Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home   |   Geotechnical Home
Report
This report is an archived publication and may contain dated technical, contact, and link information

Publication Number: FHWA-HRT-05-159
Date: June 2006

Previous | Table of Contents | Next

Design and Construction of Driven Pile Foundations – Lessons Learned on The Central Artery/Tunnel Project

Chapter 3. Construction Equipment and Methods

This chapter presents a description of the equipment and methods used during pile driving operations at the CA/T project in the selected contracts. This includes a general overview of impact hammers, how a pile is installed, and how to tell when a pile has reached the desired capacity. Construction issues associated with pile driving during this project are also presented. Pile heave was identified as an issue during construction of the arrivals tunnel at Logan Airport, which required a significant number of piles to be redriven. At another site at the airport, soil heave resulting from pile driving caused significant movement of an adjacent building and required changes to the installation process, including preaugering the piles to a depth of 26 m.

EQUIPMENT AND METHODS

Impact hammers were used to drive all of the piles for the CA/T project. An impact hammer consists of a heavy ram weight that is raised mechanically or hydraulically to some height (termed "stroke") and dropped onto the head of the pile. During impact, the kinetic energy of the falling ram is transferred to the pile, causing the pile to penetrate the ground.

Many different pile driving hammers are commercially available, and the major distinction between hammers is how the ram is raised and how it impacts the pile. The size of the hammer is characterized by its maximum potential energy, referred to as the "rated energy." The rated energy can be expressed as the product of the hammer weight and the maximum stroke. However, the actual energy transferred to the pile is much less a result of energy losses within the driving system and pile. The average transferred energies range from 25 percent for a diesel hammer on a concrete pile to 50 percent for an air hammer on a steel pile.(17)

Three types of hammers were used on the selected contracts: (1) a single-acting diesel, (2) a double-acting diesel, and (3) a single-acting hydraulic. The manufacturers and characteristics of the hammers used in these contracts are summarized in table 4, along with the pile types driven. Schematics of the three types of hammers are shown in figures 9 through 11.

Table 4. Summary of pile driving equipment used on the selected contracts.
Make and Model Type Action Rated Energy (kN-m) Pile Types Driven Contracts Designation
Delmag™
D 46-32
Diesel Double
153.5
41-cm PPC
C07D1
I
HPSI 2000
Hydraulic Single
108.5
41-cm PPC
C07D1, C07D2
II
ICE 1070
Diesel Double
98.5
31-cm PPC, 41-cm PPC, 41-cm pipe
C08A1, C09A4
III
HPSI 1000
Hydraulic Single
67.8
41-cm PPC
C19B1
IV
Delmag D 19-42
Diesel Single
58.0
32-cm pipe
C19B1
V
Delmag D 30-32
Diesel Single
99.9
32-cm pipe
C19B1
VI

A single-acting diesel hammer (figure 9) works by initially raising the hammer with a cable and then releasing the ram. As the ram free-falls within the cylinder, fuel is injected into the combustion chamber beneath the ram and the fuel/air mixture becomes pressurized. Once the ram strikes the anvil at the bottom of the cylinder, the fuel/air mixture ignites, pushing the ram back to the top of the stroke. This process will continue as long as fuel is injected into the combustion chamber and the stroke is sufficient to ignite the fuel.

Figure 9. Schematic. Single-acting diesel hammer. This figure contains six sketches that together show how a single acting diesel hammer initiates and maintains pile driving. The first sketch, entitled Tripping, names the parts of the hammer. From top to bottom, the parts are: ram, cylinder, exhaust port, anvil, recoil dampener, striker plate, hammer cushion, and helmet. An additional part, the fuel pump, is identified in the second sketch, entitled Fuel Injection. In the first sketch, Tripping, the ram is in a raised position. In the second sketch, Fuel Injection, the ram is descending. In the third sketch, Compression-Impact, the tip of the ram has reached the anvil. In the fourth sketch, Explosion, an explosion has occurred where the tip of the ram met the anvil, and the ram is ascending. In the fifth sketch, Exhaust, the ram has passed the exhaust port from which exhaust is escaping. And in the sixth sketch, Scavenging, the ram has reached its highest position and is starting to descend.
Figure 9. Single-acting diesel hammer.(17)

A double-acting diesel hammer (figure 10) works like the single-acting diesel hammer except that the system is closed at the top of the ram. As the ram rebounds to the top of the stroke, gasses are compressed in the bounce chamber at the top of the hammer. The bounce chamber temporarily stores and redirects energy to the top of the ram, allowing the stroke height to be reduced and the blow rate to be increased. Bounce chamber pressure is monitored during pile driving because it is correlated with hammer energy. The stroke of the hammer, and thus the energy, is controlled using the fuel pump. This is effective for avoiding bouncing of the hammer during the upstroke, which can lead to unstable driving conditions and damage to the hammer.(17)

A single-acting hydraulic hammer (figure 11) uses a hydraulic actuator and pump to retract the ram to the top of the stroke. Once the ram is at the top of the stroke, the ram is released and free-falls under gravity, striking the anvil. An advantage of hydraulic hammers is that the free-fall height, and thus the energy delivered to the pile, can be controlled more accurately.

Figure 10. Schematic. Double-acting diesel hammer. This figure contains five sketches. The first sketch, entitled Tripping, names the parts of the hammer. From top to bottom, the parts are: bounce chamber, bounce chamber port, ram, cylinder, inlet or exhaust scavenge port, anvil, recoil dampener, striker plate, hammer cushion, and helmet. In the first sketch, Tripping, the ram is descending from a raised position. In the second sketch, Compression, the tip of the ram is near the anvil. In the third sketch, Compression-Impact, the tip of the ram has reached the anvil. In the fourth sketch, Exhaust, the ram is ascending and exhaust is venting from the inlet or exhaust scavenge ports. In the fifth sketch, Scavenging, the ram is near the top of its ascent.

Figure 11. Schematic. Single-acting hydraulic hammer. This figure contains one sketch in which the parts of a single-acting hydraulic hammer are identified. From top to bottom, the parts are: actuator, adjustable cylinder carriage, cage, lift cylinder, segmented ram weight, shock absorber, anvil, and pile sleeve.

Figure 10. Double-acting diesel hammer.(17)

Figure 11. Single-acting hydraulic hammer.(17)

In preparation for driving, a pile is first hoisted to an upright position using the crane and is placed into the leads of the pile driver. The leads are braces that help position the piles in place and maintain alignment of the hammer-pile system so that a concentric blow is delivered to the pile for each impact. Once the pile is positioned at the desired location, the hammer is lowered onto the pile butt. A pile cushion consisting of wood, metal, or composite material is placed between the pile and the hammer prior to driving to reduce stresses within the pile during driving.

Once the pile is in position, pile driving is initiated and the number of hammer blows per 0.3 m of penetration is recorded. Toward the end of driving, blows are recorded for every 2.5 cm of penetration. Pile driving is terminated when a set of driving criteria is met. Pile driving criteria are generally based on the following: (1) the minimum required embedment depth, (2) the minimum number of blows required to achieve capacity, and (3) the maximum number of blows to avoid damage to the pile. All information that is associated with pile driving activities (e.g., hammer types, pile types, pile lengths, blow counts, etc.) is recorded on a pile driving log.

A typical pile driving log is shown in figure 12. This particular record is for the installation of a 24-m-long, 41-cm-diameter PPC pile installed at the airport as part of contract C07D2. A hydraulic hammer with an 89-kN ram and a 1.2-m stroke was used. The number of blows per 0.3 m of driving was recorded from an embedment depth of 9.5 m to a final depth of 16.5 m. At a depth of 16.5 m, the hammer blows required to drive the pile 2.5 cm were recorded in the right-hand column of the record. Driving was stopped after a final blow count of 39 blows per 2.5 cm was recorded.

Once a pile has been installed, the hammer may be used to drive the pile again at a later time. Additional driving that is performed after initial installation is referred to as a redrive or restrike. A redrive may be necessary for two reasons: (1) to evaluate the long-term capacity of the pile (i.e., pile setup or pile relaxation), or (2) to reestablish elevations and capacity in piles that have been subject to heave. Both of these issues were significant for the CA/T project, and they are discussed in the next section.

Figure 12. Form. Typical pile driving record. This figure is an example of a completed pile driving record. The title is “Precast, Prestressed Concrete Pile Driving Record.” The top portion of the form contains entries for such identifying information as the contractor, the pile contractor, and the contract number. The top portion also contains sections for hammer data, pile data, and driving data. The middle portion of the form is a table with columns for depth in feet and blows per foot. The depth columns are pre-printed. The first row of the depth column is zero to 0.305 meters parenthesis zero to 1 foot end parenthesis, and the rows proceed in increments of 0.305 meters parenthesis 1 foot end parenthesis to 53.4 meters parenthesis 175 feet end parenthesis. The rows of the blows per foot column are not pre-printed. On this particular form, the blows per foot rows, beginning with the row for which the depth is 9.5 meters parenthesis 31 feet end parenthesis and ending with the row for which the depth is 16.5 meters parenthesis 54 feet end parenthesis, have handwritten entries. On the far right of the table is a section entitled Final Resist, with columns for depth and blows per inch. The depth column is blank and the blows per inch column has handwritten entries. The bottom portion of the form has spaces for remarks and for a signature by a contractor representative.
Figure 12. Typical pile driving record.

CONSTRUCTION-RELATED ISSUES

Pile Heave

Pile heave is a phenomenon where displacement of soil from pile penetration causes vertical or horizontal movement in nearby, previously driven piles. Pile heave generally occurs in insensitive clays that behave as incompressible materials during pile driving.(17) In these soils, the elevation of adjacent piles is often continuously monitored during driving to look for heave. If a pile moves in excess of some predetermined criterion, the pile is redriven to redevelop the required penetration and capacity. From a cost perspective, pile heave is important because redriving piles can require significant additional time and effort.

Pile Layout and Soil Conditions

Of the contracts reviewed, pile heave was an issue during construction of the arrivals tunnel at Logan Airport (contract C07D2). The location of the C07D2 site is shown in figure 1. A plan view of the arrivals tunnel structure showing the pile locations is shown in figure 13. The tunnel structure is approximately 159 m in length and is located where ramp 1A-A splits from the arrivals road. The tunnel was constructed using the cut-and-cover method, and thus a portion of the overburden soil was excavated prior to pile driving.

Figure 13. Drawing. Site plan, piling layout for the arrivals tunnel at Logan Airport. This figure is an irregularly shaped drawing. On the right side is an open area labeled “Arrivals Road.” Proceeding to the left, another blank area branches at a diagonal upwards and is labeled “Ramp 1A.” Surrounding the blank areas are grid-like borders that indicate the locations of approximately 576 piles described in the text. At the bottom of the drawing is the legend “Not to Scale.”
Figure 13. Site plan, piling layout for the arrivals tunnel at Logan Airport.(18)

Approximately 576 piles were driven beneath the alignment of the tunnel structure. The piles, consisting of 41-cm-diameter PPC piles, were designed to support a concrete mat foundation in addition to a viaduct located above the tunnel. They were generally installed in a grid-like pattern, with a spacing of approximately 1.2 m by 1.8 m center to center (figure 13).

The general subsurface conditions based on borings advanced in the area prior to excavation consist of approximately 3 to 6.1 m of cohesive and/or granular fill, overlying 1.5 to 3 m of organic silt and sand, overlying 12.2 to 42.7 m of soft marine clay, overlying 0.9 to 2.8 m of glacial silts and sands, underlain by bedrock.(6) Excavation was accomplished into the clay layer, resulting in a clay layer thickness of about 6.1 m at the southeastern end of the structure to around 3.7 m at the northwestern end.(19)

The piles were designed for end bearing in the dense glacial silts and sands, and were preaugered to about the bottom of the marine clay layer to minimize heave and displacement of these soils. The preauger depths were approximately 30 to 70 percent of the final embedment depths of the piles. Preaugering was done using a 46-cm-diameter auger, which is the equivalent circular diameter of the 41-cm square pile. The piles were driven using an HPSI 2000 hydraulic hammer.

Field Observations

Pile heave was monitored during construction by field engineers. As described in the Massachusetts State building code and project specifications, piles identified with vertical displacement exceeding 1.3 cm required redriving. According to field records, 391 of the 576 piles (68 percent) installed required redriving. Of those 391 piles, 337 piles (86 percent) were driven in one redrive event, 53 piles (14 percent) required a second redrive event, and 1 pile required a third redrive event. The impact on the construction schedule or costs was not identified. Despite the use of partial preaugering, a significant portion of the piles showed excessive heave and required substantial redrive efforts. Heave is attributed to the displacement of the underlying glacial soils that were not preaugered.

Pile heave issues were not identified on the other CA/T contracts. Since partial preaugering was used on the majority of these contracts, the difference may be related to the spacing between piles. Table 5 summarizes the pile spacing used on the selected contracts. As shown in table 5, the pile spacing of 1.2 m used at the arrivals tunnel structure is significantly less than the spacing used for structures of comparable size. Therefore, it is anticipated that a pile spacing of greater than about 1.8 m may limit pile heave to within the 1.3-cm criterion.

Table 5. Summary of pile spacing from selected contracts.
Contract Structure Foundation Bent Spacing(m) Pile Spacing (m)
C07D1
Ramp ET
Slab
2.7
2.7
Pile cap
1.4
1.4
Egress Ramps
Pile cap
1.8
1.8
C07D2
Arrivals Tunnel
Pile cap
1.8
1.2
Pile cap
1.8
1.2
Pile cap
1.4
1.2
C08A1
South Abutment
Pile cap
3.05
1.8-2.4
East Abutment
Pile cap
1.1–2.7
1.4–2.6
West Abutment
Pile cap
1.1–2.1
1.4–2.7
C09A4
Utilities
Pile cap
2.0–2.7
1.8
Approach No. 1
Slab
3.7
5.6
Pile cap
1.4
2.6
Pile cap
NA
1.4
Pile cap
NA
1.5
Approach No. 2
Slab
4.57
3.1–4.6
Approach No. 5
Slab
3.7–4.9
2.1–4.3
C19B1
NS-SN
Slab
3.7
4.9
Ramp CT
Slab
3.1
4.6
Ramp LT
Slab
2.9–3.2
2.4–3.1

NA = not applicable or available

Soil Heave

Soil heave caused by pile driving was primarily responsible for the significant movement observed at a building adjacent to the construction of the east abutment and east approach to ramp ET at Logan Airport (contract C07D1). Shortly after the start of pile driving, settlement in excess of 2.5 cm was measured at the perimeter of the building and cracking was observed on the structure itself. These observations prompted the installation of additional geotechnical instrumentation, installation of wick drains to dissipate excess pore pressure generated during pile driving, and preaugering of the piles to reduce soil displacement. Despite these efforts, heave continued to a maximum vertical displacement of 8.8 cm. (See references 20, 21, 22, and 23.)

Pile Layout and Soil Conditions

The location of the project in relation to the building is shown in figure 14. The portion of the east approach that is adjacent to the building consists of two major structures, including an abutment and a pile-supported slab. Both structures are supported by 41-cm-diameter PPC piles. The layout of the pile foundation system is also shown in figure 14. The piles for the slab are arranged in a grid-like pattern with a spacing of about 2.7 m center to center. A total of 353 piles support the structures.

Figure 14. Drawing. Site plan showing locations of piles, building footprint, and geotechnical instrumentation. At the top of this drawing, extending generally horizontally, is the irregular outline of a structure labeled 'Existing Building.' Just below the building are five deformation monitoring points. The sites of an inclinometer and a multipoint heave gauge are also indicated, as are the sites of three vibrating wire piezometers. Below the building is an irregularly shaped area divided into Phase I and Phase II. Within the area, the locations of 353 piles are indicated. The locations of wick drains above and to the left of the pile area are also indicated.
Figure 14. Site plan showing locations of piles, building footprint, and geotechnical instrumentation.

Prior to construction activities, five deformation monitoring points (DMPs) were installed along the front perimeter of the building closest to the work area. The DMPs consisted of 13-cm-long hex bolts fixed to the building. These points, designated DMP-101 through DMP-105, were monitored for vertical movement. The DMPs were monitored initially by the contractor and subsequently monitored by an independent consultant.

The subsurface conditions based on borings advanced in the area consist of approximately 3 to 4.6 m of fill, overlying 3 to 6.1 m of organic silt and sand, overlying 27.4 to 33.5 m of soft marine clay, overlying 6.1 to 12.2 m of glacial silt and sand, underlain by bedrock. The piles were designed as end bearing piles to be driven into the dense underlying glacial materials. The glacial soils were encountered at depths of approximately 39.6 to 45.7 m below the ground surface and bedrock was encountered at a depth of approximately 48.8 m.

Field Observations (Phase I Pile Driving)

Pile driving for the east approach was executed in two phases. The first phase began on April 5, 1995, and concluded on June 10, 1995. The second phase began on July 13, 1995, and concluded on August 17, 1995. The piles were driven using a Delmag D46-32 single-acting diesel hammer. The extent of the first phase of pile driving is shown in figure 15. This first phase of work was performed no closer than 27.4 m from the building. The majority of the piles for the slab were installed from the west side of the site working toward the east during the periods of April 5 to April 23, and May 15 to June 2. The majority of the piles for the abutment were installed at the west end of the site during the period of April 23 to May 15.

Settlement data obtained by the contractor during the first phase of pile driving are shown in figure 15. On April 21, 1995, after approximately 2 weeks of pile driving on the west side of the site, initial heave displacements of 0.9 and 0.7 cm were measured in DMP-102 and DMP-103, respectively. Notable heave was observed at DMP-101 and DMP-104 on May 1, which registered displacements of 1.3 and 0.8 cm, respectively. An initial heave displacement of 0.4 cm was measured in DMP-105 on May 9. The heave increased steadily to maximum values as pile driving commenced toward the east side of the site.

Figure 15. Graph. Settlement data obtained during first phase of pile driving. This figure is a graph. The x axis is the date and ranges from January 25, 1995, to June 28, 1995. The y axis is the vertical heave in centimeters and ranges from minus 2 to plus 6. A double-headed arrow on the graph indicates that Phase I extended from April 5, 1995, to June 10, 1995. Five lines connecting five sets of data points are plotted on the graph, one each for deformation monitoring points 101 through 105. From January 1995 to the beginning of Phase I on April 5, 1995, the plots are, for the most part, horizontal, with a vertical heave between minus 1 and zero centimeters. The five plots slope upward to the right during Phase I. The plot for deformation monitoring point 103 has the highest apogee, a vertical heave of approximately 4.4 centimeters near the end of Phase I in early June 1995. The plot for deformation monitoring point 105 has the lowest apogee, a vertical heave of approximately 1.6 centimeters near the end of Phase I. After the end of Phase I, each of the plots begins sloping downward.
Figure 15. Settlement data obtained during first phase of pile driving.

A summary of the maximum heave values attributed to the first phase of driving is given in table 6. The greatest amount of heave occurred in DMP-103, which was centrally located relative to the pile grid. On June 2, 1995, 1 week before completion of construction, the heave measured in DMPs 101 through 103 began to level off and subside.

Table 6. Maximum building heave (in cm) observed during pile driving.
Construction Phase DMP 101 DMP 102 DMP 103 DMP 104 DMP 105
Phase I
2.5
3.5
4.3
3.8
1.6
Phase II
3.6
4.8
5.3
3.7
1.3

As a result of the excessive heave (greater than 2.5 cm) observed in the first phase of pile driving, mitigation measures were implemented for the second phase of work. This was critical considering that the second phase involved driving piles even closer to the building. The geotechnical consultant recommended three approaches to limiting heave based on schedule and cost constraints.(24) These included: (1) installation and monitoring of pore pressures in the clay during driving and adjusting mitigating measures as appropriate; (2) installation of wick drains between the Hilton and the work area to intercept and aid in the reduction of pore pressures beneath the Hilton that may be generated from pile driving; and (3) based on the performance of the wick drains, preauger phase II piles to limit soil displacement.

Field Observations (Phase II Pile Driving)

Prior to the start of the second phase of pile driving, three double-nested vibrating wire piezometers (VWPZ) were installed to measure pore pressures. These piezometers were installed in close proximity to three of the existing deformation monitoring points (DMP-102 through DMP-104). Additional instrumentation was also installed following the start of the second phase of work, including a multipoint heave gauge (MPHG) to measure vertical movement with depth and an inclinometer to measure lateral movement. The locations of the additional geotechnical instrumentation are shown in figure 14.

The second phase of pile driving began on July 13, 1995, and concluded on August 17, 1995. The extent of the work area is also shown in figure 14. Pile driving generally progressed from the west side of the site toward the east. The location of the second phase of work was no closer than 15.2 m from the existing building.

Shortly after the start of driving, 200 wick drains were installed from July 20 to July 28, 1995, around the western and northern perimeters of the work area. The drains were installed through the clay layer at a spacing of 1.2 m center to center.

Settlement data for the second phase of work, shown in figure 16, demonstrate that heave began to increase at DMP-101 through DMP-104 approximately 1 week after the start of pile driving. Based on the review of initial settlement data, preaugering was implemented from August 4, 1995, through the completion of construction. Preaugering was accomplished using a 41-cm-diameter auger to a depth of 26 m, which is approximately 50 to 60 percent of the pile's final embedment depth. The auger diameter is 11 percent less than the 46-cm equivalent circular diameter for a 41-cm square pile.

As shown in figure 16, heave continued to increase even after preaugering was initiated. Net heave values of 3.3 to 13.5 cm (table 6) were observed from the start of preaugering to the completion of pile driving, resulting in total heave values ranging from 2.6 to 8.8 cm.

Figure 16. Graph. Settlement data obtained during second phase of pile driving. This figure is a continuation of the graph in figure 14. The x axis is the date and ranges from April 5, 1995, to August 23, 1995. The y axis is the vertical heave in centimeters and ranges from zero to 10. One double-headed arrow on the graph indicates that Phase I extended from April 5, 1995, to June 10, 1995. A second double-headed arrow indicates that Phase II extended from July 13, 1995, to August 17, 1995. Five lines connecting five sets of data points are plotted on the graph, one each for deformation monitoring points 101 through 105. The plots begin at approximately May 17, 1995, or in the latter part of Phase I. The first data point for each plot falls within the vertical heave range of 0.8 to 3.3 centimeters. The plots then slope upward until the end of Phase I on June 10, 1995. The plots then slope gradually downward until the beginning of Phase II on July 13, 1995, falling to a vertical heave range of 1.2 to 3.5 centimeters. During Phase II, the plots slope sharply upward, reaching apogees near the end of Phase II on August 17, 1995. The plot for deformation monitoring point 103 has the highest apogee, a vertical heave of approximately 8.6 centimeters. The plot for deformation monitoring point 105 has the lowest apogee, a vertical heave of approximately 2.6 centimeters. A small double-headed arrow indicates that wick drains were installed between approximately July 20 and July 28, 1995. A single-headed arrow indicates that preaugering began on approximately July 17, 1995.
Figure 16. Settlement data obtained during second phase of pile driving.

Data from the multipoint heave gauge showed that the magnitude of the heave was relatively constant within the upper 30 m, as shown in figure 17. However, vertical displacement decreases dramatically below this depth to approximately zero at the bedrock depth of approximately 50 m. The maximum heave of approximately 5.1 cm at a depth of 3 m below the ground surface is also consistent with the maximum value of 5.3 cm recorded at DMP-103.

Figure 17. Graph. Multipoint heave gauge data obtained during second phase of pile driving. This figure consists of a graph of data points and connecting lines, and an adjacent bar chart. The x axis of the graph is vertical heave in centimeters and ranges from minus 1 to plus 7. The y axis is initial depth in meters and descends from zero to 60. Three lines connecting three sets of data points are plotted on the graph, one plot each for data collected on August 8, 11, and 18, 1995. The three plots begin at a vertical heave of approximately zero centimeters at an initial depth of approximately 50 meters. The three plots then rise roughly linearly to an initial depth of 30 meters. At that point, the vertical heave is approximately 1 centimeter for the August 8 plot, 2.5 centimeters for the August 11 plot, and 5.5 centimeters for the August 18 plot. From the initial depth of 30 meters to just below the surface, each plot rises in a roughly vertical fashion. The bar chart adjacent to the graph gives the depths of five layers of soil. The layers and depths are: fill, zero to 6 meters; silt and sand, 6 to 10 meters; marine clay, 10 to 44 meters; glacial soils, 44 to 50 meters; and bedrock, 50 to 60 meters.
Figure 17. Multipoint heave gauge data obtained during second phaseof pile driving.

The excess pore pressures recorded during the second phase of pile driving are presented in figure 17. The six gauges shown in figure 18 correspond to three pairs (55894–55895, 55896–55897, and 55898–55899) located adjacent to DMP-102, DMP-103, and DMP-104, respectively. There was an increase in the excess pore pressures throughout the pile driving, with maximum values ranging from 0.6 to 12.8 m of head with an average of 5.9 m. The greatest head was measured in VWPZ-55896 at a location nearest DMP-103. These data suggest that the wick drains were not effective in dissipating all excess pore pressures generated during pile driving.

Figure 18. Graph. Pore pressure data obtained during second phase of pile driving. The x axis of this graph is the date and ranges from June 28, 1995, to August 23, 1995. The y axis is excess pore pressure head in meters and ranges from minus 2 to plus 16. Six lines connecting six sets of data points are plotted on the graph, one each for gauges 55894 through 55899. The plots begin at an excess pore pressure head of zero meters on July 11, 1995. The plots then diverge, but generally increase until the final data points on August 17, 1995. On that date, the final data points range from a low excess pore pressure head of approximately 0.6 meters for gauge 55899 to a high excess pore pressure head of approximately 12.8 meters for gauge 55897.
Figure 18. Pore pressure data obtained during second phase of pile driving.

The inclinometer data that were obtained adjacent to the building are shown in figure 19. These data showed increasing lateral movement in the direction of the building during pile driving. The maximum net lateral deformations were relatively constant with depth within the upper 30 m of the profile. A maximum deformation of approximately 6 cm was recorded at a depth of approximately 34 m. Similar to the vertical deformations, the lateral deformations decreased sharply below this depth to zero at the bedrock depth. These data suggest that the lateral deformations are of the same magnitude and behavior as the vertical deformations.

Figure 19. Graph. Inclinometer data obtained during second phase of pile driving. This figure consists of a graph of data points and connecting lines, and an adjacent bar chart. The x axis of the graph is lateral deformation in centimeters and ranges from zero to 7. The y axis is depth in meters and descends from zero to 60. Three lines are plotted on the graph, one plot each for data collected on August 12, 16, and 18, 1995. The three plots begin at a lateral deformation of approximately zero centimeters at a depth of approximately 50 meters. The three plots then curve to the right and upward, each reaching a maximum lateral deformation at a depth of approximately 34 meters. The maximum deformations are: approximately 3.2 centimeters for the August 12 plot, approximately 4.6 centimeters for the August 16 plot, and approximately 6.0 centimeters for the August 18 plot. From a depth of approximately 34 meters to the surface, each plot curves in an irregular fashion to the left. The lateral deformations at a depth of zero meters are: approximately 2.2 centimeters for the August 12 plot, approximately 2.5 centimeters for the August 16 plot, and approximately 3.2 centimeters for the August 18 plot. The bar chart adjacent to the graph gives the depths of five layers of soil. The layers and depths are: fill, zero to 6 meters; silt and sand, 6 to 10 meters; marine clay, 10 to 44 meters; glacial soils, 44 to 50 meters; and bedrock, 50 to 60 meters.bar chart: fill, silt and sand, marine clay, glacial soils, bedrock
Figure 19. Inclinometer data obtained during second phase of pile driving.

SUMMARY

Soil heave was recognized early on as a potential problem and following phase I driving in contract C07D1, several mitigation efforts were initiated. These included installing wick drains to promote rapid dissipation of excess pore pressures and preaugering piles through a portion of the soft clay layer to a depth of 26 m. Additional instrumentation was installed, including piezometers, MPHG, and an inclinometer. Despite these efforts, heave during phase II pile driving continued to increase to a maximum displacement of 8.8 cm. The piezometer data indicate that the wick drains were not effective in rapidly dissipating pore pressures generated during pile driving. The deformation data indicated that soil heave can still occur in piles that are preaugered over a portion of their embedded depth.

 

Previous | Table of Contents | Next


The Federal Highway Administration (FHWA) is a part of the U.S. Department of Transportation and is headquartered in Washington, D.C., with field offices across the United States. is a major agency of the U.S. Department of Transportation (DOT).
The Federal Highway Administration (FHWA) is a part of the U.S. Department of Transportation and is headquartered in Washington, D.C., with field offices across the United States. is a major agency of the U.S. Department of Transportation (DOT). Geotechnical Engineering Federal Highway Administration's (FHWA) R&T Web site portal, which provides access to or information about the Agency’s R&T program, projects, partnerships, publications, and results.
FHWA
United States Department of Transportation - Federal Highway Administration