this document downloaded from

vulcanhammer.info
the website about Vulcan Iron Works Inc. and the pile driving equipment it manufactured

Terms and Conditions of Use:

All of the information, data and computer software (information) presented on this web site is for general information only. While every effort will be made to insure its accuracy, this information should not be used or relied on for any specific application without independent, competent professional examination and verification of its accuracy, suitability and applicability by a licensed professional. Anyone making use of this information does so at his or her own risk and assumes any and all liability resulting from such use. The entire risk as to quality or usability of the information contained within is with the reader. In no event will this web page or webmaster be held liable, nor does this web page or its webmaster provide insurance against liability, for any damages including lost profits, lost savings or any other incidental or consequential damages arising from the use or inability to use the information contained within. This site is not an official site of Prentice-Hall, Pile Buck, or Vulcan Foundation Equipment. All references to sources of software, equipment, parts, service or repairs do not constitute an endorsement.

Visit our companion site http://www.vulcanhammer.org

CYCLIC TENSILE TESTING OF A PILE IN GLACIAL TILL

R.P.L. McAnoy A.C. Cashman D. Purvis Taylor Woodrow Research Laboratories,

345 Ruislip Road,

Sou t hall,
M iddlesex

England

RESUME
For offshore s t r u c t u r e s in d e e p water, piles a r e being designed t o withstand cyclic tension, due t o uplift and wave loading, throughout t h e i r design life. However t h e r e is a s c a r c i t y of d a t a concerning t h e load levels which can b e safely applied in t h i s manner. This paper reports tensile tests on a heavily instrumented 10 m e t r e long pile jacked into glacial till. Previous work had shown s a t i s f a c t o r y pile behaviour under cyclic tensile loads peaking at up t o

48% of t h e u l t i m a t e tensile capacity, and so this work was aimed at

investigating pile response t o more s e v e r e load levels, approaching failure. Cyclic tests were performed with varying peak loads up t o 80% of t h e initial s t a t i c capacity, and up t o 13,500 cycles w e r e applied depending on pile response. The pile sustained encouragingly high loads without serious P i l e response deformation, but failure did occur during t h e most s e v e r e test, when t h e peak load was nominally 80% of t h e u l t i m a t e tensile capacity. analysis provided insight i n t o compression pile design methods when applied t o tension piles. Alpha and Lambda methods, b u t not t h e B e t a method, e s t i m a t e d
i

u l t i m a t e tensile capacity well, whilst stiffness was g r e a t e r than implied by published T-Z curves.

1.0

INTRODUCTION
/ -

The oil and a l t e r n a t i v e energy industries, by their increasing interest in large s c a l e floating structures, have stimulated research i n t o t h e behaviour of conventional s t e e l pipe piles as anchors. A major development programme has recently been completed for one such floating system, required t o provide a working platform in t h e 200-400 m e t r e range of w a t e r depths (Smith and Taylor 1980). T h e s t r u c t u r e comprises a buoyant column, t e t h e r e d t o a foundation anchorage by a series of tendons. Due t o t h e positive buoyancy of t h e column, t h e fluctuating loads imposed on t h e foundation are always tensile. The choice for t h e foundation design was between a piled or gravity base. Most of t h e foundation development programme w,as d i r e c t e d towards a piled base. Available design guidance was considered inadequate for both t h e u l t i m a t e carrying capacity of piles in tension and pile response t o working loads. Further, t h e e f f e c t on pile capacity of t h e vertical and l a t e r a l cyclic loads transmitted during s e v e r e storms could only b e t e n t a t i v e l y quantified. With this level of uncertainty t h e preliminary design method had t o ensure t h a t relatively low stress levels w e r e applied t o t h e piles during t h e 100 year storm. T o check t h e adequacy of t h e design a n experimental programme of. laboratory and onshore l a r g e scale pile tests was undertaken. T h e s e tests substantiated t h e design adopted, examined t h e stability of a soil during representative s t o r m loadings and determined t h e relative efficiency of jacking f o r c e t o tensile capacity. In t h e s e tests t h e basic pile p a r a m e t e r s and loading regimes were applicable to a jacked pile arrangement in a preliminary foundation design.

:?!
2 '

$4
. J

On completion of this section of t h e field test programme i t was f e l t t h a t it would b e of considerable value t o d e t e r m i n e t h e behaviour of a single pile under more severe loadings. Thus a single, heavily instrumented, open ended 10.4m long s t e e l pipe pile was tested u n d e r . a s e r i e s of high level tensile load packages. The results of t h e s e tests a r e presented in t h i s paper. F u r t h e r details of t h e laboratory and field test p r o g r a m m e have been reported by G a r a s (1979) and G a r a s and McAnoy (1980). DESCRIPTION OF FIELD %STS 2.1 TEST SITE Taking into account likely North S e a locations for a buoyant column s t r u c t u r e , t h e most suitable ground condition for an onshore test s i t e w a s considered t o b e a clayey glacial till. The U K Building Research Establishment (BRE) had already established a test site on t h e Holderness c o a s t at Cowden, 23 km north east of Hull, Humberside, t o investigate t h e properties of glacial till. An a r e a of this s i t e was m a d e available for t h e s e pile tests together with s i t e investigation data, including large s c a l e plate tests and pressuremeter results. The ground consists of

an

exposed

thick

deposit

of

uniform

overconsolidated, sandy silty clay till, with interposed layers of sand and gravel. Boreholes and cone p e n e t r o m e t e r results adjacent t o t h e

test position indicate zonal weathering of t h e till down t o 4.5 m e t r e s

and a dense grey black sand and gravel layer at 10.5 metres. Marsland and Powell (1980) g a v e a more detailed description of t h e till. other clay types has been presented by Lupini et a1 (1981). Table 1 shows typical soils d a t a as assessed f r o m information provided by t h e BRE, and from t h e above references. typical c o n e p e n e t r o m e t e r readings. Figure 1 (a) shows a typical borehole log f o r t h e site, a d j a c e n t t o t h e test position a n d ' l (b) A comparison of t h e residual s t r e n g t h of this Cowden till with various

2.2

TEST PILE AND INSTRUMENTATION

f7

The t e s t pile was one o f four s t e e l pipe piles (193mm diameter, 9 m m wall thickness by 10.4m long) which were each jacked 9.9m into t h e ground t o form a square group at 580mm centres. They w e r e installed open ended with a c u t t i n g shoe t o reduce plugging. External ducts were welded t o t h e piles t o p r o t e c t t h e instruments and cables. T h e location of t h e instruments on t h e most heavily instrumented pile, used f o r t h e s e single pile tests, is shown on Figure 2: The axial load along t h e pile was monitored by
1

.' f-?

.d'

bridge weldable

e l e c t r i c a l resistance strain gauges, positioned as pairs, diametrically opposite each other. The imposed head loads were monitored by a load cell built into t h e loading system. P o r e w a t e r and t o t a l l a t e r a l soil pressure measurements at t h e pile/soil i n t e r f a c e were taken by miniature e l e c t r i c a l transducers mounted in t h e ducts, flush with t h e o u t e r face. P i l e head v e r t i c a l and horizontal displacements were measured relative t o an independent r e f e r e n c e b e a m by linear displacement transducers. P o r e pressures in t h e surrounding soil at 2, 6 and 12 radii and 3.5m depth w e r e measured by hydraulic piezometers open t o t h e atmosphere. All instruments were calibrated and checked in t h e laboratory, t o ensure as f a r as possible t h a t they would perform satisfactorily during testing. Whenever practical, t h e s a m e calibration procedure was repeated on s i t e b e f o r e pile installation. 2.3 TEST PROGRAMME T h e four piles were jacked into t h e ground over an eight day period in April 1980. A pile was pitched and t h e instrumentation checked during one day, and t h e pile installed t h e next. On a v e r a g e installation took 6 hours t o reach t h e full embedded length of 9.9 metres.
..

Over a 41 day period f r o m t h e end of jacking, t h e measured excess pore pressure at t h e pile-soil hydrostatic pressure. i n t e r f a c e reduced t o within 5% of t h e this period the During

test

facility

was

rearranged t o a c c o m m o d a t e t h e load a c t u a t i n g system, and a rigid ste,el pile c a p w a s welded t o t h e head of t h e piles, leaving a c l e a r g a p b e t w e e n t h e pile c a p and t h e ground. A s e r i e s of cyclic load tests, summarised in Figure 3, w e r e then carried o u t on t h e pile group. This test work, which is t o b e discussed by

1 McAnoy et a (1982), iricluded a load package equivalent t o t h e

spectrum of a s e v e r e storm, with applied loads of up t o 48% of t h e v e r t i c a l or 6% of t h e horizontal e s t i m a t e d pull o u t capacity (U ).
t

The e s t i m a t e of Ut was based upon t h e c a p a c i t y of a single pile of identical dimensions installed and t e s t e d on t h e s a m e site adjacent t o t h e test position during t h e summer of 1979. A t t h e conclusion of t h e group tests i t was decided t h a t i t would b e of considerable value t o d e t e r m i n e individual pile performance under even m o r e demanding loading conditions. T h e pile c a p was t h e r e f o r e dismantled, and t h e loading system positioned over one of t h e four piles. A c o n s t a n t r a t e of extraction test t o failure was carried o u t as soon as possible a f t e r t h e group tests t o determine t h e u l t i m a t e capacity of t h i s pile. T h e value obtained was t h e n used as t h e initial c a p a c i t y of the single pile prior t o high level cyclic loading. Results from this initial e x t r a c t i o n test and subsequent high level cyclic loading of this pile have been used f o r t h e analysis presented. A summary of t h e test programme considered is shown in Figure 3(b). Each of t h e t h r e e cyclic tests shown was immediately followed by a c o n s t a n t r a t e of extraction test t o d e t e r m i n e t h e ultimate pullout capacity.
A r a t e of displacement of 0.75

mm/min was used, as

recommended by Whitaker (1963) f o r piles in cohesive soils. The tensile loads and numbers of cycles applied were nominally: Mean 40% UA 40% UB 30% Uc Cyclic Component No. of Cycles 11,775 564 13,597

- 20% UA + - 40% UB +
- 30% Uc +

for tests A, B and C respectively, where UA is t h e u l t i m a t e capacity determined immediately b e f o r e test A, etc. Sinusoidal loading was performed at t w o frequencies (0.1H
z

and

1.OH ) t o maintain similitude of s t r a i n and drainage respectively with

a prototype pile.

Problems of similitude for scaled pile tests a r e T o ensure t h a t viscous

discussed by Gallagher and St. John (1980).

damping did not a f f e c t cyclic displacements when drainage was being modelled, testing was performed under displacement control for t h e
'

higher frequency. 3.0 3.1 3.1.1 ANALYSIS OF FIELD TEST DATA ULTIMATE TENSILE CAPACITY Results Figure 4 shows t h e six constant r a t e of extraction tests t o failure performed on t h e pile. T h e first t w o tests E l , E2, w e r e carried o u t T h e change in t h e capacity before t h e pile had been subjected t o high level cycling and were used t o deduce t h e u l t i m a t e tensile capacity. recorded on e x t r a c t i o n tests E 3 t o E6. The maximum load from test E l was 495kN, which was comparable t o 481kN measured during test E2. The first test, E l , was carried o u t a month a f t e r t h e completion of t h e group tests, and t h e second test, E2, was performed a month l a t e r , one day b e f o r e t h e s t a r t of t h e cyclic
tests.

6 -2

a f t e r applying t h e cyclic load was assessed by using t h e c a p a c i t y

The slopes of

the load displacement curves a r e however The first c u r v e was non-linear over t h e

considerably different.

majority of its length and peak c a p a c i t y was reached at 5.2mm displacement. The second curve was linear up t o over 80% of t h e u l t i m a t e capacity, which was reached at 3.2mm displacement.
..

Extraction test E3, performed immediately

a f t e r cyclic test A,

indicated a d e c r e a s e in ultimate load capacity, t o 457kN. Extraction

test E4, a f t e r cyclic test B, indicated a f u r t h e r decrease t o 400kN. T h e

'-1

last two extraction tests, E 5 and E6, showed a slight increase of u l t i m a t e capacity. T h e shape of t h e load-displacement curves a r e The corresponding displacement linear over 80% of t h e i r range. 2.5mm. 3.1.2 Predictions of U l t i m a t e Tensile C a p a c i t y T h r e e conventional methods of predicting u l t i m a t e pile capacity have been assessed; t h e Alpha, Beta, and Lambda methods. None of t h e s e methods distinguish b e t w e e n tensile and compressive loading. In using t h e m a number of assumptions have been made based upon s i t e investigation d a t a and earlier results of pile load distribution. Firstly i t is assumed t h a t the uppermost 1.0m of t h e ground contributed nothing t o t h e side resistance, since strain gauge readings indicated t h a t resistance was very low in this region. Secondly t h e C u vs depth profile 2 c a n b e idealised to 170 kN/m for depths 1.0m t o 4.5m and 120 2 kN/m f r o m 4.5m t o 10.0m depth, as suggested by pressuremeter test results (Table 1). Alpha Method API RP2A (1980) recommends t h a t this t o t a l stress method is t o b e used for soils of medium t o low plasticity and c a n b e generally applied t o overconsolidated clays. I t is assumed t h a t t h e u l t i m a t e skin friction on t h e pile wall is proportional to t h e undrained shear strength, according t o t h e equation.

required t o mobilise t h e s e values were reasonably consistent at about

r s where

7 s

= aCu = u l t i m a t e shear s t r e s s on pile wall

cx

0.5 for t h e values of C u assumed above.

The alpha method predicted a s t a t i c tensile capacity of 410 kN.

B e t a Method
'J--.\

A number of methods have recently been developed using t h e e f f e c t i v e s t r e s s p a r a m e t e r s c',

n
--

9'.

These methods generally require a n e s t i m a t e In heavily

of t h e radial s t r e s s on t h e pile s h a f t a f t e r set up. overconsolidated clays t h i s is difficult t o obtain. The basic e f f e c t i v e stress equation involving Burland (1973) is;

P as suggested by

where

= =

Kstan@' softened drained angle of shear resistence coefficient of l a t e r a l e a r t h pressure insitu vertical e f f e c t i v e s t r e s s

@'
K
S
/

=
=

Ov 0

I t is assumed throughout t h a t c' = 0 due t o disturbance of t h e soil during installation. Meyerhof (1976) h a s analysed a large number of pile tests on bored and driven piles t o derive P values. H e defined P a s above but f r o m his work s t a t e s t h a t Ks varies from roughly KO t o 2 KO for driven piles, and suggests t h a t on average Ks may b e taken as 1.5 KO in stiff clays. An empirical relationship between t h e overcons;lidation is also suggested:KO

ratio,

p', and

KO,

=
=

(l-sin@')@EiT coefficient of e a r t h pressure at r e s t overconsolidation r a t i o 1.5 (1

where Hence

KO

OCR
Ks

= =

- s i n @'){a

Meyerhof's formula predicted a n u l t i m a t e pile c a p a c i t y of 270kN. L a m b d a Method

The Lambda method (Vijayvergiya & Focht, 1972) has been developed from t h e s a m e pile t e s t d a t a as t h e alpha method, but incorporates t h e

F45

_<,-

e f f e c t of penetration length on u l t i m a t e skin friction, which is calculated from t h e equation

dimensionless constant, which is a function of pile penetration s u r f a c e a r e a of pile signifies average over pile length

As

= =

The Lambda method predicted a n u l t i m a t e pile c a p a c i t y of 430 kN. 3.1.3 Discussion on U l t i m a t e C a p a c i t y

A summary of measured and predicted u l t i m a t e capacities is given in

Table 2. Two of t h e extraction tests have been used t o highlight t h e measured c a p a c i t i e s before and a f t e r failure under cyclic loading (Tests E l & E4 respectively). It is possible t h a t t h e 19% change in measured c a p a c i t y during this s e r i e s of tests was influenced by t h e formation of continuaus slip surfaces, as 18mm of cumulative displacement was recorded. The reduction in c a p a c i t y was close t o t h a t which could b e e x p e c t e d if t h e e f f e c t i v e angle of friction of t h e soil reduced t o t h e residual. T h e alpha and lambda methods provided reasonable e s t i m a t e s of the u l t i m a t e tensile capacity, and compared particularly well with t h e measured c a p a c i t y a f t e r cyclic failure. In order t o have predicted t h e measured capacity before cycling t h e alpha values used would have had

to b e approximately 0.6.

Previous case studies indicate t h a t alpha is F u r t h e r i t is generally

o f t e n higher t h a n t h e recommended value of 0.5.

expected t h a t jacked piles have a g r e a t e r capacity than driven piles. T h e g r e a t e r measured capacity may also reflect t h e influence of t h e installation of t h e adjacent t h r e e piles due t o a change of insitu stresses acting on t h e pile shaft. The b e t a method provided poor correlation between measured and predicted results. However, Meyerhof (1976) does show a wide s c a t t e r of p values from case records of piles in stiff clays. I t h a s been

suggested t h a t p a r t of the reason for this lies in t h e values of t h e e f f e c t i v e stresses assumed (Parry & Swain, 19771, as i t is difficult t o assess t h e c o r r e c t values. This uncertainty is reflected in the poor correlation noted above. However, given a b e t t e r determination of t h e e f f e c t i v e s t r e s s e s to b e used and t h e manner in which they change due t o installation and loading, t h e method could in principle allow for t h e e f f e c t s of adjacent piles, method of installation and direction of loading. B e t t e r correlation h a s been achieved in s o f t clays, in which e f f e c t i v e s t r e s s c a n b e more accurately assessed. Difficulties w e r e experienced in assessing t h e undrained shear strength values t o b e used in t h e calculations as t h e various insitu and laboratory

.,--,

,fl
-*

test methods indicated very different values. Most methods indicated

t h a t for 0-10 m e t r e depth t h e t o p 4.5 m e t r e s of soil had a higher Cu value t h a n t h e soil below. varied considerably. RESPONSE DURING CYCLIC TESTS Displacements during tests A, B and C a r e shown in Figures 5 and 6 . T e s t s A and C produced surprisingly small permanent displacements 0.12 and 0.14mm respectively high load levels applied. However, t h e magnitude of t h e difference

a f t e r over 11,000 cycles despite t h e linear relationship between mean

displacement and log number of cycles is evident, with t h e gradient of t h e line being slightly g r e a t e r f o r A than f o r C, reflecting t h e difference in peak loads. During t h e f i r s t

4b

cycles or so, t e s t B

exhibited a similar linear relationship, although t h e gradient of t h e line w a s considerably g r e a t e r than for t h e other tests. However, from 40 cycles onwards an entirely different response occurred. Mean displacements increased drastically, and continued t o do so up t o t h e point where t h e change of mean displacement per cycle was increasing with e v e r y cycle, indicating pile failure. clearly passed test B was t e r m i n a t e d necessary. Once this point had been only 564 cycles had been

No change in cyclic displacement occurred within any of the t h r e e

tests, indicating t h a t t h e r e was no degradation of cyclic soil modulus,

contrary to the suggestion of Poulos (1980) t h a t cyclic shear degradation was a function of stress level and number of cycles. I t .is particularly surprising t h a t no degradation in cyclic stiffness occurred during test B, especially when t h e pile was failing as is clearly shown in Figure 6(a). 3.2.1 Axial Load Distribution Figure 7 shows t h e axial load down t h e pile under cyclic and s t a t i c loading at t h e beginning and end of tests A, B and C. There is no significant c h a n g e with number of cycles, indicating t h a t no load shedding occurred during cycling. F u r t h e r m o r e t h e r e is no change in t h e relative proportion of load taken by each e l e m e n t during t h e s e cyclic tests and t h e u l t i m a t e values given by extraction test E3. I t is c l e a r t h a t t h e a v e r a g e shear stress, represented by t h e r a t e of d e c r e a s e of axial load with depth, varies considerably from t h e 1-3m depth t o t h e 3-8m depth, which is broadly consistent with t h e change in soil s t r e n g t h
at a b o u t 4-5m below ground, as discussed in Section 3.1.2.

3.2.2

P o r e Pressures Readings of excess pore pressure at t h e pile/soil interface, taken at t h r e e depths as shown in Figure 2 during tests A, B and C showed t h a t initial loading g a v e rise to a small excess pore pressure. A maximum 2 increase of 15 kN/m w a s measured at 3.15m level. For tests A and
C, cyclic loading did not lead t o any build-up in excess pore pressure

and t h e initial pressure reduced. There was s o m e evidence t o suggest a slight increase in e x c e s s pore pressure during t h e first 40 cycles of test 0 t o a maximum of 20 kN/m 2 at t h e 3.15m level. A s pull-out increased t h i s pressure dissipated, and at t h e end of all t h r e e tests excess pore pressures were continuing t o decrease. None had quite returned t o zero, but t h e maximum value at t h e end of cycling was down t o approximately 5 kN/m 2

3.2.3

Pile Head Stiffness and Apparent Soil Shear Modulus Values of pile and soil stiffness deduced from t h e cyclic tests a r e shown in Figure 8. T h e initial s t a t i c stiffness was reasonably consistent between t e s t A, B and C, whilst t h e final stiffness was similar for t e s t s A and C b u t dropped by 12% in B. The e f f e c t i v e stiffness reduced as t h e pile gradually pulled-out, t h e drop being drastic during test B. T h e cyclic stiffnesses remain constant throughout each test, b u t vary from one test t o another. F i g u r e 9 shows a plot of pile head stiffness vesus load range (i.e. maximum minus minimum head load) for t h e t h r e e cyclic tests. I t a p p e a r s t h a t load range and hence stress range, has been t h e dominant f a c t o r in t e r m s of cyclic stiffness, reflecting t h e non-linearity of t h e soil stress-strain curve, particularly at high load levels. Mean load level and r a t e of test would also b e expected t o influence cyclic stiffness, but a r e not thought t o b e significant in this
< ,

case. For example, t h e highest loading r a t e , which occurred during test

B, would b e expected t o produce t h e highest cyclic stiffness, when in f a c t test B exhibits t h e most flexible response. T h e a p p a r e n t shear modulus of t h e soil at t h e pile/soil i n t e r f a c e has been derived for an e l e m e n t of soil at 4.5m depth and exhibits t h e s a m e trend as pile head stiffness (Figure 8). T h e a p p a r e n t shear moduli derived, which range f r o m 50-76 MN/m 2 , a r e 3 t o 7 t i m e s g r e a t e r t h a n values determined from plate and pressuremeter tests, Table I. Similar behaviour was observed by Gallagher &'St. John f r o m pile t e s t s on t h e s a m e site. This is to b e expected as t h e shear strain during t h e
-.

tests was considerably lower t h a n those produced by most insitu testing.

NUMERICAL ANALYSIS OF LOAD-DISPLACEMENT BEHAVIOUR One method of estimating t h e load-displacement behaviour of a pile is t o use a d i s c r e t e e l e m e n t technique. T h e soil is modelled as a series of non-linear ground springs, characterised by IT-Z curves' relating pile displacement t o mobilised shear stress. The pile is divided into e l a s t i c e l e m e n t s a c t e d on by the relevant ground spring. A beam-column computer program c a n t h e n b e used t o d e t e r m i n e t h e pile head

n
.

---.'

load-displacement curve. Results from t h e extraction tests have been analysed to derive experimental T-Z curves. Results from e x t r a c t i o n
test E4, which produced the lowest u l t i m a t e capacity, were considered

to b e of g r e a t e s t interest, and a r e presented here. 3.3.1 Derivation of T-Z Curves f r o m Experimental D a t a T h e pile was divided i n t o seven e l e m e n t s as shown in Figure 10. Using t h e measured drop in axial load across a given element, t h e average s h e a r stress on t h a t e l e m e n t was computed. axial load down t h e pile; Displacement, z, of a n e l e m e n t was found f r o m t h e measured pile head displacement and t h e

where

Q(x)
A

=
= =

axial load in t h e pile at any depth x sectional a r e a of pile e l a s t i c modulus of pile material pile head displacement.

=t
3.3.2

Results and Predictions F r o m extraction test E4, t h e development of shear stress on each e l e m e n t for four load levels (a) - (d) a r e shown on Figure 11. Load level (d) represents t h e approximate u l t i m a t e capacity. The corresponding displacements of each e l e m e n t a r e also shown. Using information of this t y p e t h e T-Z c u r v e s for e l e m e n t s 3-6 have been plotted on Figure 12. Also shown a r e 'predicted' curves, which have been produced f r o m t h e work by Coyle & R e e s e (1966), who suggested a series of T-Z c u r v e s f o r piles installed in cohesive soils. Three standard curves w e r e proposed for varying soil depths, each being presented in t e r m s of undrained shear strength, a f t e r a correction to t a k e into account t h e effect of pile installation. The predicted curves of Figure 12 have been produced using t h e in-situ undrained shear strength profile assumed in
Sec tion 3.1.2.

Vij ayver giy a (1 977) developed a different approach f r o m results of load t e s t s on compression piles. I t was suggested t h a t t h e equation:

x - 7

' I-",
..J

=
$where T T
Z

(2 zlzc) max average shear stress on an e l e m e n t

u l t i m a t e a v e r a g e shear s t r e s s on an element

\IZ/ZcZ/Zc-

max

= =

v e r t i c a l displacement of an e l e m e n t c r i t i c a l displacement of a n e l e m e n t required t o mobilise Tmax.,

Zc

could b e used t o define T-Z curves for cohesive soils up t o failure. F u r t h e r m o r e Vijayvergiya suggested t h a t Tmax should b e derived using conventional methods for axial capacity, as discussed in Section 3.1.3, and t h a t 5-8mm is a typical range of Z
C

for piles of 300mm

d i a m e t e r and g r e a t e r . Figure 13, produced f r o m Figure 12 shows t h a t measured values of Z shows the
C

decrease with depth f r o m a maximum of Figure 14 plots of in shape between normalised

approximately 1.5mm (0.8% of pile d i a m e t e r ) t o 0.4mm. difference

Vijayvergiya's equation, measured values, and Coyle and Reese's c u r v e for d e p t h s g r e a t e r t h a n 6 metres. 3.3.4 Discussion I t is clear f r o m Figures 12 and 14 t h a t t h e T-Z curves exhibit a bilinear behaviour, and t h a t t h e curve suggested by Vijayvergiya is inappropriate in this case. The c u r v e for e l e m e n t 3 i k l i c a t e s a slightly s o f t e r response t h a n t h e remainder, but this is considered t o b e due t o t h e existence of a low radial stress over t h e upper part of this element. Each c u r v e c a n b e described by t w o parameters, Tmax and Zc. The values of Tmax predicted by Coyle & R e e s e for element 6 a g r e e s with measurements, but curves for e l e m e n t s 3, 4 and 5 yield gross max from c o n e p e n e t r o m e t e r results is desirable, since some correlati.on c a n b e observed between t h e shape of t h e u l t i m a t e shear stress distribution of Figure 1l(a), and t h e sleeve friction measurements of Figure l(b). underestimates. I t would appear t h a t a method of predicting T

c"r;,

f$
.-/

With r e s p e c t t o critical displacement, t h e recommendations of Coyle and Reese, and Vijayvergiya, again significantly o v e r e s t i m a t e measured values. Neither of t h e s e methods incorporate t h e e f f e c t of diameter on Zc, although t h e measured results appear t o conform t o t h e suggestion of 1% pile d i a m e t e r as an upper value for Z c (Whitaker and Cooke, 1976) for t h e piles tested. T h e reduction of Z c with depth which w a s observed is implied by t h e cu,rves of Coyle and Reese, and is possibly due to a n increase in confining pressure with depth. In any assessment of existing methods of analysis, i t is important t o note t h e conditions for which t h e methods w e r e designed. The curves of Coyle and R e e s e and Vijayvergiya were suggested f o r piles under sustained compression, and may b e a f f e c t e d by direction and r a t e of loading. On f i r s t impressions, therefore, i t might s e e m inappropriate t o c o m p a r e t h e curves with results from a c o n s t a n t r a t e of e x t r a c t i o n

a s t i f f e r response, particularly at high load levels. However, t o put this i n t o perspective

test,
which might b e expected to produce t h e cumulative displacements recorded a f t e r t h e completion of cycling in T e s t A have been plotted on Figure 12. The pile was still under t h e m e a n load and this displacement is considered as a n upper bound t o t h e displacements which would have been recorded during sustained s t a t i c loading over t h e s a m e period of 6 4 hours. The point still indicates t h a t t h e e l e m e n t response is s t i f f e r t h a n t h a t predicted by Coyle and Reese. However at higher load levels displacements due t o c r e e p a r e more significant and thus may approach or even exceed predicted displacements. 4. CONCLUSIONS E f f e c t of Cycling on U l t i m a t e Tensile Capacity I. The u l t i m a t e tensile capacity (Ut) of t h e pile reduced by 19% during t h e t e s t programme. This reduction was close t o t h a t which could b e expected if t h e e f f e c t i v e angle of friction of t h e soil reduced from t h e remoulded t o t h e residual value.

2.

The lowest U value occurred immediately a f t e r t h e test B, when

t

r--.

t h e pile failed under cyclic loading, and t h e reduction was probably due to
a

' 0
-* , '

combination

of

cyclic

loading

and

cumulative

displacem e n ts. Accuracy of Compression P i l e Methods for Determining Ut

3.

Good correlation between measured and predicted U t values was achieved using t h e alpha and lambda methods, which erred by 10-20% on t h e conservative side before cycling, and were within 7% of t h e measured value a f t e r cyclic failure.

4.

The b e t a method provided a gross underestimate before and a f t e r cycling.

Response During Cycling 5. T h e pile sustained encouragingly high levels of cyclic loading, peaking at up t o 60% Ut, with permanent displacements of only 0.14mm a f t e r 11,000 cycles.

6.

A t a higher load level, peaking at nominally 80% Ut, a d r a m a t i c change in behaviour occurred a f t e r a small number of cycles, and failure followed.

7.

The cyclic stiffness of t h e pile did not vary with number of cycles in any of t h e tests, even during failure.

8.

No significant

build-up

of pore pressure due t o cycling was

observed.

9.

No load shedding down the pile occurred during cycling, even at failure
..

T-Z Curves 10. The relatioriship

(2)

between For

mobilised

shear

stress

(T)

and

displacement constant.

on a pile e l e m e n t was linear f o r e

Zc, t h e

'critical' displacement.

Zc, T remained approximately

11. U l t i m a t e shear stress, Tmax, predicted by t h e T-Z method of Coyle and R e e s e underestimated measured values in t h e upper levels of t h e pile, a p a r t from t h e t o p m e t r e , which took no significant load. 12. The distribution of Tmax with d e p t h showed some correlation with c o r e p e n e t r o m e t e r sleeve friction results. 13. The c r i t i c a l displacement reduced with depth, possibly due t o an increase in confining pressure. 14. C r i t i c a l displacements were considerably less than t h e values suggested by Coyle and Reese. This may b e partly due t o t h e test pile being of a smaller d i a m e t e r than those from which t h e design curves w e r e derived. F u t u r e Work This investigation was limited t o the tensile behaviour of jacked piles in one soil type. Considerably more research is necessary before efficient design m e t h o d s can b e relied on for dynamically loaded tension piles. This research should c o n c e n t r a t e on t h e load transfer mechanism, c r e e p and sensitivity of soils t o dynamic loads. Taylor Woodrow and t h e U.K. Building Research Establishment have proposed

an extensive development programme t o measure t h e performance of different

t y p e s of piles under representative tensile axial and l a t e r a l loading conditions
..

and t o assess appropriate design methods.

This will incorporate and extend (Negotiations with potential

both organisations' present work in this area. still being sought).

sponsors for this project a r e well advanced, although f u r t h e r sponsorship is

Acknowledgements The Authors would like t o thank t h e Directors of Taylor Woodrow for their permission t o publish t h e s e results, t h e U.K. Building Research Establishment f o r t h e use of their test s i t e and s i t e investigation results, and all Research Laboratory personnel involved in t h e project. LIST O F REFERENCES Andersen, K. "Behaviour of Clay Subjected t o Undrained Cyclic Loading". Proc. Conf. Behaviour of Offshore Structures, Trondheim, Vol.1, 1976. Burland, J.B. "Shaft Friction of Piles in Clay

'/1
.."

A Simple Fundamental

Approach", Ground Engineering, Vol. 6,3, 1973. Coyle, H.M. & Reese, L.C. J n l ASCE, SM2, March 1966. Caras, F.K. "The Behaviour of Tension Piles for Offshore Applications with R e f e r e n c e to a Particular T e t h e r e d Buoyant Structure", 2nd Int. Conf. on "Load Transfer for Axially Loaded Piles in Clay".

Behaviour of Offshore Structures, London, 1979. Garas, F.K. and McAnoy, R.P.L. Mechanics, Mexico 1980. Idriss, I.M., Dobry, R. and Singh, R.D. "Non-Linear Behaviour of S o f t Clays "Offshore Piled Foundations with Particular

R e f e r e n c e t o Behaviour Under Cyclic Tension", Int. Symp. on Marine Soil

During Cyclic Loading". Jnl. Geot. Eng. Divn, ASCE, ~ o l 104, No. GT12, 1978. .

Lee, K.L. & Focht, J.A.

"Strength of Clay Subjected T o Cyclic Loadingt1.

Marine Geotechnology, Vol. 1, No. 3, 1976. Lupini, J.F. Skinner, A.E. and Vaughan, P.R. "The Drained Residual S t r e n g t h

of Cohesive Soils". Geotechnique 31, No. 2, 198 1. Marsland, A. and Powell, J.J.M. Loading, Vol. 2, Swansea, 1980. "Cyclic Loading T e s t s on 865mm Diameter

P l a t e s in Stiff Clay Till", Proc. Int. Symp. on Soils under Cyclic and Transient

Matlock, H.,

Meyer, P.L.

and Holmquist, D.V.

"Axcol 3: A Program for

Discrete E l e m e n t Solution of Axially Loaded Members with Linear or Non-Linear Supports". Texas, March 1976. Meyerhof, G.G. "Bearing C a p a c i t y and S e t t l e m e n t of P i l e Foundations", Jnl. Report t o t h e American Petroleum Institute, Austin,

Geot. Eng. Divn., ASCE, Vol. 102, GT 3, 1976. McAnoy, Cashman and Williams. T o b e Published Poulos, H.G. "Analysis of Cyclic Axial Response at a Single Pile", Research

Report R362, Univ. Sydney, Sch. Cw. Eng, March 1980. Poulos, H.G. "Cyclic Axial Response of Single Pile" Jnl. Geot. Eng. Divn.

ASCE, Vol. 107 No. G T l , 1981. Randolph, M.F. and Wroth, C.P. "Analysis of Deformation of Vertically

Loaded Piles". Jnl. Geot.Eng.Divn, ASCE, Vol. 104, No. GT 12, 1978.
--

S m i t h J.R. and Taylor R.S.

"The Development of Articulated Buoyant Column

Systems as an aid t o Economic Offshore Production" P a p e r 266, Europec 1980. Vij ayvergiya, V.N. "Load-Movement C h a r a c t e r istics of Piles". P o r t s '77

Conference, Long Beach, California, 1977. Vijayvergiya, V.N. and Focht, J.A. "A New Way t o P r e d i c t t h e C a p a c i t y of

Piles in Clayu, Offshore Technology Conference, Houston, 197 2, API R P ZA, Planning, Designing and Constructing Fixed Offshore Platforms, API Washington D.C. 1980. Whitaker, T. "The Constant R a t e of Penetration T e s t for t h e Determination of t h e U l t i m a t e Bearing Capacity of a Pile", Proc. Inst. Civ. Eng. Vol. 26, . . 1963. Whitaker, T. and Cooke, R.W. Inst. Civ. Engs. London 1966. "Investigation of S h a f t and Base Resistance of

L a r g e Bored Piles in London Clay", Proc. Symposium of L a r g e Scale Piles.

Table 1. Typical Soil Properties 0-10.0m Depths Property

-

Value 5 1m

Comment

Over consolidation r a t i o Depth t o ground w a t e r Index P r o p e r t i e s Water C o n t e n t Plast Limit Liquid Limit Strength

16% 20% 40 %

- Deformation P r o p e r t i e s - 4.'5m d e p t h s - 10.0m

170 120 K N / ~ ~ ) Pressuremeter

Undrained Shear S t r e n g t h (Cu) 0.0 4.5

K N / ~ ~ ) result*

Effective Strength Parameter:

C'

o0
27' 24 O 12 17 MN/m2 MN/m2

) Undrained triaxial ) test result

6'
@rl
Shear Modulus

Ring shear result Pressuremeter resultt 865mm P l a t e result0

*
+
0

Calculated f r o m t h e equation C u = (PL Average G for 1% strain

- PH0)/6.18,

where

PL = limiting pressure & PHO = original horizontal stress in t h e ground. S e c a n t modulus over range P v O t o P V O+ 1 (qU - PVO),where qu = u l t i m a t e base pressure & P v o = original vertical stress in t h e ' ground.

Table 2. Summary of Ultimate Tensile Capacity T e s t No. Measured Value (kN) Predicted Values (kN) and PredictedIMeasured Values (%) Alpha kN El E4 495 400 410 410
%

Lambda kN 430 430

Beta kN 270 270

%

16
87

83 102

54 67

107

Depth

SOIL DESCRIPTION TOP SOlL WEATHERED DARK BROWN CLAY TlLL

OLL

PL

'10C LAY

(ml

10% 20

30

40

10

20

30

40

2.0- WEATHERED DARK

BROWN STIFF 3.0- FISSURED T l L L

4'05.0

WEATHERED BROWN FlRM TlLL

UNWEATHERED 6.0- DARK GREY BROWN FlRM T l L L

7.0 8.0-

r q 9.0-

- 10.0
Sleeve F r i c t i o n ( l o 3 k

~m 2 l)

Cone R e s i s t a n c e ( lo3 k N l m 1

0.3

0.2

0.1

FIGURE 1 (a) Borehole Log (b) Cone Penetrometer Log.

Ground Level

Pore Pressure Transducer Flush With Cover Plate Total Pressure Transducer Flush With Cover Plate

Cover P l a t e

Strain Gauge on Pile Wall

Cover PlateJ

\pressure

Transducer

FIGURE 2 Pile Instrumentation

17-7-80 4.8 -80 5 -8.80

Date

6 -8.80

FIGURE 3. Load History o f the Piles (a) Previous

Loading During Group Test ( b ) Single Pile Test.

Cyclic Test A

Cyclic Test 0

Cyclic Test C.

30 Cumulative Displacement

15
(mm)

FIGURE 4 .

Constant Rate o f Extraction Tests To Failure, E - E6 l

_ Y -

Peak
Cyclic Test A

1.0e

- Mean
--

0 1
No. o f Cycles

Cyclic Test 0 (1942 193kN) 0.50

,t $ 4* +
I

E E
C

. ,

a J
E
d

aJ U
V)

'

'

' I

No. o f Cycles

i3 1.51.00.5-8

.
I

Mean

Cyclic Test C (123 2,122 k N )

" " I

'

' 1

10

1,000 10,000 No. o f Cycles

100

FIGURE 5 . Pile Head Displacements During

Cyclic Test I3

Cyclic Test C
a

- a

Cyclic Test A

Number Of Cycles

1.5-

~ Y~ ~ ~
-(194

Cyclic Test B kN 1

E
Y

1.0Cyclic TestA t 97 k N 1

+ c

E a
m

U
d

V) .-

n
a

0.5:,
C
rg

r :

Cyclic TestC (123 2 122 kN

Number O Cycles f

100

I ,000

FIGURE 6 Pile Head Displacements(a) Cyclic Amplitude

(b)

Pile Head Load

Test

Static, Before Cycling

Peak, During Cycling

1 I
0

Ultimate During ' Extraction

A,C

B / C

E3
Axial Load (kN)

FIGURE 7 Axial Load Distributions During Testing.

lnlnai Static

Final

-- - -

-.

---

Description o f Modulus Initial Static Loading Cyclic Effective,

loth

Cycle
I1

100 t h 250 t h 500 th 1,000 'h 10,000 t h

Test A K G 262 76 (252) 255 (215) 249 231

II II

Test B K G 262 77 (252) 226 (198) 2 11 198 139 50

Test

K G 251 63 (256) 242 (233 1 233 2 28

217 196

, I

220 217 255 70 1 2 2 8

/

F i n a l Static Unloading

50

251

K - Pile Head S t i f f n e s s I k N l mm ( K ) - S t i f f n e s s , Derived From CRE Tests Over Same Load Range
G

A p p a r e n t Shear Modulus a t Mid-Depth of Pile (.MN l r n 2 )

FIGURE 8 Variations in Pile Head Stiffness and Apparent Soil Shear Modulus
P 17

No Load

Element No.

Tensile Load

Depth t o Centre (m) -0.25 - m +0.5 -

FIGURE 10 ( a ) Division o f Pile ~ n t o Elements ( b ) R e p r e s e n t a t i o n o f T- Z Curves

Pile Head Load

Depth (m

a 0 Shear Stress ( k ~ l r n ~ )

100 200 300 400 P i l e Head Load ( k N) (b)

FIGURE 11 ( a ) Distribution of Average Shear Stress & ( b) Corresponding Elemental Displacements a t Various Stages During Extraction E4

Element 3

I Depths l- 3 m 1
A f t e r Cycling, Test A Predicted
C

--/+ -

-----

-_----___I

---c--

A f t e r Cycling, /Test A
&

Predicted
-CC-

----

Displacement Z

( mm )

FIGURE 12 Comparison Between Measured and Coyle & Reese T - Z Curves.

C r i t i c a l V e r t i c a l Displacement , Zc 0.5 1.0

I

, (mm)

'

1.5
I

FIGURE 13 Critical Vertical Displacement Versus Depth

KEY Measured, Element 3 Measured , Element 4 x Measured ,Element 5 A Measured ,Element 6 . - Measured ,Elements 3 - 6 --- Coyle & Reese ,Curve C ' Vijayvergiya

--

FIGURE 14 Normalised

T-Z

Curves