disturbance are considered especially important at
Fig. 2. qs=s9vo versus OCR from pile load tests using odometer tests to assess OCR
Table 3. Summary of Pile Load-Test Results, Laboratory Data, and Sources for Nearly Normally Consolidated Sites
Load test no. L (m) D (mm) L=D. 50Open or closed
ended su=s9vo qs=s9vo ðqs=s9voÞNC Assigned OCRSource/referenceadapted from Comments
See Table 1 7 155 No C 0.24 0.225 0.19 1.25 Seed and Reese (1955) Disturbed su4 53.5 400 Yes O 0.25 0.22 0.22 1 Trenter and Burt (1981) Disturbed suLake St. Peter 24.4 219 Yes C 0.225 0.2 1.2 Blanchet et al. (1980)1 14.3 324 No C 0.21 0.21 1 Robertson et al. (1985)Pipe pile 15.2 457 No C 0.22 0.2 1.2 Finno et al. (1989) Disturbed suSorenga 53 470 Yes C 0.18 0.18 1 Bjerrum et al. (1969) DDSorenga 41 500 Yes C 0.23 0.23 1 DDHeroya 32 500 Yes C 0.25 0.25 1 DDHeroya 30 300 Yes C 0.26 0.26 1 DDBerthierville 49 324 Yes C 0.2 0.2 1 Bozozuk (1972);
Fellenius (2006)DD
0.26 0.26 1 Dawson (1970) DD0.2 0.2 1 DD0.2 0.2 1 DD
Note: DD 5 downdrag.
1066 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Fig. 3. Pile load test data ðqs=s9voÞNC versus Ip
Fig. 4. (a) Normalized undrained shear strength versus OCR: AGS site (data from Koutsoftas and Ladd 1985 and Koutsoftas and Fischer 1976);(b) assessment of OCR using the normalized undrained shear strength: screened UUC, UC, and odometer data; (c) su=s9vo versus laboratory OCR usingUUC (data from Chen and Kulhawy 1993)
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1067
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
OCR values less than about 2. Direct measurement of the OCRwithodometer tests is considered to provide the most reliable means toassess OCR because these results are believed to be less sensitive todisturbance than the triaxial shear and unconfined compression tests.Odometer test data are particularly desirable to define theOCR valuewhere OCR, 1:5.
Disturbance of samples is a concern for all laboratory test data.The effects of significant sample disturbance limit the usefulness of
the empirical correlations developed with measured strength data.The writers have attempted to address this issue by assigning anOCR of 1 for sites where the reported su=s9vo is less than 0.3 andusing laboratory odometer test data to assess OCR instead of thelaboratory su data. To help identify disturbed samples and poor-quality laboratory strength tests, the su data can be normalized tos9vo, and values of su=s9vo from the su UUC tests of less than 0.3 areconsidered disturbed (recognizing that in rare occurrences soils
Table 4. Data for the Empirical Correlation to Assess OCR with UUC and UC Strength Data Screened for Sample Disturbance
Site Depth (m) su (kPa) s9vo (kPa) su=s9vo
LaboratoryOCR Site Depth (m) su (kPa) s9vo (kPa) su=s9vo
LaboratoryOCR
AGS clay 1.02 3.7 Leda clayKabir (1988)
4.25 75 54 1.4 6.9
Koutsoftas and Fischer(1976)
0.89 3.9 RRC 5.2 45 62 0.73 3
1.46 6.8 5.5 45 65 0.69 31.7 8.9 7.6 50 85 0.59 1.851.8 8.1 SLS 4.3 25 39 0.64 2.92.28 9 5.8 40 49 0.82 3.052.37 11.1 9.3 45 80 0.56 2.13.35 15.2 Plum Point, Arkansas 8.8 33 96 0.34 1.24.42 26 Unpublished 3.7 24 60 0.42 1.4
AMS 4 19 26 0.69 2.9 Quiros et al. 20 70 110 0.64 2.55Unpublished 5.2 17 31 0.56 2.2 (1983) 40 110 230 0.48 1.75
7.9 23 42 0.55 1.9 60 160 405 0.4 1.49.5 28 48 0.59 1.9 80 205 580 0.35 1.3
ACNJ 8.2 26 69 0.38 1.4 Ramalho-Oritagoet al. (1983)
2 5.1 6.4 0.8 3.1
Unpublished 3 5.7 9.6 0.59 2.3Bangkok Clay 2.2 34 29 1.17 4.7 4 6.2 12.8 0.48 2.1Moh et al. (1972) 3 24 33 0.71 3.1 St. Alban 2.6 16 22 0.73 2.27
3.6 29 35.3 0.83 2.6 La Rochelle et al. 4 16 32 0.5 1.75 28 43 0.64 2 (1974) 4.8 19 34 0.56 1.827.5 33 59 0.57 1.65 5.5 24 36 0.67 1.94
Beaumont clay 1.5 2.2 0.62 3.55 21.5 6.2 25 40 0.63 1.95O’Neill and Reese 3 2.4 1.25 1.92 9 7 22 44 0.5 1.93(1972) 4.5 2.6 1.88 1.38 6.5 7.7 27 46 0.59 2.06
6.1 2.8 2.18 1.28 5.2 8.4 29 48 0.6 2.08Beaumont clay 6.1 1.6 1.5 1.07 5.7 9.4 32 52 0.62 2.11Woodward-ClydeConsultants
7.7 1.8 2 0.91 5.3 Storz Expressway 1.7 0.8 0.47 0.8 3.4
(1982) 1.5 1.1 0.31 3.45 21.5 Omaha, NE 3.2 0.8 0.67 0.6 2.35.6 1.1 1.08 1 5.7 Unpublished 5.2 35 44 0.39 1.65.8 0.9 0.94 0.94 5.7 Station 264 5.9 36 49 0.37 1.5
Beaumont clay 7 125 92 1.36 6.4 9.4 43 68 0.38 1.2University of Houston Tilbrook Grange 4.7 360 72 5 34O’Neill (2000) Lambson et al. 7.5 560 108 5.2 23Chicago Till HDR 4 8.8 29 86 0.34 1.2 (1993) 8.5 420 122 3.45 22Finno and 9.5 430 135 3.2 19Chung (1992) 11 460 152 3 16Northwestern NGES-Finnoet al. (1989)
9.8 40 113 0.36 1.15 12.2 450 170 2.65 12
Finno et al. (1989) 3.2 145 67 2.16 12 13 440 180 2.45 12Cowden, England 4 110 76 1.4 8.5 13.2 400 185 2.15 11.8Lehane and Jardine 5 95 88 1.08 5.5 15.2 420 205 2.05 10(1994) 6.4 120 105 1.14 6.5 17 340 230 1.5 8.5
7 90 112 0.8 4 19 390 252 1.55 88 110 124 0.89 3.8 22.5 530 300 1.76 119.3 105 139 0.76 2.7 24 680 320 2.12 12
GEB VBC 13 47 102 0.46 1.55 25.5 670 335 2 9.5Unpublished 28.5 540 380 1.42 7
1068 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
could be underconsolidated). Themeasurement of volumetric strainsduring reloading to s9vo described by Andresen and Kolstad (1979)also can be used to assess the specimen quality to help eliminatesignificantly disturbed specimens from consideration and to rank thequality of the laboratory test results. In nearly normally consolidatedsoils, in situ tests can provide a reliable means to characterizethe site conditions with less uncertainty related to sample distur-bance; correlations with in situ test results may be used to assessthe OCR profile for pile design, particularly where laboratoryodometer tests confirm the OCR assessment made with the in situtest correlations.
qs=r9vo in Overconsolidated Soils: su Reference Data
The data available from pipe pile-load tests in overconsolidatedcohesive soils where su data have been reported provide a basis toextend the SHANSEP method to these conditions. Semple andRigden (1984) reported pile load-test data relating laboratory su datato the measured average qs of driven pipe piles installed entirely inclay. The average qs is related to the average s9vo at middepth of thepile. The data from Semple and Rigden (1984) are presented inTable 1 with the reported average undrained strength data, the typeof strength test, index property tests when reported, and the OCRcalculated from the su UUC tests using Eq. (5). Where available, thelaboratory OCR from odometer tests was used. These su data werescreened to remove undrained strength results considered disturbedin caseswhere su=s9vo was less than 0.3, andTable 1 shows that abouthalf the strength data were either eliminated, assigned an OCR5 1,or assigned a higher OCR based on the odometer data. The averagescreened su=s9vo data then are used to calculate an average OCRusing Eq. (5). The screened data are identified in Table 1 as thewriters’ assigned OCR column. The qs values also were screened forthe effect of incomplete set-up due to short wait times betweeninstallation and testing. Data considered reliable are identified inTable 1 as thewriters’ assignedqs=s9vo column.Thesevaluesofqs=s9voversus OCR are combined with the data from Table 2 reflectingthe laboratory odometer test results to develop the relationship inFig. 5. The calculated OCR values developed with the screenedsu=s9vo data combined with the qs=s9vo data correspond well with thedata developed from laboratory odometer tests and expand theempirical assessment of the exponent m to yield the empiricalrelationship
qs=s9vo ¼ 0:19ðOCRÞ0:7 (6)
Incorporation of Cone-Penetration Test Data into theStress History and Normalized Soil EngineeringParameter–Based Approach
Almeida et al. (1996) present cone-penetration test (CPT) andpiezocone test (CPTu) data formany of the sites presented in Table 2.Whereas Almeida et al. (1996) develop a direct relationship betweenuncorrected cone resistance qc or corrected cone resistance qT andqs, the SHANSEP-based approach uses the CPT/CPTu data first toevaluate the OCR profile and then to evaluate qs=s9vo using Eq. (6).The OCR values developed from the CPT/CPTu data are related tothemeasured qs=s9vo in Fig. 6 for the sites described byAlmeida et al.(1996) and others. These data fit reasonably well to the trend line forEq. (6), which is based on the OCR values developed from labo-ratory odometer tests and correlations with the laboratory su data.
Open-Ended Piles and Closed-End Piles, Long Piles,and Short Piles
When the concepts described in this paper are applied to availabletest-pile data in cohesive soils, the effects of closed-end conditionsand pile length do not appear to significantly affect qs. Fig. 7 presentsthe screened data fromTables 1 and2 sorted as either full-displacement(closed-end) or partial-displacement (open-ended) piles.No significantvariations are seen in the relationship between qs=s9vo and OCRbetween the two pile types. Fig. 8 presents the screened data fromTables 1 and 2 sorted by pile length. The long piles (length/diameter. 50) follow the same trend line as the short piles (length/diameter,50), suggesting that the available data do not provide a reason toseparate the evaluation of qs for pile length, as previously recom-mended by Semple and Rigden (1984). Three long piles fromTable 1with qs=s9vo , 0:18 (LoadTests 868, 451 and 444, 450)wereplotted in Fig. 8 with a designation o?. These data were excludedfrom the trend line for length variations in Fig. 8. Other factorsappear to affect the capacity of these piles.
The long piles reported by Peck (1961), designated 444, 450, and451 in Table 1 represent the pile tests at Drayton, North Dakota.
Fig. 5. Screened qs=su versus OCR from odometer tests with calculated OCR using undisturbed UC and UUC strength data (data from Semple andRigden 1984)
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1069
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Several issues are associated with the pile tests at this site that limitthe usefulness of the data to develop a general empirical correlationfor qs. First, the piles were installed with an oversized bottom platethat could reduce the qs of these piles. Second, su=s9vo 5 0:21 in thelower half of the piles, suggesting disturbance. Third, the long pileswere sequentially driven and tested and then redriven and retested.This method of installation is thought to be significantly differentfrom that of many of the other piles in the empirical database. Thepiles at Drayton were driven initially to a depth of 22.8 m and testedafter a set-up time of 10–18 days. The two tests at this length (443and 449 in Table 1) show a qs=s9vo of 0.30. The writers calculate anaverage su=s9vo of 0.96 from Peck (1961) and a calculatedOCR of 4,yielding results lower than the trend line for Eq. (6) likely due to boththe oversized bottom plate and short set-up time for these dis-placement piles. These piles then were redriven to a depth of 45.7 mand retested after set-up times of 8 and 9 days. One pile was redrivena second time to a depth of 66.4 m and retested after a set-up timeof 22 days. The redriven piles show qs=s9vo values of about 0.15,well below the trend line for Eq. (6). The redriving of the pilesafter significant set-up appears to affect the capacity of the Draytonpiles rather than the fact that the piles are long, and these long pilesare not included in the SHANSEP-based approach.
Examples
The writers present the following two examples to illustrate appli-cation of the SHANSEP-based approach.
St. Alban, Quebec, Canada
Konrad and Roy (1987) presented the results of a series of tests ontwo 22-cm-diameter closed-end pipe piles bearing in sensitivecemented Champlain clay installed to a length of 7.6 m. Test Pile Awas repeatedly loaded and tested at set-up times of 4, 8, 20, and 33days after installation. Test Pile B was tested after a set-up time of 2years. Both piles were jacked into the ground. La Rochelle et al.(1974) andRoy et al. (1982) present profiles ofOCRdeveloped fromlaboratory odometer tests that form the reference point for assess-ment of the stress history at this site. The soil-index properties andnormalized strength tests are presented in Fig. 9(a). The OCR profiledeveloped using the laboratory odometer data, the field vane-strengthdata, and the CPTu data are presented in Fig. 9(b). The measuredaverage qs=s9vo for full set-up (Test Pile B) is plotted with respect tothe averageOCRalong the side of thepile developedwith thedifferentlaboratory and in situ test methods in Fig. 9(d) corresponding well
Fig. 6. qs=s9vo versus OCR from pile load tests using CPT/CPTu tests to assess OCR
Fig. 7.Pile load-test data qs=s9vo versusOCRusing data screened for disturbance, open-ended and closed-end variations (data fromSemple andRigden1984)
1070 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Fig. 8. Pile load-test data qs=s9vo versus OCR using data screened for disturbance, length variations (data from Semple and Rigden 1984)
Fig. 9. Soil properties and pile load-test data qs=s9vo from St. Alban, QB, Canada, with assessments of OCR (adapted from Konrad and Roy 1987 andRoy et al. 1982)
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1071
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
with Eq. (6). The overconsolidated surficial crust appears to havea measurable effect on the average OCR values. The OCR valuesdeveloped using CPT/CPTu and field vane undrained strength dataare numerous and provide a definition of the OCR variations in thedesiccated crust within the upper 2mof the ground surface.No su UUCtestsweremade in the crust, and the average calculated averageOCR isthe lowest of the different test methods. The CPT/CPTu test methodprovided the highest assessment of the OCR profile.
Pentre, England
Almeida et al. (1996) describe the results of two closed-end pipepiles driven through surface casings at the Pentre site (Karlsrud et al.1993) tested in tension and one large-diameter open-ended pipe piletested in compression that is described further by Lambson et al.(1993). The soil-index properties are presented in Fig. 10(a),with thenormalized field-vane, piezocone, and laboratory su data presentedin Fig. 10(b). Fig. 10(c) presents the profile of OCR versus depthusing odometer tests from Lambson et al. (1993) and Almeida et al.(1996) with the writers’ assessment of the OCR using piezocone andsu data. These data profiles show that the soils have been variablystrengthened by desiccation to a depth of 10–12 m. The piezoconedata provide a good assessment of the OCR in the desiccated crust
but underestimate the OCR at depths below 20 m. The su data un-derestimate the OCR developed from the odometer tests, reflectingthe effects of sample disturbance in the lightly overconsolidatedsoil. Lambson et al. (1993) describe the site as normally consolidated,but the writers assign an average OCR of 1.45 and 1.6 to the soilprofile along Test Piles A5 and A6, respectively, based on theodometer data.
Test Pile A6 was cased to a depth of 22.5 m and extended toa depth of 32.5 m. In Fig. 10(d), Test Pile A6 shows an averageqs=s9vo value of 0.27 with an average OCR of 1.45, consistent withthe trend line for Eq. (6). Test Pile A5 was cased to a depth of 15 mand extended to a depth of 22.5 m. The load test for Test Pile A5yields a qs=s9vo value of 0.15, and the writers assign an average OCRof 1.6. In Fig. 10(d), this qs=s9vo value plots well below the trend linefor Eq. (6). Karlsrud et al. (1993) report very low horizontal effectivestresses alongside Test Pile A5; they also report that driving wasstopped at a depth of 19mbefore completing the driving to a depth of25 m. Significant drainage of the supporting soil likely occurredwhile the pile set-up at a depth of 19 m. Redriving Test Pile A5 anadditional 6 m is thought to have destroyed much of the side ad-hesion alongside the pile, remolding the silt and resulting in the lowqs=s9vo value. The writers believe that the lower capacity of longpiles in Fig. 8 is controlled by the pile installation procedure and the
Fig. 10. Soil properties and pile load-test data qs=s9vo from Pentre, England, with odometer, CPT, and su assessments of OCR (data fromAlmeida et al.1996 and Lambson et al. 1993)
1072 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
length of time needed to splice and drive subsequent pile sections.This concern and likely reduction in pile capacity are consideredespecially important for large-diameter pile sections.
Summary
An empirical approach has been developed to evaluate the side re-sistance of driven-pipe pile foundations in cohesive soil through anadaptation of the SHANSEP concept (Ladd and Foott 1974). Theaverage strength at the pile-soil interface is considered an undrainedadhesion qs, as described by Tomlinson (1957, 1970), that is nor-malized to the effective overburden stress s9vo, as described byBurland (1973), and related toOCR in themanner described byLaddand Foott (1974) for laboratory undrained strength tests. The pro-posed SHANSEP-based method separates the behavior of normallyconsolidated soils (where sample disturbance is most likely) fromthe behavior of overconsolidated soils (where sample disturbanceis less likely). The value of qs=s9vo is proportional to ðOCRÞm. TheSHANSEP-based approach incorporates the following concepts: (1)Assess theOCRusing a variety of approaches, including in situ tests,especially where the OCR is less than 2, and confirm the OCRassessment with laboratory odometer tests; (2) select soil samplingtechniques that minimize the effects of sample disturbance andconsider the effects of sample disturbance on the values of laboratorydata; discard laboratory test results where su=s9vo is less than 0.3; and(3) consider the effects of pile installation, pile set-up, and capacitygain with time.
Conclusions
The SHANSEP-based approach referencing the soil OCR providesa format to evaluate the unit side adhesion qs for pipe piles driveninto cohesive soils. By combining in situ tests, laboratory undrainedstrength tests, and odometer tests to assess the design OCR profile,an improved characterization of the site conditions was obtainedwith a single calculation of the pile side adhesion. In low-OCR clays,where disturbance of laboratory strength data are a significant factor,the use of odometer tests combinedwith CPT/CPTu tests reduces theeffects of sample disturbance on site characterization and improvesthe reliability of the design OCR profiles and the calculated pilecapacity. Using the SHANSEP-based approach with undisturbedand screened data, no variation between long and short piles wasidentified, but an adverse impact on qs was identified for piles in-stalled with long interruptions in driving where partial set-up occursduring the installation process. The SHANSEP-based approachprovides a method to assess the effects of specific constructionmethods on pile capacity, such as the detrimental effect of redrivingTest Pile A5 at the Pentre, England site.
Acknowledgments
The authors thank Jorge Santos and Bryan Kumm for their assis-tance in preparation of the figures for this paper.
Notation
The following symbols are used in this paper:Ap 5 pile end area;
CKo UTC 5 Ko consolidated undrained triaxial shear test;Ip 5 plasticity index;
KS 5 lateral stress ratio;m 5 exponent representing the increase in
normalized undrained strength or normalized sideadhesion with the increase in OCR;
o? 5 questionable data point;qc 5 uncorrected cone resistance;qs 5 pile side adhesion;
qs=svo9 5 normalized pile side adhesion;ðqs=svo9 ÞNC 5 normally consolidated normalized pile side
adhesion;qT 5 corrected cone resistance using a Type 2
piezocone;su 5 undrained strength using UC or UUC tests (no
test method referenced);sufv 5 field-vane undrained strength;
su UC 5 laboratory undrained strength measured withunconfined compression tests;
su UUC 5 laboratory undrained strength measured withunconsolidated undrained triaxial compressiontests;
ðsu=svo9 ÞNC 5 normally consolidated normalized undrainedstrength;
ðsu=svo9 ÞOC 5 overconsolidated normalized undrainedstrength;
WC 5 water content;a 5 qs /su alpha;b 5 beta 5 qs=svo9 ;w9 5 drained soil friction angle adjoining a pile; andsvo9 5 effective overburden stress.
References
Almeida, M. S. S., Danziger, F. A. B., and Lunne, T. (1996). “Use of thepiezocone test to predict the axial capacity of driven and jacked piles inclay.” Can. Geotech. J., 33(1), 23–41.
Andresen, A., and Kolstad, P. (1979). “The NGI 54-mm samplers for un-disturbed sampling of clays and representative sampling of coarsermaterials.” State of the art on current practice of soil sampling, In-ternational Society of Soil Mechanics and Foundations Engineering,London, 13–21.
Azzouz, A. S., Baligh,M.M., andLadd, C. C. (1983). “Cone penetration andengineering properties of the soft Orinoco clay.” Proc., Behavior ofOff-shore Structures, Vol. 1, Massachusetts Institute of Technology,Cambridge, MA, 161–180.
Azzouz, A. S., and Lutz, D. G. (1986). “Shaft behavior of a model pile inplastic empire clays.” J. Geotech. Eng., 112(4), 389–423.
Blanchet, R., Tavenas, F., and Garneau, R. (1980). “Behavior of frictionpiles in soft sensitive clays.” Can. Geotech. J., 17, 203–224.
Bjerrum, L., Johannessen, I. J., and Eide, O. (1969). “Reduction of negativeskin friction on steel piles to rock.” Proc., 7th Int. Conf. on Soil Me-chanics and Foundation Engineering, Vol. 2, A. A. Balkema, Rotter-dam, Netherlands, 27–34.
Bozozuk, M. (1972). “Downdrag measurements on a 160-ft floating pipepile in marine clay.” Can. Geotech. J., 9(2), 127–136.
Burland, J. B. (1973). “Shaft friction of piles in clay: A simple fundamentalapproach.” Ground Engineering, British Geotechnical Society, 6(3),1–15.
Chen, Y.-J., andKulhawy, F. H. (1993). “Drained strength interrelationshipsamong CIUC, UU, and UC tests.” J. Geotech. Eng., 119(11), 1732–1750.
Clarke, J., Rigden, W. J., and Senner, D. W. F. (1985). “Reinterpretation ofthe West Sole platform ‘WC’ pile load tests.” Geotechnique, 35(4),393–412.
Darragh, R. D., and Bell, R. A. (1969). “Load tests on long bearing piles.”Performance of deep foundations, ASTM STP-444, ASTM, WestConshohocken, PA, 41–67.
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013 / 1073
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Dawson, A. W. (1970). “Downdrag on pile foundations.”M.Sc.Eng. thesis,Dept. of Civil Engineering, Massachusetts Institute of Technology,Cambridge, MA.
Esrig, M. I., and Kirby, R. C. (1979a). “Advances in general effective stressmethod for the prediction of axial capacity for driven piles in clay.”Proc., 11th Offshore Technology Conf., Houston, 437– 449.
Esrig, M. I., and Kirby, R. C. (1979b). “Soil capacity for supporting deepfoundation members in clay.” Behavior of Deep Foundations (ASTMSTP-670), Raymond Lundgren, ed., ASTM, West Conshohocken, PA,27–63.
Esrig, M. I., Kirby, R. C., Bea, R. G., and Murphy, B. S. (1977). “Initialdevelopment of a general effective stress method for the prediction ofaxial capacity of driven piles in clay.” Proc., Ninth Offshore TechnologyConf., Vol. 3, Houston, 495–506.
Fellenius, B. H. (2006). “Results from long-term measurement in piles ofdrag load and downdrag.” Can. Geotech. J., 43(4), 409–430.
Finno, R. J., and Chung, C.-K. (1992). “Stress-strain-strength responsesof compressible Chicago glacial clays.” J. Geotech. Eng., 118(10),1607–1625.
Finno, R. J., Cosmao, T., and Gitskin, B. (1989). “Results of foundationengineering congress pile load tests.” Predicted and observed axialbehavior of piles, ASCE GSP-23, ASTM, West Conshohocken, PA,338–385.
Kabir, M. G. (1988). “Interpretation of the piezocone and piezoblade testsin clay.” M.S. thesis, Dept. of Civil and Environmental Engineering,Clarkson Univ., Potsdam, NY.
Karlsrud, K., Hansen, S. B., Dyvik, R., and Kalsnes, B. (1993). “NGI’s piletests at Pentre: Review of testing procedures and results.” Large-scalepile tests in clay, Thomas Telford, London, 405–435.
Konrad, J.-M., and Roy, M. (1987). “Bearing capacity of friction piles inmarine clay.” Geotechnique, 37(2), 163–175.
Koutsoftas, D., and Fischer, J. A. (1976). “In-situ undrained shear strengthof two marine clays.” J. Geotech. Eng. Div., 102(GT9), 989–1005.
Koutsoftas, D., and Ladd, C. C. (1985). “Design strengths for an offshoreclay.” J. Geotech. Eng. Div., 111(3), 337–355.
Kraft, L. M. (1982). “Effective stress capacity model for piles in clay.” J.Geotech. Eng. Div., 108(11), 1387–1403.
Ladd, C. C. (1991). “Stability evaluation during staged construction.” J.Geotech. Eng. Div., 117(4), 537–615.
Ladd, C. C., and Foott, R. (1974). “New design procedure for stability of softclays.” J. Geotech. Eng. Div., 100(GT7), 763–786.
Ladd, C. C., Foott, R., Ishihara, K., Schlosser, F., and Poulos, H. G. (1977).“Stress-deformation and strength characteristics.” Proc., 9th Interna-tional Conf. on Soil Mechanics and Foundation Engineering, Vol. 2,A. A. Balkema, Rotterdam, Netherlands, 421–494.
Lambson, M. D., Clare, D. G., Senner, D. W. F., and Semple, R. M. (1993).“Investigation and interpretation of Pentre and Tilbrook Grange soilconditions.” Large-scale pile tests in clay, Thomas Telford, London,134–196.
La Rochelle, P., Trak, B., Tavenas, F., and Roy,M. (1974). “Failure of a testembankment on a sensitive Champlain clay deposit.” Can. Geotech. J.,11(1), 142–164.
Lehane, B. M., and Jardine, R. J. (1994). “Displacement pile behavior inglacial clay.” Can. Geotech. J., 31(1), 79–90.
Moh, Z.-C., Brand, E. W., and Nelson, J. D. (1972). “Pore pressures undera bund on soft fissured clay.” Proc., Performance of Earth and Earth-Supported Structures, Vol. 1 Part 1, ASCE, West Conshohocken, PA,243–272.
O’Neill, M. W. (2000). “The national geotechnical experimentation site:University of Houston.” National geotechnical experimentation sites,Vol. 93, J. Benoit and A. J. Lutenegger, eds., ASCE, West Con-shohocken, PA, 72–101.
O’Neill, M. W., and Reese, L. C. (1972). “Behavior of bored pilesin Beaumont Clay.” J. Soil Mech. and Found. Div., 98 (SM2), 195–213.
Peck, R. B. (1961). “Record of load tests on friction piles.” HighwayResearch Board Special Rep. 67, Highway Research Board, Wash-ington, DC.
Quiros, G. W., Young, A. G., Pelletier, J. H., and Chan, J. H. C. (1983).“Shear strength interpretation for Gulf of Mexico clays.” Geotechnicalpractice in offshore engineering, ASCE, Austin, TX, 144–165.
Ramalho-Ortigao, J. A., Werneck, M. L. G., and Lacerda, W. A. (1983).“Embankment failure on clay near Rio de Janeiro.” J. Geotech. Eng.,109(11), 1460–1479.
Randolph, M. F. (1983). “Design considerations for offshore piles.” Geo-technical Practice inOffshore Engineering, ASCE, Austin, TX, 422–439.
Robertson, P. K., Campanella, R. G., Brown, P. T., Grof, I., and Hughes,J. M. O. (1985). “Design of axially and laterally loaded piles using in situtests: A case history.” Can. Geotech. J., 22(4), 518–527.
Roy, M., Tremblay, M., Tavenas, F., and La Rochelle, P. (1982). “Develop-ment of pore pressures in quasi-static penetration tests in sensitive clay.”Can. Geotech. J., 19(2), 124–138.
Seed, H. B., and Reese, L. C. (1955). “The action of soft clay along frictionpiles.” Trans. ASCE, Proc. Paper 842, 731–754.
Semple, R.M., andRigden,W. J. (1984). “Shaft capacity of driven pipe pilesin clay.” Analysis and design of pile foundations, ASCE, West Con-shohocken, PA, 59–79.
Sladen, J. A. (1992). “The adhesion factor: Applications and limitations.”Can. Geotech. J., 29(2), 322–326.
Steenfeld, J. S., Randolph, M. F., and Wroth, C. P. (1981). “Instrumentedmodel piles jacked into clay.” Proc., 10th Int. Conf. on Soil Mechanicsand Foundation Engineering, Vol. 2, A. A. Balkema, Rotterdam, Nether-lands, 857–864.
Tomlinson, M. J. (1957). “The adhesion of piles driven into clay soils.”Proc., 4th Int. Conf. on Soil Mechanics and Foundation Engineering,Vol. 2, A. A. Balkema, Rotterdam, Netherlands, 66–71.
Tomlinson, M.J. (1970). “Some effects of pile driving on skin friction.”Behavior of piles. Institution of Civil Engineers, London, 107–114.
Trenter, N. A., and Burt, N. J. (1981). “Steel pipe piles in silty clay soilsat Belawan, Indonesia.” Proc., 10th Int. Conf. on Soil Mechanics andFoundation Engineering, Vol. 2, A. A. Balkema, Rotterdam, Nether-lands, 873–880.
Vesic, A. S. (1977). “Design of pile foundations.” National CooperativeHighway Research Program Synthesis of Practice 42, TransportationResearch Board, Washington, DC.
Woodward-Clyde Consultants. (1982). “Study to investigate the effects ofskin friction on the performance of drilled shafts in cohesive soils.” Tech.Rep. GL-82-1, Geotechnical Laboratory, U.S. Army Engineer Water-ways Experiment Station, Vicksburg, MI.
1074 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JULY 2013
J. Geotech. Geoenviron. Eng. 2013.139:1062-1074.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Dan
Bro
wn
on 0
4/18
/14.
Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.