Mark R Svinkin Cleveland, Richard D Woods, University of Michigan, USA
Accurate and reliable determination of pile capacity is very important for the proper design, construction and estimation of the cost of foundations. Traditionally, the static loading test is used to determine ultimate capacity of the pile-soil system or the value of a service load to be supported by a pile. In recent decades, because of advances in data acquisition during pile driving and restrikes, dynamic testing has become an integral part of pile capacity prediction and measurement.
Dynamic methods have certain advantages and some uncertainties in their application. Wave equation analysis of driven piles is a prevalent method of pile driving stress calculations. Unfortunately, in most cases, computed pile capacity differs substantially from the results of both static and dynamic load tests.
Dynamic measurements of force and velocity at the upper end of the pile during pile driving, followed by a signal matching procedure, is the most common method for dynamic determination of pile capacity. However, though dynamic methods have been used in practice for years, the actual reliability of dynamic methods is vague because in most cases, their comparison with static loading tests is made incorrectly.
This paper considers some aspects of the verification of dynamic capacity formulas and dynamic testing methods. It also considers the improvement of pile capacity prediction by wave equation analysis.
It is imperative to consider time effects for accurate determination of pile capacity by both static and dynamic methods.
The prediction of pile capacity in pre-diving wave equation analysis can be improved by the use of variable damping as a function of time. Variable damping is the key parameter to enhance the accuracy of wave equation solutions because this damping takes into consideration soil consolidation after pile installation.
The main criterion for the accurate assessment of pile capacity prediction based on dynamic measurements of force and velocity at the upper end of the pile during driving is the ratio of capacities obtained by dynamic and static tests. Such a ratio, taken for arbitrary time between compared tests, is not a verification of dynamic testing results.
Dynamic testing and analysis yield the real, not predicted, static capacity of piles at the time of testing. The static capacity from a static loading test is not a unique standard for the assessment of dynamic testing results. Both static loading test and dynamic testing yields the pile capacity at the time of testing.
In soils with time-dependent properties, a comparison of static loading test and dynamic testing must be made only for tests performed immediately, in short succession.
Installation and loading tests of deep piles in Shanghai alluvium
Warren Pump, Woodward-Clyde International, Melbourne, Australia; Stan Korista, Skidmore, Owings and Merrill, Chicago; James Scott, Woodward- Clyde International, Denver, USA
Static and dynamic loading testing were performed on two identical open- ended steel pipe piles installed to depths of 80m in Shanghai, China. The 914mm diameter piles were each fully strain gauged and driven through soft Quaternary alluvium to found on dense alluvial sands. Load testing established the capacity of both piles. The work formed part of the pre- construction investigations for the 88-storey Jin Mao building, located in the Pudong district of the city. Pile driving was monitored using a Pile Driving Analyser. Static load testing of the two piles was in accordance with ASTM D 1143 with cyclic application of load to a maximum of 18,000kN. The paper provides the results of geotechnical site investigations; descriptions of pile installation; and pile load-settlement behaviour and shaft load distribution.
The test pile programme has also shown that the piles were driven as non-displacement piles, where after driving the soil plug remained in place over most of the pile length; densification of sandy layers due to the driving of neighbouring piles did not significantly add to pile driving resistance for pile spacing of about five diameters; stresses in the piles during periods of heavy driving were well within acceptable limits; static test loading of two test piles indicates similar load-settlement behaviour. Static test loading using the ASTM procedure (D 1143, that is) provides pile load-settlement results that are similar to those obtained using the usual Shanghai test loading schedule, the latter being considerably less time consuming; the method of Davisson appears to represent a rational way of determining load capacity from a pile load test at this site; dynamic monitoring and CAPWAP analysis have provided a reliable indication of static load capacity of both piles and a large proportion of the load resistance of each pile was mobilised by shaft resistance; complete soil plugging appears not to have occurred.
Capacity of model grouted piles in calcareous sediments
Hackmet Joer and Mark Randolph, Centre for Offshore Foundation Systems, The University of Western Australia
Pile foundations are widely used in calcareous sediments around the world. In many locations, these materials are renowned for their compressibility and susceptibility to crushing, resulting in very low frictional capacity for driven piles. The offshore location of these materials has led to a very limited database of field pile tests, and pile design has to rely heavily on laboratory testing.
Piles driven into such sediments generally have very low shaft capacity, due to the crushing and rearrangement for the soil grains adjacent to the pile shaft. However, the frictional capacity may be fully recovered by injecting grout into the zone. Such 'grouted driven piles' are significantly cheaper to install than conventional drilled and grouted piles, and avoid the potential problems of hole stability in the latter type of pile. The reluctance to adopt grouted driven piles more widely has usually been due to uncertainties in the grouting operations, and lack of quantified 'quality control' during construction.
The paper describes a systematic programme of laboratory tests undertaken to explore the grouting operation and subsequent shaft friction in reconstituted calcareous soil. The drivability of single and multiple piles were investigated and the feasibility of grouting multiple piles studied. The effect of depth, or stress level, was also investigated by consolidating the samples under K0 conditions at various confining pressures.
Closed-ended piles of 38mm and 51mm diameter were used in this study and consistent results were obtained between the two diameters. In order to study the time effect ('ageing effect') on the shaft friction, pull- out tests were made at various intervals of time. Although the elapsed time between tests was of the order of seven days, the results showed a relatively small increase in the friction capacity with time. Other parameters, such as the interaction between piles, the grout pressure and the pile diameter, were also investigated.
Effects of construction technique of the behaviour of plain bored cast in situ piles constructed under drilling slurry
Brian Littlechild and Glen Plumbridge, Ove Arup & Partners, Hong Kong
The Bangkok Elevated Road and Train System (BERTS) requires the construction of some 16,000 1.5m diameter bored cast in situ concrete piles (bored piles) founded in sand and clay.
Due to the large number of piles to be constructed for the BERTS project, a number of fully instrumented trial piles were carried out on piles constructed using different techniques to enable the selection of the most economical foundation type for various load cases. This trial piling was aimed at identifying the design parameters compatible with different construction techniques, and from this to then identify the most cost effective foundation solution.
A number of the early trial piles for the project failed, highlighting the impact that construction methodology has on plain bored piles. As a consequence of the failures, some changes were made to the piling specification, and a number of additional instrumental trial piles constructed to verify the compatibility of the design parameters and the construction methodology.
Shaft resistance capacity of the plain bored pile is highly dependent on the construction methodology and materials used for construction.
The shaft friction resistance for bored piles constructed under Bentonite slurry tends to decrease as the construction time increases and as the Bentonite slurry viscosity increases.
The shaft resistance of bored piles constructed under polymer slurry shows no significant change with construction times up to 42 hours, but does show a slight reduction in shaft resistance with decreasing polymer slurry viscosity.
Design parameters and assumptions are therefore applicable only to piles constructed in accordance with a given methodology and material types, and it is necessary for trial piles to be carried out to confirm the applicability of the design parameters and assumptions for a given pile construction methodology. In addition, all contract piles should be constructed in a manner at least as favourable as that of the trial piles. The trial piles should supersede the specification.
Quality assurance and testing on a contract for 16,000 large diameter bored cast insitu piles
AR Chodorowski and MR Duffy, Ove Arup & Partners International
This paper describes the quality assurance (QA) system and procedures adopted during the construction of the bored pile foundations for the Bangkok Elevated Road and Train System (BERTS). The QA system not only checked that pile construction was carried out correctly, but was also an important extension of the design process.
The QA system was used to identify variations in ground conditions along the route and to propose changes in design details and remedial measures during the actual construction of the piles. This ensured that there was a high degree of confidence in the capacities of the as built piles, based on an economical design approach, and avoided unnecessary additional costs and disruptions to the construction programme due to potentially unacceptable foundations.
An important aspect of the QA system on such a large piling contract was the processing, interpretation and summarising of data of pile construction. This data was reviewed such that trends of critical parameters were highlighted and action taken before any implications became serious.
Creating customer confidence
MC Putnam, R Fernie and JA Elliott, Kvaerner Cementation Foundations, UK
Applying quality, marketing and service-based techniques to construction is often considered less appropriate than in manufacturing and service industries. Reasons given vary from the environment on a construction project, to the transient labour forces to the uniqueness of each project.
An even stronger argument may be raised when the product is out of sight below ground, as in foundations or ground treatment. The reality is, of course, that these are the very reasons for the adoption of standardised working practices and controls with continuous records for the duration of construction. Wherever possible the human control, though not supervision, of production standards should also be removed.
This then gives confidence in reliability and repeatability to enable undertakings to be made to clients on programme, price and quality. As clients grow towards taking this reliability for granted, then suppliers can move towards marketing themselves as service organisations able to influence project values through partnership, teamwork and employee empowerment.
The construction industry has needed to improve performance and become more sophisticated as a result of recession and its reputation for conflict and poor performance. This paper discusses how a major specialist sub- contractor, recognising the change, has developed its strengths and assets such that it can sit comfortably within this market place and offer itself as a service company.
This has been achieved by developing its quality system alongside its instrumented and automated products, and undergoing a process of cultural change. The company is now able to offer a flexible, customer oriented foundation service to clients who may now view commitments offered within this service with confidence. This confidence arises from a product confidence based on process and quality system capabilities, programme confidence from the application of proven key productivity indicators to tenders and site controls, and trust from relationships with a supplier who recognises that customer satisfaction is the cornerstone of future growth and profitability.
As the company develops with this approach so it moves confidently from having to resolve product deficiencies to offering service compliance.
Sludge detector for quality assurance of large diameter bored pile and barrette construction
John G Wang & CS Ho, Fugro International (Hong Kong) Limited
The ability to clear the sludge accumulated at the bottom of a bored pile, which is formed in a drilled hole stabilised with a Bentonite slurry, is critical to the load bearing capacity of the pile. If the sludge is too thick that it cannot be totally displaced during concreting, then it is difficult to carry out remedial work afterwards. The determination of the sludge thickness has often led to disputes between the designer and the contractor, since until now there has been no commonly accepted means to measure the thickness of the sludge quantitatively and effectively.
A slurry sludge detector has now been developed as a quality control device. It is portable and can be lowered into a drilled hole or a slurry trench to measure the depth and thickness of the sludge automatically. This paper describes the design principles and operation procedure of the device.
The working procedure of the test, which is carried out after flushing of the drilled hole, involves the following steps: The downhole detector is lowered into the bored hole or a slurry trench by the winch at a speed of 6m/min. For a 40m hole, lowering the detector takes seven minutes. The depth of the downhole detector below ground level is measured by a photo-electric sensor on the winch and recorded by computer.
The contact pressure acting on the detecting disk is measured by a strain gauge. When the pressure reaches 5KPa, a signal is sent from the downhole detector to the computer, which reduces the speed of the winch to 5mm/sec.
As the downhole detector penetrates into the sludge, the contact pressure acting on the detecting disk is recorded against the depth measured by the winch sensor. When the contact pressure reaches 25KPa, the internal locking mechanism attached to the detecting disk is automatically released, allowing the electric cones to continue to penetrate into the sludge while the detecting disk remains stationary. The displacement of the cones relative to the detecting disc measured by a LVDT, and the tip resistance, are recorded by the computer.
A test is terminated either when the cones register a tip resistance of 1.0MPa, or when the displacement exceeds 200mm. After a test is completed, the downhole detector is retrieved to the ground surface by the winch.
A prototype of the SSD has been completed and has been tested under a controlled condition. The early testing results have shown that it has satisfied all design objectives. Field tests will be carried out before it is put into production.
The bearing capacity of driven steel piles in weathered chalks
M Bustamante and L Gianeselli, Laboratoire Central des Ponts et Chausses, France; and L Weber, International Sheet Piling Company, Luxembourg
The mechanical properties of chalks differ strongly from conventional rocks such as granite, marl and sandstone. Indeed, the entry of a pile pulverises the chalk and breaks down its cellular structure. In the presence of groundwater, this powdered rock forms a slurry which lubricates the surfaces of the pile shaft. As a result, piles can be driven deeply into moderate-to-high weathered chalks below the water table without any marked increase in driving resistance. It is also stated by some authors that the shaft resistance in chalks increases substantially with time after driving; but the tests presented here do not support this.
Within this paper the results of loading tests, performed in France and Belgium over the last few years on five job sites, are analysed in order to deduce helpful design parameters of the unit shaft resistance and unit toe resistance of driven steel piles in weathered chalks.
Testing deep foundations with virtual instruments
Erez I Amir and Joram M Amir, GeoComp, Israel
The first generation of equipment for pile-testing (by stress wave analysis) consisted of purely analogue components. With its dbut in the early seventies, it consisted of makeshift equipment evolving to commercially available systems. The early eighties saw the transition to the second-generation, with digital systems based on purpose-built computers and proprietary operating systems. With the availability of more rugged laptop personal computers, third-generation pile-testing equipment based on DOS was a natural consequence. Today, practically all sonic testing systems in use are computerised, belonging to either the second or third generation.
The computer gap
In the course of a few years, a large gap has opened between the systems used by pile testing firms on site and those used in the office (word processors, worksheets, Internet browsers, etc). Office computers are constantly becoming more powerful, user-friendly, and, at the same time, more affordable.
As a result, existing pile-testing instruments look very outdated and, by contemporary standards, quite slow. Their user-interface is rather awkward, forcing the operator to remember a long list of acronyms. Moreover, updating such instruments to include modern computing power at a reasonable cost is practically impossible. Thus, the ability of these systems to perform serious, real-time analysis on-site, is severely limited.
Fourth generation pile testing equipment, which has recently become available, abolishes this dichotomy. It is based on the recognition that, on a world scale, the market for pile testing equipment is tiny. Therefore, to produce a cost-effective instrument, developers must dedicate their efforts to the job-specific part of the system.
By using standard operating system software and hardware components, the system developers can concentrate on the real heart of the system. Instead of designing and constructing displays, pointing devices, printer drivers, file storage, communication etc, these are readily picked from specialised manufacturers at attractive prices.
After more than two decades in which pile testing equipment has evolved, the industry is now ripe to switch over to fourth generation instrumentation.
A comparison of static and statnamic load tests in sand: A case study of the Bayou Chico bridge in Pensacola, Florida
Michael D Justason, Berminghammer Foundation Equipment, Canada;
Gray Mullins, University of South Florida; Donald T Robertson, Applied Foundation Testing; and William F Knight, Florida Department of Transportation, US.
This paper details the static and statnamic load testing on a 600mm square pre-stressed concrete pile. The pile was located at Pier 15 on the $20M Bayou Chico Bridge Project in Pensacola, Florida. The pile was 10.5m long and had a design load of 1.3MN.
Three static load test cycles were performed in November 1996. The statnamic load test was performed in January 1997. The Davisson Failure load of the first cycle of the static test was 3.7MN, and the Davisson Failure load of the Statnamic test was 3.2MN. The load displacement curves for the two types of tests were similar, indicating that the Statnamic load test performed well in the sandy soils.
Strain gauges embedded in the pile showed 35% skin friction and 65% end bearing for both test methods. Strain gauges were located at equal spacing at five elevations in the pile. A toe accelerometer verified the rigid body assumption of the unloading point method currently used in the analysis of statnamic load test. The data collected from the pile instrumentation was of excellent quality and represents some of the best research to date comparing static and statnamic load tests.