boredpileglostrext aziz lee 10thicpdf 020606

8/2/2019 BoredpileGlostrext Aziz Lee 10thICPDF 020606

1/8

1

Paper published in 10t International Conference on Piling and Deep Foundations, 31st May 2n June 06, Amsterdam.

APPLICATION OF GLOBAL STRAIN EXTENSOMETER (GLOSTREXT)METHOD FOR INSTRUMENTED BORED PILES IN MALAYSIA

Abdul Aziz Hanifah, Senior Assistant Director, Public Works Department, Kuala Lumpur.Lee Sieng Kai, Managing Director, Glostrext Technology Sdn. Bhd., Kuala Lumpur, Malaysia.

A novel instrumentation and analysis technique, called GLOSTREXTmethod has recently been introduced for bored pile load tests inMalaysia. This technique provides an innovative and improvedalternative for conventional bored pile instrumentation methodscommonly practiced for the past few decades. Results for five casehistories involving full scale static load tests for high capacity boredpiles with both new and conventional instrumentation details placedin within the same instrumented piles are presented to demonstratethe advantages of this novel technique. Results show goodagreement between the new and conventional instrumentation.

INTRODUCTION

Over the past few decades, there is anobvious lack of innovation in the area ofinstrumentation and monitoring for theclassical static load tests, while other indirector alternative pile test methods such asDynamic Load Tests, Statnamic Load Testsand Bi-Directional O-Cell Load Testing hadundergone significant improvements in recentyears.

A conventional bored pile instrumentationmethod for static load testing is shown inFig.1(a). In general, vibrating wire strain

gauges and mechanical tell-tales are installedand cast within the pile to allow for monitoringof axial loads and movements at various levelsdown the pile shaft including the pile toe level.

The constraints with this method includes longlead-time required for instrumentation,instruments have to be pre-assembled andinstalled onto the steel cage prior to concretingof the pile, information on pile length andinstrument locations has to be planned andpredetermined before the drilling andinstallation of the test pile.

Strain gauges give localized strainmeasurements and are sensitive to variationsin pile cross-section. Sleeved rod tell-talesoften gives unsatisfactory results due to rodfriction, bowing, eccentricity of loading andreference beam movement. The movementfor lower portions of pile shaft is particularlydifficult to be reliably measured most of thetime.

When instrumentation levels increases for a

particular complex soil strata or geologicalstructure, it is sometimes not practical to puttell-tales at every level due to congestion ofthe sleeved pipes in the piles, as well asdifficulty in monitoring set-up at the pile head.

While development of removableextensometers in full scale static load tests(Bustamante et al 1991) had been verysuccessful and unique in Europe for manyyears now, the advancement had not been

Fig.1(a) : Conventional pile instrumentation

Instrumented Test Bored PileApllied load measured by vw load cells

Pile headPTop Platform level

Verify and back-calculate Ec Strain Gauges Lev. A

PB Strain Gauges Lev. B

Tell-tale Extensometer 1

PC Strain Gauges Lev. C

PD Strain Gauges Lev. D

PE Strain Gauges Lev. E

Tell-tale Extensometer 2PF Strain Gauges Lev. F

Pile toe

Legend:denotes Vibrating Wire Strain Gauges

denotes mechanical tell-tale extensometer


2/8

2


materialized in this part of the world. Thanks toa new generation of pneumatic retrievableextensometer anchors coupled with high-precision spring-loaded vibrating-wire sensors(Geokon, 1995, 2003), highly accuratemeasurements of the relative deformations ofanchored segments across entire pile lengths

are now possible to be logged with relativeease during static load testing.

The papers main focus is to highlight theseveral advantages of GLOSTREXT methodand high correlation between GLOSTREXTmethod and conventional instrumentation fromthe results of five case histories involving bothinstrumentation details placed in within thesame instrumented piles.

DESCRIPTION OF GLOSTREXT METHOD

The GLOSTREXT Method for static load

testing on bored piles is a deformationmonitoring system using advancedpneumatically anchored extensometers and anovel analytical technique for determining axialloads and movements at various levels downthe pile shaft including the pile base level.This method is particularly useful formonitoring pile performance and optimizingpile foundation design.

To appreciate the basic innovation containedin the GLOSTREXT Method, the deformationmeasurement in the pile by strain gauges andtell-tale extensometers are reviewed.Normally, strain gauges (typically short gaugelength) are used for strain measurement at aparticular level or spot, while tell-taleextensometers (typically long sleeved rodlength) are used purely for shorteningmeasurement over an interval (over a lengthbetween two levels).

From a strain measurement point of view, thestrain gauge gives strain measurement over avery short gauge length while the tell-taleextensometer gives strain measurement overa very long gauge length! Tell-taleextensometer that measure strain over a verylong gauge length may be viewed as a verylarge strain gauge or simply called globalstrain extensometer.

With recent advancement in the manufacturingof retrievable extensometers such as state-of-the-art vibrating wire extensometers, it is nowpossible to measure strain deformation overthe entire length of piles in segments withease during static load testing.

Fig.1(b) illustrates typical arrangement ofretrievable vibrating wire extensometers in testpile, permitting improved monitoring of axialloads and movements at various levels downthe pile shaft including the pile base levelusing GLOSTREXT Method. Generally twostrings of retrievable vibrating wire

extensometers with 6 or 7 anchors areadequate to yield improved results forinstrumented static load tests. This system isequivalent to the conventional method of using24 no. VW Strain Gauges and 6 no. sleevedrod extensometers, which might not bepossible to be installed satisfactorily due tocongestion in small-sized piles.

CASE HISTORIES

The results of five instrumented sacrificialbored test piles namely PTP1, PTP2, PTP3,PTP4 and PTP5, involving full scale static loadtests with both new and conventionalinstrumentation details placed within the samepiles are presented. The load tests wereconducted with great care and control asdescribed in following subsections. All themonitoring instruments were measuredautomatically during the loading and unloadingcycles using a data-logger system. Fig.2 givesa typical illustration for instrumentation andmonitoring set-up.

Fig.1(b) : GLOSTREXT pile instrumentation

Instrumented Test Bored Pile

Apllied load measured by vw load cells

Pile headPTop

Anchored Lev. A Verify and back-GLOSTREXT Sensor 1 calculate Ec

Anchored Lev. B

GLOSTREXT Sensor 2 PAve(BC)

Anchored Lev. C

GLOSTREXT Sensor 3 GaugePAve(CD) Length

Anchored Lev. D

GLOSTREXT Sensor 4PAve(DE)

Anchored Lev. E

GLOSTREXT Sensor 5 PAve(EF)

Anchored Lev F

GLOSTREXT Sensor 6 PAve(FG)

Anchored Lev. G

Pile toeLegend:

denotes GLOSTREXT anchored level

denotes GLOSTREXT Vibrating Wire Sensor


3/8

3


Subsurface Conditions and PilesInstrumentation Scheme

The location of the site [Figure 3(a)] is atInterchange No.1 (ICW01), part of SouthernIntegrated Gate Project located at JohorBahru, Johor, Malaysia. The site is underlainpredominantly by weathered residual soils,which consist mainly of silty sand.

Standard penetration tests (SPT N values inblows/30cm) from borehole result at each test

pile location and pile instrumentation detailsare graphically represented in Figure 3(b).

For each of the sacrificial bored piles, twotypes of instruments, namely, the vibratingwire strain gauges and retrievable vibratingwire extensometers were installed internally in

the pile. The strain gauges were installed atsix levels with four strain gauges at each level.The Geokon A-9 retrievable vibrating wireextensometers sensors, housed in a 51mminternal diameter sonic logging pipe wasinstalled at six levels at corresponding straingauges levels, with 2 sets per level.

Piles Structural Properties

The instrumented test piles PTP1, PTP2,PTP3, PTP4 and PTP5 constructed were allbored cast-in-situ reinforced concrete pileshaving structural properties as listed in Table1:

Fig.2: Typical pile test instrumentation andmonitoring set-up

x x

x

x x

x

x x

x

x x

x x x x x - - -

x x x x x - -

x x x x x - - -

x x x x - -

x x x x x - - -

x x

x

x x

x

x x

x

x x

x

x x

N.T.S

0

5

10

15

20

25

30

35

40

45

50

55

60

Depth(m)

Silty

SAND

Platform

Level

Soil Profile

Clayey

SILT

Clayey

Silty

SAND

Silty

SAND PTP 1

Legends:

VW Strain Gauge (4 Nos/Level)

GLOSTREXT Anchored Level

GLOSTREXT VW Sensor

5

6

5

5

9

7

17

18

12

13

16

18

42

39

18

21

43

45

12

14

28

2645

50

23

8

9

21

43

36

100

67

33

3836

0 50 100 1 50

SPT value, N

(blows/30cm)

7

6

4

5

14

9

19

20

23

25

15

41

68

42

49

64

18

40

30

33

51

73

103

143

143

0 50 100 150

SPT value, N

(blows/30cm)

5

9

7

0

6

0

9

7

8

9

11

3

4

11

10

12

14

25

29

16

22

28

24

17

19

22

21

19

21

32

33

37

37

26

32

45

50

27

4045

0 50 100 150

SPT value, N

(blows/30cm)

8

8

14

4

6

11

8

3

4

9

9

10

10

12

11

30

18

22

17

18

64

83

33

47

57

55

79

107

107

0 50 100 1 50

SPT value, N

(blows/30cm)

PTP 2

PTP 3

PTP 4

PTP5

5

7

8

6

7

18

22

19

27

20

50

19

39

32

37

45

37

35

39

25

27

2728

40

43

36

50

50

34

42

50

75

67

7935

0 50 100 150

SPT value, N

(blows/30cm)

Lev. A

Lev. B

Lev. C

Lev. D

Lev. E

Lev. F

DB 3 DB 5 BH 4 DB 6 BH 2

Fig.3(b) : SPT N values and pile instrumentation scheme for PTP1 to PTP5

Fig.3(a): Site location plan

ICW01 site

Johor Strait


4/8

4


Table 1:Structural Properties of Piles

TestPileNo.

PileDiameter

(mm)

PileLength

(m)

MainReinfor-cement

ConcreteOver-break

PTP1 750 47.0 12T20


5/8

5


Load transferred (P) at each level is calculatedas follows:

P = (Ec*Ac + Es*As )

where = average change in strain gauge

readingsAc = cross-sectional area of concreteEc = Concrete ModulusAs = cross-sectional area of steel

reinforcementEs = Young's Modulus of Elasticity in

steel ( = 200 kN/mm2)

The modulus of concrete, Ec was back-calculated with the aid of the strain gaugeresults at level A and the pile top loads. Foreach stage of loading, Ec is back-calculated byassuming that the load at the strain gaugelevel A was equal to the applied load at the

pile top. Load distribution curves acquired from

VW Strain Gauges test results for the testcycle for PTP1 to PTP5 are presented in Fig.5.

Axial Load Distribution From RetrievableVW Extensometers

In the GLOSTRET method, the modulus of

concrete, Ec was back-calculated bymeasuring the strains in the top 2.0m ofdebonded length of the pile using theGLOSTREXT vw sensors and the pile toploads. For each stage of loading, Ec is back-calculated by assuming that the load at themid-point of the 2.0m debonded length levelwas equal to the applied load at the pile top.

Load distribution curves acquired fromGLOSTREXT test results for the test cycle forPTP1 to PTP5 are plotted and presented inFig.6. It is worthy to note that Fig.5 and Fig.6show very similar characteristics.

0

1000

2000

3000

4000

5000

6000

7000

0 20 40 60 80

Pile Head Settlement (mm)

PileHeadLoad(kN)

PTP1

0

1000

2000

3000

4000

5000

6000

7000

0 20 40 60 80

Pile Toe Settlement (mm)

PileHeadLoad(kN)

PTP1

0

1000

2000

3000

4000

5000

6000

7000

0 10 20 30 40

Total Shortening (mm)

PileHeadLoad(kN)

PTP1

0

2000

4000

6000

8000

10000

12000

0 20 40 60 80 100 120


PileHeadLoad(kN)

PTP2

0

2000

4000

6000

8000

10000

12000

0 20 40 60 80 100 120


PileHeadLoad(kN)

PTP2

0

2000

4000

6000

8000

10000

12000

0 10 20 30 40


PileHeadLoad(kN)

PTP2

0

2000

4000

6000

8000

10000

12000

14000

0 5 10 15 20 25


PileHeadLoad(kN)

PTP3

0

2000

4000

6000

8000

10000

12000

14000

0 5 10 15 20 25


Pile

HeadLoad(kN)

PTP3

0

2000

4000

6000

8000

10000

12000

14000

0 5 10 15 20 25


PileHeadLoad(kN)

PTP3

0

2000

4000

6000

8000

10000

0 5 10 15 20 25 30


PileHeadLoad(kN)

PTP4

0

2000

4000

6000

8000

10000

0 5 10 15 20 25 30


PileHeadLoad(kN)

PTP4

0

2000

4000

6000

8000

10000

0 5 10 15 20 25 30


PileHeadLoad(kN)

PTP4

0

2000

4000

6000

8000

10000

0 5 10 15 20 25


PileHeadLoad(kN)

PTP5

0

2000

4000

6000

8000

10000

0 5 10 15 20 25


PileHeadLoad(kN)

PTP5

0

2000

4000

6000

8000

10000

0 5 10 15 20 25


PileHeadLoad(kN)

PTP5

Fig.4: Pile Head Settlement, Pile Toe Settlement and Total Shortening behaviour, PTP1 to PTP5

0

10

20

30

40

50

0 1000 2000 3000 4000 5000 6 000 700 0

Loads ( kN)

Depth(m)

PTP1

0

10

20

30

40

50

0 2000 4000 6000 8000 10000 12000

Loads ( kN)

Depth(m)

PTP2

0

10

20

30

40

0 20 00 40 00 6 00 0 80 00 1 00 00 1 2 00 0 1 40 00

Loads ( kN)

Depth(m)

PTP3

0

10

20

30

40

50

60

0 2000 4000 6000 8000 10000

Loads ( kN)

Depth(m)

PTP4

0

10

20

30

40

0 2000 4000 6000 8000 10000

Loads ( kN)

Depth(m)

PTP5

Fig.5: Load Distribution Curves computed from vw strain gauges test results for PTP1 to PTP5


6/8

6


Fig. 7(a): Ec vs Measured Axial Strain forTest Pile PTP 1

COMPARISON OF RESULTS FROMCONVENTIONAL AND GLOSTREXTMETHOD

Comparison Of Back Calculated ConcreteModulus Values

The plots of back-calculated concrete modulus

values, Ec, versus measured axial strain atlevel A from both conventional strain gauges(at 1.0m depth) and Global StrainExtensometers (from 0.0m to 2.0m depth) forthe test cycles for 5 piles are plotted andpresented in Fig.7 (a) to Fig.7 (e) respectively.

Fig. 7(d): Ec vs Measured Axial Strain forTest Pile PTP 4

Fig. 7(e): Ec vs Measured Axial Strain forTest Pile PTP 5

Fig. 7(c): Ec vs Measured Axial Strain forTest Pile PTP 3

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900

Measured Axial Strain at Lev. A ( x 10-6

)

Concretese

cantmodulus,

Ec

(kN/mm

2)

PTP1, VWSG

PTP1, GloStrExt

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900


)

Concretesecantm

odulus,Ec(kN/mm

2)

PTP2, VWSG

PTP2, GloStrExt

10

1214

16

18

20

22

24

26

28

30

32

34

36

38

40

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900


)

Conc

retesecantmodulus,

Ec

(kN/mm

2)

PTP3, VWSG

PTP3, GloStrExt

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900


)

Concretesecantmodulus,Ec(kN/mm

2)

PTP4, VWSG

PTP4, GloStrExt

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900


)

Concretesecantm

odulus,

Ec

(kN/mm

2)

PTP5, VWSG

PTP5, G loStrExt

0

10

20

30

40

50

0 1000 2000 3000 4000 5000 6000 7000

Loads ( kN)

Depth(m)

PTP1

0

10

20

30

40

50

0 2000 4000 6000 8000 10000 12000

Loads ( kN)

Depth(m)

PTP2

0

10

20

30

40

0 2000 4000 6000 8000 10000 12000 14000

Loads ( kN)

Depth(m)

PTP3

0

10

20

30

40

50

60

0 2000 4000 6000 8000 10000Loads ( kN)

Depth(m)

PTP4

0

10

20

30

40

0 2000 4000 6000 80 00 10000Loads ( kN)

Depth(m)

PTP

Fig.6: Load Distribution Curves computed from GLOSTREXT test results for PTP1 to PTP5

Fig. 7(b): Ec vs Measured Axial Strain forTest Pile PTP 2


7/8

7


From the plots presented, it is clear that theback-calculated concrete modulus valuesmeasured by two independent systems(conventional Strain Gauges and Global StrainExtensometers) agree reasonably well.

These plots are also extremely useful to study

the correction of Ec according to variation ofstrain level with pile depth, which can furtherimprove the accuracy of the axial loaddistribution computation using back-calculatedEc.

Comparison Of Measured Axial StrainAlong Pile Shaft

The plots of measured axial strain at variouslevels along pile shaft from both conventionalstrain gauges and Global StrainExtensometers for 5 piles are presented inFig.8(a) to Fig.8(e) respectively.

Fig. 8(b) : Measured Axial Strain from VWSGvs Axial Strain from GloStrExt forPTP 2

Fig. 8(a) : Measured Axial Strain from VWSGvs Axial Strain from GloStrExt forPTP 1

Fig. 8(c) : Measured Axial Strain from VWSGvs Axial Strain from GloStrExt forPTP 3

Fig. 8(d) : Measured Axial Strain from VWSG

vs Axial Strain from GloStrExt forPTP 4

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700

Axial Strain from GloStrExt ( x 10-6 )

AxialStrainfromVWSGs(x10-6) Lev A

Lev A to Lev BLev B to Lev CLev C to Lev ELev E to Lev FControl Line for VW SG=GloStrExt

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700


Ax

ialStrainfromVWSGs(x10-6)

Lev ALev A to Lev BLev B to Lev CLev C to Lev DLev D to Lev ELev E to Lev FControl Line VWSG = GloStrExt

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700

Axial Strain Fr A9 ( x 10-6 )

AxialStrainFrVWSGs(x10-6)

Lev ALev A to Lev BLev B to Lev CLev C to Lev DLev D to Lev ELev E to Lev FControl Line "VWSG = GloStrEt"

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

800

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800


AxialStrainfromVWSGs(x10-6)

Lev ALev A to Lev BLev B to Lev C

Lev C to Lev DLev D to Lev ELev E to Lev FControl VWSGs=GloStrExt

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

800

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800


AxialStrainfromVWSGs(x10-6)

Lev ALev A to Lev BLev B to Lev CLev C to Lev DLev D to Lev ELev E to Lev FControl Line VWSG=GloStrExt

Fig. 8(e) : Measured Axial Strain from VWSGvs Axial Strain from GloStrExt forPTP 5


8/8

8


From the plots presented, it is shown that theaxial strains measured by the two independentsystems are in good agreement. Consideringthat the Global Strain Extensometers measurestrains over an entire section of a pile, thus itintegrates the strains over a larger and morerepresentative sample than the conventional

strain gauges.

CONCLUSIONS

The results of five instrumented bored testpiles involving full-scale static load tests withboth GLOSTREXT method and conventionalinstrumentation show the following behavior:

i) The back-calculated concrete modulusvalues measured by two independentsystems (conventional Strain Gauges andGlobal Strain Extensometers) agree

reasonably well.

ii) The axial strains measured by the twoindependent systems are in goodagreement. The Global StrainExtensometers measure strains over anentire section of a pile, thus it integratesthe strains over a larger and morerepresentative sample than theconventional strain gauges.

iii) Using the global strain extensometers,measurement of the pile shortening overthe whole pile length can now be reliablymeasured in segments. This enables themovement of the pile and strains atvarious levels down the pile shaft to bedetermined accurately, thus permitting animproved load transfer distribution of pilesin static load tests.

The GLOSTREXT method significantlysimplifies the instrumentation effort byenabling the sensors to be post-installed aftercasting of pile. It also minimizes the risk ofinstruments being damaged during theconcreting process.

Acknowledgement

The authors wish to thank the Public WorksDepartment Malaysia for permission to publishthis paper. The authors also wish to thank theGerbang Selatan Bersepadu Public WorksProject Team for their support.

References:

Bustamante M., Doix B (1991) A new model ofLPC removable extensometer, Proc. 4

th

International Deep Foundation InstituteConference, pp 475-480.

Geokon Inc. 1995, revised 2003. InstructionManual, Model A-9 Retrievable Extensometer.

boredpileglostrext aziz lee 10thicpdf 020606

Documents