Design and Construction of Pile Foundations for the Seoul Worldcup Stadium

 

B. H. Kang

Professor, Inha University

K. B. Kim

Chief Engineer, Samsung Engineering Co., Ltd.

C. W. Cho

Director, Piletech Consulting Engineers

H. S. Hong

Principal, Piletech Consulting Engineers

M. W. Lee

Principal Researcher, Piletech Consulting Engineers

 

ABSTRACT: The site for the construction of the new stadium for the 2002 Worldcup is in the western part of Seoul. The geotechnical condition of the site is rather poor and pile foundations should be adopted. In the pile foundation design, various methods such as conventional static analysis, wave equation analysis and test piling were carried out. From the test piling, hammer efficiency, suitability of driving system, time effect, etc. were confirmed. By taking various parameters into consideration, a most cost effective pile design was made. And the quality control scheme was prepared based on the results of test piling so that the safety of the structure could be ensured.

 

1.     INTRODUCTIon

In 2002, Korea will host the 17th Worldcup Football Games jointly with Japan. The Opening Ceremony will be held in Seoul, Korea. The Seoul City Government decided to construct a new football stadium for this purpose. The new stadium is to be built at Sang Am Housing Development Area, Mapoku which is in the western part of Seoul.

The main goal of the construction is to provide convenient and comfortable facilities for the players, spectators, administrators, press and VIPs during the game, and is to provide the best media facilities, and safer stadium, so as to achieve a successful and flawless Worldcup results.

For the design of the new stadium, FIFA(Federation International de Football Association) guidelines were strictly followed. The all-seater stadium was designed to accommodate 63,930 seats which include 805 VIP seats and 2,024 seats for the press.

The general appearance of the superstructure was developed from the shape of the Korean traditional soban(a small rectangular dining table) and moban(a small circular or octagonal dining table). The roof of the stadium was from the idea of Korean traditional bangpaeyun(a kite in the shape of the ancient shield) and Korean traditional roof lines. Overall the global image of the stadium resembles the images of hwangpodotbae(traditional ship with yellow canvas sails) which operated in the nearby Han River until several decades ago. Fig. 1. shows the general appearance and the general plan of the stadium.

The total project area for the main stadium and training ground is 216,712m2 and the total floor area is 128,138m2(one basement and 6 floors above ground level). The construction started in October, 1998 and it is scheduled to be completed in December, 2001. The architect of the project is Mr Choon Soo Ryu of the Beyond Space Group and the leading contractor of the consortium of five companies is Samsung Engineering Co., Ltd. The pile foundation design and quality control were carried out by Piletech Consulting Engineers.

As the geotechnical characteristics of the site clearly indicated that relatively soft layers located at some depths, deep foundations should be required. Considering the surrounding circumstances around the project site, PHC(spun pretensioned high strength precast concrete) piles were chosen to be driven by hydraulic hammers. To establish the initial design criteria, pile driving was simulated via WEAP(wave equation analysis of pile driving) program. Test piling was carried out prior to the final design.

At the time of driving test piles, pile bearing capacity and driveability were investigated by performing dynamic pile loading tests using PDA(pile driving analyzer). Time effect was confirmed by carrying out dynamic pile loading tests at some time after the pile driving. And the reliability of dynamic pile loading tests was confirmed by performing the static pile loading tests for the same piles which were load tested dynamically.

Prior to driving of the actual working piles, trial piles were driven to confirm the performance of the delivered hydraulic hammers and to make sure of the geotechnical conditions of the whole project site area. From the results of the investigation, quality control criteria were drawn for each hydraulic hammer as well as the location. During the period of the construction of the permanent piles, both the dynamic and static pile loading tests were performed for randomly selected piles.

This paper briefly introduced the design process and the quality control scheme for this project.

 

 

 

 

(a) General appearance

 

 

 

(b) General plan

 

Fig 1.  General appearance and plan of the stadium


 

 2.     Geotechnical Characteristics of the Site

The site is located near the Han River. To investigate the geotechnical characteristics of the site, a total of 80 boreholes were drilled. The boreholes were arranged at 40m interval. It is usual to carry out SPT(standard penetration test) in most of the site investigation in Korea. In addition to the SPT, undisturbed samples were taken at some selected locations for the laboratory tests. Also in-situ permeability tests and pressuremeter tests for the rock were performed.

Table 1 shows the summary of the site investigation. Since the site is relatively large the detailed geotechnical characteristics vary rather significantly. In general, however, the subsurface layers can be divided into four main categories, namely fill, alluvial deposit, residual soils and the bedrock.

Fill layer consists of sandy gravel(GP-GM) and silty sand(SM). In some area, boulders and wastes exist in fill layer. Although SPT results show very irregular change, relative density generally represents medium to very dense state. Municipal wastes were found in southern part of the ground. Its thickness ranges 1.3m to 8.8m and density represents soft state.

The upper part of the alluvial deposit layer is composed mainly of fine grained soils, while the lower layer exhibits rather granular particles. The soils deposited in the upper part are classified as ML or CL. The SPT N values of the layer vary from 4 to 37. The lower layer soils are classified into SM, SP, GP or GW. The thickness of the alluvial deposit ranges from 0.4m to 16.2m. The relative density of the deposit increases with depth and the SPT N value at the bottom part of the lower layer alluvial deposit exceeds 50.

In the northern part of the site residual soil layer was found and the thickness of this layer was in the range of 0.8m to 4.4m. The soil particles were identified as silty sand formed by weathering of the bedrock. The measured N values were in the range of 17 to 50.

The bedrock is identified as banded biotite gneiss. The degree of weathering decreases rapidly with depth.

 

 

 

Table 1. Summary of investigation of bore holes

Name of layers

USCS

Thickness

(m)

N

value

Remarks

Fill

Back

fill

SP, SM

GP, GM

0.9~13.2

2~50

Artificial Fill

Wastes

-

1.3~8.8

3~25

Western part of the ground

Alluvial

Deposit

upper

CL, ML

1.1~12.1

4~37

Alluvium

under

SP, SM

GP, GW

0.4~16.2

5~50

Residual Soil

SM

0.8~4.4

17~50

Northern part of the ground

Bedrock

Banded biotite gneiss

 

 

 

3.     Design of Pile Foundation

3.1   Preliminary Design

In designing pile foundations, various parameters should be considered, such as safety, cost effectiveness, durability, constructability, environmental effects, etc. Considering the surroundings of the project area, it was postulated that driving of precast piles was acceptable. This was later confirmed through actual monitoring of both the ground vibration and noise level during the test piling period. PHC(spun pretensioned high strength concrete) pile was chosen as the suitable pile material. Steel tubular pile was also taken into consideration, however, the idea was discarded because of the corrosion possibility due to the leachate from the municipal wastes deposit.

Two sizes of PHC piles, Φ400mm and Φ500mm were considered for the final design. Pile driving was simulated via WEAP(wave equation analysis of pile driving) program in which hydraulic hammers were utilized for the driving. The driveability of both sized PHC piles were also confirmed by performing test piling at the site. The results of WEAP analysis as well as the test piling clearly indicated that Φ500mm PHC pile was not suitable for this project due to the limitation of the available hydraulic hammer capacity. Consequently it was decided to drive Φ400mm PHC piles by 7ton capacity hydraulic hammer. Table 2 shows the basic material properties of Φ400mm PHC pile.

 

 

 

3.2   Axial Load Carrying Capacity

3.2.1    Analytical Methods

To determine the axial load carrying capacity of the Φ400mm PHC pile both the static and dynamic analyses were performed. In the static analysis, modified Meyerhof’s formula was utilized as the N values were the only available information. The results of the static analysis indicated that the allowable axial load carrying capacity differs from 104 tons to 143 tons depending on the location. In the calculation, piles were assumed to be driven until the N value reaches 50 and the factor of safety is fixed as 3.0 against the ultimate bearing capacity.

In the dynamic analysis driveability was analysed via GRLWEAP(GRL Inc., 1996) program. Fig. 3. is an example of the driveability analysis performed for this project. As shown in this figure, the compressive stress during pile driving pile remained reasonably safe state even at the instant of refusal. This was judged from the allowable compressive stress level of the PHC pile material. On the other hand the tensile stress during penetration of the relatively soft alluvial deposits or the municipal wastes deposit exhibited relatively high level judging from the allowable tensile stress level of the PHC pile material.

 

 

 

In addition to the expectation of relatively high tensile stress, the existence of obstacles such as large boulders, debris from building demolition, worn out tires, etc., was also taken into consideration for the finalization of the construction method. For the problem of relatively high tensile stress during driving, lowering of the hammer stroke as well as reinforcing the pile cushion material would be helpful to reduce the tensile stress level. This was later confirmed by performing PDA(pile driving analyzer) monitoring during the test piling. The problem of subsurface obstacles, however, was not a matter of prediction. It was decided to prebore a hole, when obstacles were identified.

 

3.2.2    Test piling

According to the experiences gained during last several years in Korea, the computer simulation of pile driving by GRLWEAP program has been regarded as an excellent tool of analysing the pile behaviour. Especially, the driveability analysis can provide  more realistic information regarding the pile bearing capacity, suitability of pile driving equipment, damage potential of the pile material during driving, etc. Thus, the conventional pile design can be improved. However there exist some unavoidable limitations in the application of the wave equation analysis of pile driving. Some of the limitations are listed below.

 

¡     the assumed hammer efficiency may differ quite significantly from the actual efficiency of the mobilized hammer

¡     the assumed geotechnical input parameters as determined from the results of site investigation which represents only the limited locations may be different from the real subsurface conditions

¡     the analysed pile bearing capacity may be wrong where significant set up or relaxation occurs

 

In order to overcome these limitations it is necessary to carry out test piling prior to finalize the pile design(Cho et al, 1998). PDA(pile driving analyser) monitoring during test piling period is quite useful to confirm the various points which influence the pile behaviour. From PDA monitoring at the time of test pile driving(EOID, end of initial driving), the hammer efficiency and the suitability of dolly and pile cushion material can be confirmed. Also both the induced compressive and tensile stresses can be investigated so that integrity of the pile material can be confirmed. To take the time effect into consideration it is necessary to perform PDA tests some time after driving for the same pile/s which was/were tested at the time of driving(RESTRIKE). By comparing the EOID and RESTRIKE results, the time effect can be clarified.

Table 3. shows the brief description of the test piling for this project. Test piling of Φ500mm PHC piles were also included to confirm the results obtained from the wave equation analysis. Most of the analyses were done by interpreting the PDA results by CAPWAP(PDI, 1996) program. To confirm the reliability of PDA tests and the CAPWAP analysis, two piles were tested by conventional static load testing method.

 

 

Table 3. The contents of test piling

 

Test Position

No. of Test Piles

Pile Type

DLT

SLT

E.O.I.D

1st

2nd

3rd

1

BH-40

BH-40A1

PHC

Φ400

 

 

 

 

2

BH-40T

 

 

 

 

3

BH-39

BH-39T

4

BH39-500

PHC

Φ500

 

 

 

5

BH-39S

PHC

Φ400

 

 

 

 

6

BH-42

BH-42T

 

7

BH42-A1

 

 

 

8

BH42-A5

 

 

 

 

9

BH42-500

PHC

Φ500

 

 

 

10

BH-26

BH-26T

PHC

Φ400

 

 

 

11

BH-26S

 

 

 

 

12

BH-62

BH-62T

 

 

 

13

BH-56

BH-56T

 

 

 

Remarks : 1) E.O.I.D : End of Initial Driving

          2) DLT : Dynamic Load Test

          3) SLT : Static Load Test

 

3.2.3    Comparison of Static and Dynamic Loading Test Results

Fig. 4. shows the comparison of load-settlement relationship obtained from static loading test and CAPWAP analyses for the same pile. The load-settlement relationships of two different testing methods exhibit almost identical pile behaviour. In this comparison, time effect was one of the most important factors. In order to minimize the influence of time effect, the dynamic loading test was performed immediatly before the commencement of the static loading test. The result of the other comparison was more or less similar, so that the reliability of PDA testing method as well as the reliability of PDA engineer/s was confirmed.

 

 

Fig 4. The comparison of dynamic and static pile loading tests

 

3.2.4    Time Effect

In the traditional soil mechanics and foundation engineering theory, pile bearing capacity is expressed in terms of the effective stress. It has been well known that excessive porewater pressure might be developed due to the impact of pile driving. It takes time for the developed excessive porewater pressure to dissipate and the required time for the dissipation depends on the permeability of the ground. The development and dissipation of the excessive porewater pressure will change the effective stress value and consequently would result in the change of pile bearing capacity.

In cohesive soils with low permeability, the required time for the dissipation of the developed excessive porewater pressure is relatively long, so that pile bearing capacity would increase as time passes. This phenomenon has been well known to the foundation engineers and it is usually called  set up.

This may be true even in cohesionless soils with high permeability. However the time required for the developed excess porewater pressure to dissipate would be much shorter than the time required in cohesive soils. Therefore it has been usually assumed that the pile bearing capacity would not change with time in cohesionless soils.

During the last several years, much research has been focused on this subject, variation of pile bearing capacity with time in cohesionless soils. The results of many researches clearly indicated that pile bearing capacity in cohesionless soils also varies with time. Such achievement is only possible with the rapid development of dynamic pile loading test technique. In Korea, numerous verification examples are available on this subject. In many cases the pile bearing capacity increases with time(set up) in cohesionless soils. In many cases, not as much as the set up cases, the pile bearing capacity in cohesionless soils remains constant with time. In several cases, however, it was confirmed that the pile bearing capacity in cohesive and/or cohesionless soils decreases with time. This phenomenon has been known as relaxation.

From a design engineer's point of view, the time effect is one of the most important factors to consider in the design of pile foundations. Unfortunately, there is not one method available for the practicing engineers which can predict this time efect prior to performing actual field tests.

In order to verify the time effect, a total of 6 piles were selected for PDA monitoring. Pile bearing capacity at the time of driving was tested and pile bearing capacity for the same pile some time after driving was also tested. Two of the piles were tested three times to confirm the time effect. Table 4 briefly summaries the test results and Fig. 5. shows one example.

 

 

Fig 5. Change of load-settlement curve with time

 

Table 4. Summary of the results of dynamic load test

No.

No. of Test Pile

Pile Type

Test Type

Bearing Capacity(ton)

Failure Load by Davisson

Remark (elapsed time after driving)

Shaft Resistance

Toe Resistance

Total Resistance

1

BH-39T

PHC

Φ400

E.O.I.D

57.0

117.0

174.0

104.0

 

Restrike

1st

67.0

139.0

206.0

150.0

1hr

2nd

98.0

140.0

238.0

238.0

3days

3rd

162.0

98.0

260.0

260.0

10days

2

BH39-500

PHC

Φ500

E.O.I.D

46.0

150.0

196.0

117.0

 

Restrike

100.0

158.0

258.0

240.0

4days

3

BH-42T

PHC

Φ400

E.O.I.D

56.0

146.0

202.2

175.0

 

Restrike

1st

140.7

124.7

265.4

265.4

2days

2nd

256.0

88.0

344.0

344.0

9days

4

BH42-500

PHC

Φ500

E.O.I.D

42.0

178.0

220.0

141.0

 

Restrike

122.0

178.0

300.0

300.0

1day

5

BH-62T

PHC

Φ400

E.O.I.D

83.0

142.0

225.3

160.0

 

Restrike

85.0

158.0

235.0

191.0

7days

6

BH-56T

PHC

Φ400

E.O.I.D

80.8

124.7

205.4

164.0

 

Restrike

84.0

194.0

278.0

206.0

7days

 

 

 

As shown in Table 4, the time effect of each pile was not the same. Some piles exhibited quite significant set up phenomenon, while the effect was not so significant in some other piles. Anyhow the results were more than enough for the design purposes, since the allowable axial load of all the tested piles exceeds the maximum allowable load of the pile material as determined by the national building code.

3.3   Final Design

In the design of pile foundations for the Worldcup Stadium, various methods were used:conventional static analysis, wave equation analysis ,test driving and pile loading tests. From the results followings were derived for the final design.

 

¡     Driving of PHC piles does not cause any serious environmental hazards to the surrounding areas.

¡     Φ500mm PHC pile is not suitable for this project, because the available hydraulic hammer capacity is not sufficient.

¡     Φ400mm PHC pile is suitable for this project if driven by 7ton ram capacity hydraulic hammer.

¡     Dynamic pile loading tests using PDA provides reliable information regarding pile bearing capacity.

¡     Time effect is variable depending on the location. In general, set up of pile bearing capacity was found in most of the site area.

¡     Hammer efficiency plays an important role for the final quality of the piles. The hammer efficiency is not constant but varies with driving location, time, etc. Therefore hammer efficiency should be confirmed for every delivered hammer.

¡     In some locations unexpected underground obstacles were identified. If pile driving is found to be impossible, a prebored hole should be drilled prior to pile installation.

¡     The final design load of Φ400mm PHC pile was 85 tons.

 

 

4.     CONSTRUCTION

The design of the pile foundations for the Seoul Worldcup Stadium is one of a few examples in which the state-of-the-art technology has been used. As a result, 85 tons design capacity for the Φ400mm PHC piles was the highest design value in Korea. Consequently a lot of cost and time could be saved.

In order to meet the design requirements, a comprehensive quality control scheme was established. Some of the important quality control schemes are as follows.

 

¡     Perform trial driving of piles to confirm the geotechnical conditions, existence of underground obstacles, suitability of driving system(hammer efficiency, cushion materials, etc.).

¡     Perform EOID tests for every hammer mobilized for this project, to investigate,

- pile bearing capacity at the time of driving

- hammer efficiency

- integrity of pile material

¡     Perform RESTRIKE tests for selected number of piles to confirm the time effect

¡     Perform RESTRIKE tests for several piles after completion of group pile installation in order to confirm the possibility of pile heaving due to group pile installation

¡     Perform EOID and RESTRIKE tests for piles which were installed after preboring

¡     Perform EOID tests for every hammer at every predetermined period to confirm the variation of hammer efficiency with time

¡     For some selected piles static pile loading test is recommended to confirm the reliability of PDA test

 

REFERENCES

Cho, C. W., Lee, M. W., Hong, H. S., Kim, S. H., & Jun, Y. S. 1998. An alternative to enhance the reliability of wave equation analysis of piles. Proc. of the KGS fall '98 national conference. Korean Geotechnical Society : 137-144.

GRL Associates, Inc. 1996. CAPWAP User Manual.

GRL Associates, Inc. 1996. GRLWEAP User Manual.

Lee, W. J., Jun, Y. S., Hong, H. S. & Lee, M. W. 1995. A study on the bearing capacity increase of driven piles with time. Proc. of the KGS spring '96 national conference. Korean Geotechnical Society : 69-89.

NAVFAC. 1982. Foundations and earth structures (DM-7). Department of the US Navy: 7.2-213~7.2-241.

Pile Dynamics, Inc. 1995. PDA User Manual.

US Dpt. of Transportation. 1996. Design and construction of driven pile foundation, FHWA Workshop Manual: 9-67~9-95.