Your browser does not support script

Fourteenth Southeast Asian Geotechnical Conference - Hong Kong

The use of high capacity hydraulic injection piles for buildings in limestone ex-tin mining sites in Kuala Lumpur

1. INTRODUCTION

Tin mining activities in the limestone formation of Kuala Lumpur were rampant in the past. The mining activities left behind numerous ponds and remnants mainly consisting of sand and clay slime, forming a highly heterogeneous overburden on the limestone. Many ex-mining areas have become dumping grounds thus making the overburden material above the limestone more complicated.

As land is getting scarce due to rapid development, previously abandoned ex-mining areas become valuable lands for developments. Conventional pile driving and bored piling encounter uncertainties and difficulties due to heterogeneous nature of the overburden materials and erratic Karstic features of limestone as illustrated in Figure 1.

Gobbett & Hutchison (1973) describe the Kuala Lumpur limestone as "Upper Silurian marble, finely crystalline grey to cream, thickly bedded, variably dolomitic rock. Banded marble, saccharoidal dolomite, and pure calcitic limestone also occur".

The unconfined compressive strengths (UCS) of the Kuala Lumpur limestone from a few locations are summarised in Table 1 below.

Locations

Borehole No.

Data no.

UCS, MPa

Range

Avg.

Sentul, site 1

29

46

25 - 85

48

Sentul, site 2

62

113

8 - 117

46

Ampang

15

19

12 - 92

40

Komoo (1989)

-

-

28 - 120

-

Tan & Ch'ng

(1987)

-

108

10*- 120

-

*Low values due to failures along fractures/fissures

A typical soil profile from an ex-mining site is shown in Fig. 2. The soil is erratic, consists of very soft clayey to very loose sandy materials. Concrete blocks are encountered in some boreholes. Some abrupt increases in N values within the dump are suspected due to construction wastes. The thickness of the dump is up to 14 m in Borehole ABH-2. A test excavation carried out at the site showed the dump material highly heterogeneous, consisting of concrete blocks, wood, steel bars, metal scraps etc alongside with earth.

Cavities, with vertical dimensions of up to 3 m, are encountered. A limestone overhang is suspected in borehole ABH-8 because the material above and underneath the overhang is of the same nature.

2. PILE FOUNDATION

Bored piles are usually preferred over driven piles in limestone areas due to the following concerns :
  • Should there be cavities with roofs of inadequate thickness, there will be risk of collapse of the roofs if piles bear on the roofs. Bored piles can penetrate through the cavity roofs and socket a sufficient depth into the bedrock. The capacity of the bored piles will be ensured. Bored piles also overcome the problem of premature termination of driven piles on hard lenses, floaters or overhangs above bedrock.
  • Due to erratic limestone rock surface, piles tend to deviate during driving although provision of pile shoes and proper control of driving energy may be able to reduce this phenomenon. Pile deviation results in excessive pile length and pile damages. Quite often the integrity of piles can be affected without showing visible signs of damages.
  • Driving of piles is not allowed in many populated parts of Kuala Lumpur due to noise pollution.
However, bored pile solution is costly and slow in construction compared to driven piles. Stringent construction control is required to ensure the quality of bored piles. Due to the nature of overburden material and limestone bedrock, difficulties can be encountered :
  • Many ex-mining areas become dump grounds. Boring would be frequently hindered by debris in the dump such as concrete blocks, steel bars, metal scraps etc
  • Bored holes are not stable when slime or very soft/very loose material, a common feature of ex-mining remnants (such as those in boreholes ABH-5 & -6, Figure 2), is encountered. Although casing can be provided to overcome this problem, collapses of bored holes are still possible as slimy soil can enter the casing through gaps between the casing and uneven limestone rock surface. There were cases where grouting treatments were needed prior to bored pile construction due to the existence of slimy material.

3. JACK-IN PILING SYSTEM

In Malaysia, jack-in piling system was first developed about 12 years ago. It started with a low jacking capacity of 20 tonnes for small RC piles of 10 tonnes working load as an alternative to the tanalised timber piles. The jack-in capacity was increased to 60 tonnes and later 90 tonnes.

In the mid-nineties, there were enquiries from the consultants and the government authorities to look into developing a jacking system that could go beyond the 100 tonne capacity. Such a system was found in China and the jacking capacity was 400 tonnes and later increased to 600 tonnes. It handles square concrete, 250 mm to 550 mm in size; circular concrete spun piles, 0.3 m to 0.6 m diameter, and steel H-piles. The working capacity of piles installed is similar to normal driven piles, in the range of 1000 kN to 2500 kN if ground condition permits. Figure 3 illustrates the set-up of the jacking system. The specifications of the system are given in Table 2.

In view of the above shortcomings and difficulties associated with bored piles and driven piles, pile jack-in piling method offers an alternative.
The advantages of pile jack-in piling method include :
  • Noise- and vibration-free, no mud slurry, no excavated material to be disposed.
  • Sound quality compared to bored piles, as piles are pre-cast and installed by jacking in.
  • No hard driving, no uncertainty of in-situ underground concrete casting.
  • Much faster than construction of bored piles.
  • Capacity of each pile installed is verified by a jack-in force up to two times design loads (DL) or higher.
  • Obstructions in the dump material as mentioned above are not a concern. When a pile is jacked under a force of 2DL or higher, the obstacles will be pushed aside or dragged all the way down to bearing stratum or bedrock (In the latter case, the decaying of the material might be a concern).

Table 2.  Specifications of a 6000kN capacity jack-in piling system

Item

Unit

Value

Main Body

Movement

X-Direction

m

3

Y-Direction

m

0.55

Maximum Turning Angle

Degree

20

X-Y Direction

Movement Speed

Forward

M/min

1.8

Reverse

M/min

4

Pile Size

Maximum

m

0.55x0.55

Minimum

m

0.30x0.30

Bearing Pressure

Long Sleeper

Ton/m2

14.2

Short Sleeper

Ton/m2

17.5

System

Dimension

(L x W x H)

m

11.1x10x9.1

Self Weight

ton

200

Counter Weight

ton

430

4. DESIGN CONSIDERATIONS

4.1 Pile capacity

During pile installation, the jack-in force for every 0.3 m (1ft) penetration is recorded. A 'set' is taken when the jack-in force is picking up indicating the pile has reached a bearing stratum. Normally, the pile is jacked to sustain a force of 2DL. This mechanism is similar to a static load test conducted in an accelerated manner on the pile. The set adopted is verified by pile load tests. Since every pile is installed in the same manner, the capacity of each pile is verified. More important, in erratic limestone bedrock, a good contact between the pile tip and bedrock is ensured by the high jack-in force. The problems of pile deviation and damages due to erratic limestone surface and hard driving experienced by pile driving are significantly reduced.

4.2 Premature pile termination

The concern of premature pile termination arises due to the existence of floating hard lenses (high N value material). The problem can be overcome by applying a higher jack-in force when necessary.

With the help of boreholes, locations of hard lenses can be first identified. 'Pilot' piles can be installed at such locations to see if pile penetration is really a problem. Most of the time, hard lenses or boulders can be penetrated or pushed aside with some increase in jack-in force. When a thicker hard lens is encountered, jack-in force of up to 2.5DL or above should be tried. When proven futile, although quite unlikely, pre-boring can be introduced.

Premature pile termination is also a concern when limestone floaters and overhangs are encountered. In a previously mined site, floaters would be exposed during mining operation and sank to bedrock after the materials underneath had been mined; Overhangs are normally cliff-like rather than in the form of a 'diving board'. This can be observed from the exposed limestone in various parts of Kuala Lumpur and in Gobbett & Hutchison (1973).

Boreholes can be carried out to identity the feature of limestone and subsoil condition when pile length varies significantly within a pile group. If the material underneath the shorter piles is less competent, an appropriate remedy that may include downgrading of pile capacity with respect to the strength of the material underneath is required in order to control differential settlement. Wyllie (1992) suggests performing pile group analysis to ensure the load distribution amongst the piles to be within DL. When the difference in pile lengths is significant between adjacent pile groups, differential settlement amongst the pile groups should be assessed.

4.3 Cavities

Cavities are a common feature in limestone. They associate with few types of failure mechanisms, namely local crushing, punching shear failure and bending failure of cavity roofs.

4.3 Local crushing

Kaderabek & Reynolds (1981) highlight the possibility of local crushing failure when contact pressure exceeds the confined compressive strength of limestone. They further suggest that factor safety of 2 is available when the unconfined compressive strength (UCS) of the limestone is not exceeded.

This failure is not a concern for jack-in piles as high jack-in forces verify the contact between pile toes and limestone.

4.3.2 Punching shear failure

Since every single pile is jacked to sustain a jack-in force of up to 2DL, The pile should punch through any weak roof and thus rule out the possibility of punching shear failure. In the case of pile groups, the possibility of punching shear failure can be assessed using the similar approach for the calculation of pile group effect. With adequate pile spacing, pile group effect can be ignored simply because the perimeter of the pile group is larger than adding up the circumferences of all individual piles in the pile group. A spacing of 3 pile diameters is adequate for pile groups consisting of up to 4 piles arranged in a quadrilateral pattern. For larger pile groups, pile spacing should be assessed according.

4.3.3 Bending failure

The bending moment (M) and tensile stress (T) of a cavity roof can be evaluated by the methods of Roark as suggested by Wyllie (1992). The cavity roof is assumed circular in shape and simply supported around the edges.

M = Q [(1+P)ln(r/ro)+1] / (4P)       (1)

st = 6M/H2        (2)

where Q = load on the roof; P = Poisson's ratio of the limestone; r = radius of the cavity; ro = a parameter depends on the radius of loaded area and thickness of the roof.
ro = P 1.6(D/2)2+H2 - 0.675H for D<H        (3a)

ro = D/2 for D>H       (3b)

where D = equivalent diameter of the loaded area. The authors assume the loaded area is the area enclosed by the pile group; H = thickness of the cavity roof.

Tensile strength tests for rock are not commonly carried out, not to mention rock masses. Hoek (1983) approximates the tensile strength of rock mass by:
qt(m) = 0.5qu(r)[m-(m2+4s)0.5]         (4)

The tensile strengths of carbonate rock (e.g. dolomite, limestone and marble) masses are given in Table 3.

Kaderabek & Reynolds (1981) report that full-scale load tests were carried out to induce a beam tension failure. Such a failure did not occur even the maximum tensile stress, based on theoretically computation (by Roark's method) exceeded by a factor of 2 the tensile strength determined by laboratory splitting tests on 4" core samples. They further remark beam tension failure has not been reported in south Florida area. The median strength of UCS of the Miami limestone is 1.5 MPa (Wyllie 1992).

Table 3. Tensile strength of carbonate rocks expressed as percentage of UCS of intact rock samples (Hoek 1983).

Description

Empirical constants

qt(m) /qu(r)

Intact rock samples

m=7, s=1

14%

Very good quality rock mass, joints 1 - 3m

m=3.5, s=0.1

2.8%

Good quality rock mass, joints 1 - 3m

m=0.7, s=0.004

0.6%

Fair quality rock mass, joints 0.3 - 1m

m=0.14, s=.0001

0.07%

Poor quality rock mass, joints 0.03 - 0.5m

m=0.04, s=0.00001

0.02%

Roof failure by bending may not be such a concern. The shape of small caves in limestone is more of arch then flat slab as observed by Wilford (1964), as illustrated in Figure 4. Equations (1) and (2) tend to overestimate the bending moment and tensile stress experienced by cavity roofs significantly for an arch shape cavity where loads are mainly converted into compression on the arch.

Nonetheless, under jacking force of up to 2DL on every pile installed, competency of cavity roofs is verified in a manner that can only be superseded by a static load test conducted on the pile. Since the number of load tests is limited, jack-in piling method implements a verification process far more reliable than bored piles and conventional pile driving.

Using equations (1) to (3) above, the tensile stresses experienced by the roofs of cavities of various radii and roof thickness, supporting a 4-pile group are tabulated in Table 4.

It can be observed that tensile stress decrease rapidly with cavity roof thickness H. When H exceeds 5 m, tensile stress reduces to less than 0.35 MPa for cavities of up to 10m in diameter. This tensile stress is about 0.8% of the average UCS of Kuala Lumpur limestone of 45 MPa.

Table 4. Tensile strength (in kPa) of cavity roofs supporting a quadrilateral patterned 4-pile group. Piles are 0.4 m square piles at 1.2 m spacing. Column load is assumed 4DL, DL being 1500 kN.

Roof thick-ness, H(m)

Cavity radius, r (m)

2

3

5

1

5173

7165

8995

2

1380

1742

2200

3

527

688

891

4

250

340

454

5

133

191

265

It is considered cautious to postulate that, for small pile groups consisting up to 4 piles, with pile spacing of 3 pile diameters and DL of up to 1500 kN, bending failure of cavities is unlikely when their roof thickness exceed 5 m.

Nevertheless, where there is uncertainty, the soundness of limestone underneath major pile groups should be verified by boreholes. Rafts or isolated mats that enable larger pile spacing is a safe approach as ro in Equation (1) can be increased, thus reducing the tensile stress on cavity roofs.

5. CASE HISTORY

A case history of jack-in piling method is presented. The development included ten blocks of 16 to 18 storey low-cost public housings. The site was located in an ex-tin mining land. A typical subsoil profile is shown in Figure 2.

The foundations are pile-supported rafts thus eliminating the concern of limestone cavities described in Section 4.3 above, particularly bending failure of cavity roofs. Nevertheless boreholes were sunk at lift cores and major column positions to identify the features of limestone. For a building block with typical footprint of 30 m by 88 m, 40 numbers boreholes were sunk. The total number of boreholes was some 400. Cavities were encountered in 15% boreholes. The maximum vertical dimension of the cavities was 3 m.

The piles adopted were 0.4 m grade 45 square RC piles, DL of 1200 kN. In areas where cavities were found to be more prevalent, DL of piles was downgraded to 800 kN. Total pile points were about 4700. Jack-in piling method was used. Three jacking rigs, two 6000 kN capacity and one 4000 kN capacity, were employed. The average production rate was 8 piles points/rig/10-hours-shift. Pile lengths varied from 15 m to 45 m, average 22 m.

The piles were jacked to sustain a jack-in force of 2400 kN to 3000 kN regardless of DL of 1200 kN or 800 kN. The higher jack-in force was tried when premature pile termination was suspected or when a more stringent set was desirable. A total of 15 pile load tests were conducted. Maximum test load was 2DL. All piles tested were reported to perform satisfactorily.

The percentage of unset piles (most likely due to toe slippage) encountered was about 2%, considered low compared to much higher percentage for driven piles in similar soil condition.

At two lift cores where multiple cavities were encountered and the limestone was considered less favourable, mini bored piles (locally known as micropiles) penetrating through the less favourable limestone were adopted.

6. CONCLUSIONS

High capacity hydraulic injection (or jack-in) piling with capacity of 6000 kN is able to install piles of design load up to 2500 kN when ground condition permits. It is a reliable alternative to conventional pile driving and bored piles with same range of pile capacity.

Because every pile is jacked to sustain a force of up to two design loads or higher, the pile capacity is verified in a manner similar to a static load test on the pile. Since load tests can only be conducted on limited number of piles, jack-in piling offers a pile capacity verification process far more reliable than bored piles and conventional pile driving.

With proper design considerations and the help of boreholes, uncertainties encountered by jack-in piling in limestone can be minimised.

Premature pile termination due to obstacles can be minimised by applying higher jack-in force. The differential settlement between adjacent columns has to be assessed should there be a concern of piles of one column terminating above the actual bearing stratum or limestone.

Pile toe slippage is much reduced for jack-in piling as compared to conventional driven piles.

Punching failure of cavity roofs is not a concern when there is adequate pile spacing to avoid pile group effect.

Local crushing of cavity roofs is unlikely as high jack-in force verifies the contact between pile toes and limestone.

In view of the strength of Kuala Lumpur limestone, bending failure of cavity roofs is unlikely for small pile groups consisting up to 4 piles, with spacing of 3 pile diameters and design load of up to 1500 kN.

Bending failure of cavity roofs may not be such as concern as the roofs of small caves in limestone is more of arch shape then flat slab as observed by Wilford (1964). Nonetheless, rafts or isolated mats that enable larger pile spacing is a safe approach.

It is always prudent to sink boreholes at major pile group positions to determine the soundness of limestone underneath. At locations where limestone features are found to be unfavourable, mini bored piles (micropiles) should be introduced as mitigation.

A case history illustrates successful implementation of jack-in piling for a development consisting of 10 blocks of 16- to 18-storey buildings on an ex-mining land in Kuala Lumpur limestone. The pile points were some 4700 numbers.

7. REFERENCES

Gobbett, D.J., 1965, The formation of limestone caves in Malaya, Malaya Nature Journal, 19(1): pp. 4 - 12.

Gobbett, D.J., Hutchison, C.S. 1973, Geology of the Malay Peninsula, New York: Wiley-Interscience.

Hoek E., 1983, 23rd Ranking Lecture: strength of jointed rock masses, Geotechnique, Vol. 33 No. 3, pp. 187 - 223.

Kaderabek, T. J., Reynolds, R. T., 1981, Miami limestone foundation design and construction, J. Geot. Engr. Div. GT7, Jul. 1981, pp. 859 - 872.

Komoo, I. 1989, Engineering geology of Kuala Lumpur, Malaysia, Proceedings of the int. confr. on engineering geology in tropical terrains, Bangi, Malaysia, 26-29 June 1989, pp. 262 - 273.

Sowers, G. F. 1975, Failures in limestones in humid subtropics. J. Geot. Engr. Div. GT8, Aug. 1975, pp. 771 - 787.

Tan B.K., Ch'ng, S.C. 1987, Urban geology of Ipoh and Kuala Lumpur, Forum on urban geology and geotechnical engineering in construction, 1st July, 1993, P. J. Malaysia, organised by IEM & GSM, pp. 1-25.

Wilford, G.E. 1964, The geology of Sarawak and Sabah Caves: 12 - 16, Bull. 6, Geol. Survey Dept. Borneo Region, Malaysia.

Wyllie, D. C. 1st ed. 1992, Foundations on rock: 124 - 126, London: Chapman & Hall, E & FN Spon.

Copyright 2006 G-PILE SISTEM SDN BHD, All Rights Reserved.