The $280 million C710 Beacon Hill Tunnels and Station contract forms part of the 14-mile (22km) long initial segment of Sound Transit’s Central Link Light Rail Line, which extends between downtown Seattle and SeaTac Airport. The 4,300ft (1.3km) long twin running tunnels under Beacon Hill were recently completed using a 21ft (6.4m) diameter Mitsubishi EPBM, while the deep mined station has been built using slurry walls for the shafts and the Sequential Excavation Method (SEM) for the tunnels. With an excavated depth of 160ft (49m), the station is the deepest constructed in soft ground in North America.
Beacon Hill Station (Figure 1) consists of twin shafts and a complex configuration of vehicle, pedestrian and ventilation tunnels. From the Station Headhouse, a 181ft deep (55m), 46ft (14m) i.d. Main Shaft was constructed that will house four high-speed elevators, emergency staircases, ventilation shafts and M&E equipment. A 161ft (49m) deep, 26ft (8m) i.d. Ancillary Shaft will accommodate another set of emergency staircases and ventilation shafts.
From the Main Shaft, the 41ft (12m) wide Concourse Cross Adit (CCA) will provide passenger and emergency access to the Northbound and Southbound Platform Tunnels (NBPT and SBPT respectively). These are 380ft (116m) long by 32ft (9.8m) wide and were designed to accommodate the light rail tracks, platforms, artwork and architectural finishes. Two pedestrian Cross Adits connect the Platform Tunnels, and Ventilation Tunnels will provide airflow in normal operation and for emergencies, these range in size from 16ft (4.9m) to 26ft (8m) in diameter.
A JV of Hatch Mott MacDonald and Jacobs Civil (HMMJ) completed the final design of the Beacon Hill segment of the project between 2002 and 2004. Dr G Sauer Corporation (DSC) was awarded a subcontract by HMMJ for the design of the station platform and concourse tunnels, monitoring and waterproofing. In addition to design support during construction, both HMMJ and DSC have provided SEM resident engineering services throughout. Puget Sound Transit Consultants (PSTC), a JV of Parsons Brinckerhoff Quade & Douglas, Earth Tech and URS Corporation, performed the preliminary engineering and provided program management services as an integrated team with the client, Sound Transit. Parsons Brinckerhoff Construction Services is providing construction management as Resident Engineer for the contract.
Final design of the C710 contract was completed in February 2004 and bids were solicited following a pre-qualification process in which prospective contractors had the opportunity to examine a 16ft (4.9m) diameter Test Shaft[1], constructed within the footprint of the future Main Shaft.
Three teams of bidders pre-qualified: Obayashi, a JV led by Kiewit, and another JV led by Impregilo. However, Impregilo’s team ultimately withdrew its bid. On May 15, 2005, the bids were opened, with Obayashi at $280 million and the Kiewit JV at $305 million. The Engineers estimate was $240 million. On June 28, Sound Transit issued Notice-To-Proceed to Obayashi.
Obayashi subsequently entered into an agreement with Beton und Monierbau to provide key SEM staff.
Geology
The Beacon Hill geology consists of glacial and interglacial soil units with a high degree of variation. This is evident locally to the extent that some units cannot be reliably correlated between adjacent borings. At the outset of the project, an extensive subsurface exploration program, with over 70 bore holes drilled, showed that most of the Beacon Hill Station would be excavated within glacial, over consolidated, partly fractured or slickensided clays and tills.
Intermittent sand and silt layers were also found present, with multiple perched groundwater horizons. In order to simplify the descriptions of the large number of geological units identified, and to reduce the complexity of the geological profile, the soils were grouped into six classes:
1) Loose to dense granular deposits
2) Soft to very stiff clay and silt
3) Till and till like deposits
4) Very dense sand and gravel
5) Very dense silt and fine sand
6) Very stiff to hard clay
The complex layering of these units, particularly thick zones of sands and silts, was extremely challenging for the design of shafts and tunnels[2].
Ground treatment
For the reasons described above, the designers specified a post contract award subsurface exploration program to refine the geotechnical interpretations. Boreholes, drilled down to platform tunnel level using mud rotary and sonic core recovery methods, were combined with instrumentation including inclinometers, extensometers and piezometers.
The variable ground conditions, with the potential for saturated and unstable sand lenses, led to the consideration of large-scale ground improvement measures prior to sequential excavation. A series of deep dewatering wells and vertical jet grouting program was therefore specified[3].
Depending on the soil properties encountered during exploration and instrument drilling, the need for additional jet grouting or dewatering measures would be determined.
Between August 2004 and November 2005, 50 bore holes were drilled, with core extraction and sampling conducted under the engineering oversight of Shannon & Wilson[4].
It became evident during these investigations that sand layers, subject to water pressures of approx. 50ft (15m), within the Station tunnels were far more extensive than previously anticipated. This would have meant a 300% increase in the quantity of jet grouting specified in the contract[5]. Not only would the additional jet grouting cause significant delay to the construction progress and have major cost implications, there would also be a significant impact on the local community, as a large portion of the work would have to occur outside of the worksite, within street right-of-way and on private property.
Ultimately, Sound Transit and Obayashi developed the concept of a major “shift” in the configuration of the station components, including reversal of the Platform Tunnels lengths on either side of the Main Shaft and repositioning the East Damper Chamber. The final volume of jet grouting required as a result of this “Ultimate Shift” was only 10% greater than the original bid quantity.
The jet grouting was carried out by a JV of Condon-Johnson/Soletanche and provided a very competent soil stabilisation mass, with virtually no water influx encountered.
Primary lining & SEM pre-support
As discussed, the tunnel sizes required for the station set a new record for SEM tunneling in the US. The stability assessment for the excavation sequences and support measures was performed with 3-D Finite Element (FE) models using the programs ABAQUS and FLAC.
The lining forces derived from the numerical analysis were subsequently used to design the primary and final lining[6] using capacity limit curves based on ACI 318. Due to the mixed face conditions present over large portions of the tunnel alignment, no ‘prescriptive’ support was designed, but a baseline scenario (Figure 2) was chosen that included excavation sequence and standard support measures, such as shotcrete, welded wire fabric (WWF) and lattice girders.
The contractor later paid for a redesign and pre-construction performance flexural beam tests in order to substitute steel fibre reinforced wet-mix shotcrete for the double mats of wire mesh. This was primarily improve productivity and reduce worker risk at the tunnel face, as well as improve the overall quality of shotcrete application.
Additional SEM “Toolbox” items[7] were also included in the contract (supplemental unit price bid items) to be used in conjunction with conventional excavation support. These included: Pre-support measures such as rebar spiling, grouted pipe spiling, metal sheets and grouted barrel vaults (pipe arch); Face stabilisation measures including the use of a face stabilisation wedge, pocket excavation, reduction of round length, face bolts and additional shotcrete; Ground improvement measures such as gravity and vacuum detwatering, permeation grouting, fracture grouting and jet grouting; and annular support measures, including additional shotcrete, soil nails and use of a temporary invert.
The contract required a certain amount of shotcrete nozzle time for nozzlemen, which proved difficult to find in a tight labor market. In order to assist in boosting human resources available for SEM crews, a variance was made to allow for on-the-job training of nozzlemen, albeit under the direction of the certified nozzlemen and foremen already on site.
Shaft construction
Slurry wall construction of the Main and Ancillary Shafts, and the East and West headhouses, began in October 2004 and was completed in February by Obayashi’s subcontracted JV partners Condon-Johnson/ Soletanche. Main shaft excavation, including West Headhouse tiebacks and demolition of the remnants of the original Test Shaft, began in March 2005 and by the end of May 2005 the shaft had been sunk to a depth of 120ft (36.5m), ready for the West Longitudinal Vent Adit (WLVA) breakout.
WLVA breakout through the Main Shaft slurry wall began in June 2005 and SEM work, albeit for a 12ft (3.7m) stub tunnel, began in earnest. Originally sequenced for excavation in nine stages for the three lattice girder rounds, the work was adjusted in the field to five rounds, by combining bench and invert rounds. Only a few dozen rebar spiles were used in the top heading second round to bridge the gap between the remnants of the jet grout columns and the roof over the headwall.
The on-site wet mix batch plant was still being assembled and fine-tuned while having trial batches run through and brought up to strength, so the initial lining for this breakout section was sprayed in with a 10” (250mm) thick dry-mix steel fiber flashcrete mix. A total of 33 shifts were needed to to complete the breakout, including securing the headwall for the better part of a year and the installation of deformation monitoring instrument pins.
Concourse construction
Excavation work in the Main Shaft continued after completion of the WLVA breakout in July 2005. The shaft was lowered down to just below the crown level of CCA for installation of the upper row of barrel vault pipes.
Northwest Cascade was subcontracted to drill 134 lost-casing pipes, evenly split with 67 each over the North and South Concourse brow of each heading. However, drilling over the North Concourse encountered more sandy soils and drew in groundwater along several of the first round of holes, which led to the lengthening of 12 holes along the inner row. Drilling over the South Concourse was relatively dry and clayey prompting deletion of the outer row entirely. Barrel vault drilling, surveying and double-packer staged grouting with microfine cement was completed by mid-August 2005, setting the stage for further shaft sinking to the CCA springline.
A Cat 325 and a Cat 320 excavator, both with hoe rams, were used to break through the 4ft thick slurry walls for the NCCA and SCCA top headings. A series of line-drilled holes were drilled with a twin-boom drill jumbo to outline the breakout area and to help weaken the wall. Initially designed for localized breakout of each individual top heading, Obayashi broke through the wall in five stages, including a portion of the center drift top heading, leaving a concrete pedestal for interim face support.
Initial excavation of the first Concourse Cross Adit (CCA) top headings began in late August 2005 for both the North and South Concourses. The permanent lattice girders along the side drifts and the temporary TH-girder sections in the center drift were installed. A single lattice girder set from the center crown was also installed, connecting the East and West sides of each heading. This was used to help align the temporary interior sidewalls as the initial top heading excavation began. The South CCA East side drift was the first top heading excavated starting in September, followed by the South CCA West side drift. The North CCA East side, and then West side, followed.
After completion of all four top headings in December 2005 the Main Shaft was excavated down to the final invert elevation of 181ft (55m) below surface. A 30” (750mm) thick concrete invert slab was poured, then backfilled up to a working level.
The South CCA sidewall drift benches and inverts were then excavated before again filling the shaft temporarily to complete the centre drift top heading and upper bench of the South CCA.
The South CCA center bench and invert were excavated in February 2006, including demolition of the temporary sidewalls.
Platform tunnels
The Southbound Platform Tunnel (SBPT) was on the contract’s critical path, with its completion necessary before the 21ft (6.4m) diameter Mitsubishi EPBM driving the running tunnels could break-in at the west end of the platform. The TBM would then be “walked” through the station and re-launched on the remainder of the Southbound running tunnel bore, before being removed and reassembled for the Northbound bore[8]. Work to complete the North Concourse and subsequent North Platform Tunnel (NBPT) excavation followed start-up on the SBPT concurrently.
A single sidewall drift method was specified for the Platform Tunnels, comprising a total of six segments per round (Figure 2). The first drift was excavated as a pilot tunnel, with the second drift following behind the first with a minimum specified offset. This offset rule allowed time to complete ring closure and gave the shotcrete time to develop strength in the first drift.
Obayashi investigated several ways to try and improve cycle times, in the interests of the schedule, and several minor changes were implemented. The single most effective of these was to increase the round length when possible, taking into account the results of probings, face maps and feedback from instrumentation in the shell.
In the design given to Obayashi, the lattice girder served only as a template and was not taken into account in the structural analysis. Therefore, the span length could be increased if the ground permitted. In the end, Obayashi adjusted the lengths in places, from 4’ to 4’ 6” or 5’.
Given the space constraints and practical limits of equipment resources, the way excavation crews were organized was also analyzed and adjusted.
Since at times there were two headings working both side drifts, there were four available faces. In these cases, Obayashi established excavation crews, girder crews and shotcrete crews that rotated between headings, rather than establishing these capabilities in each individual heading team. This proved to be a more efficient use of resources and improved the overall progress globally.
SEM decision making
Experience levels were specified in the contract for key personnel responsible for the SEM activities. Under the oversight of the Obayashi Tunnel Manager, these key personnel include the SEM Manager, SEM Project Engineer and SEM Superintendents.
The SEM Superintendents worked shifts to facilitate immediate decision-making at the face during the six day, 24 hour week, schedule and were supported at the headings by Walkers and Shift Engineers. Generally two crews were working three 8 hour shifts. Sound Transit recognized the inherent risks involved and agreement was reached during the design stage to have the Designer represented on site during the implementation of the SEM design. To this end HMMJ and DSC provided experienced SEM engineers and inspectors to support the Construction Management team (Parsons Brinckerhoff) and oversee the SEM activities. Shannon & Wilson were also represented on site, providing oversight on geotechnical activities.
Daily SEM Meetings held at the site office followed joint inspections at the headings each morning. Sound Transit, Obayashi and HMMJ/DSC were represented at all of these meetings, to discuss the status of the works and the planned activities for the next 24 hours. Any necessary changes to the Construction Work Plans were agreed, and face maps were presented and discussed as well as the results of probing. A review of the latest instrumentation readings was also included to confirm stability of the headings.
On a weekly basis, shotcrete strength results were presented and discussed. RESS sheets (Required Excavation and Support Sheets) confirmed the required support and pre-support for each tunnel section and were used to assist communications.
Dewatering, monitoring & settlement
Geotechnical instruments specified for ground monitoring outside the excavated tunnels included extensometers, inclinometers, open standpipe and vibrating wire piezometers, as well as optical survey points. Shannon & Wilson technicians read each instrument periodically, with the exception of the survey points that are checked by CH2M crews. Vertical and lateral displacements were tracked from the tunnel invert up through the crown to street level prior to excavation, then just above tunnel crown up to ground surface after SEM mining was completed, for any given tunnel section.
Data from the instrument readings are reviewed at the Daily SEM meetings and to date have shown a maximum of 1.6” (40mm) of vertical settlement over the station tunnels roughly 150ft (46m) below ground.
Street level surface settlement has only reached a maximum of 0.6” (150mm), most of which appears to be attributed to some water line replacement work in Beacon Avenue. In general surface settlement contours have followed the tunnel excavation sequence albeit with very small measurements, and not surprisingly there has been little to no measurable surface settlement over the jet grouted zones. All of this data compares quite favorably to the expected 4” (100mm) anticipated upon completion of tunneling.
Similarly, dewatering wells have proved quite effective with groundwater levels in the predominant sand layers above the tunnel crowns having dropped 20-40ft (6-12m). Groundwater levels are measured at 17 individual locations on site, typically with open standpipe bottom casings and vibrating wire piezometers at intermediate elevations in the upper sand layers. The vacuum-assisted dewatering well system has dropped from an initial 50gpm, through a steady state of 30gpm, down to 10-15gpm.
Waterproofing
The Beacon Hill Station is designed as a tanked structure and equipped with a membrane waterproofing system, which is placed between the initial shotcrete lining and the final lining. It consists of a non-woven geotextile to protect the waterproofing layer and flexible PVC membrane sheets welded together to form a continuous, impervious layer.
A remedial system consisting of a sectioning system and control and grout pipes is also installed. Obayashi contracted with specialty subcontractor Wisko America for the installation of the waterproofing.
Current status
As of May 2008, tunneling at Beacon Hill is complete: The TBM completed the second running tunnel drive by breaking through at the East Portal in early March. Subsequently, the excavation of the two remaining cross passages at the end of April marked the completion of the SEM works.
Currently, waterproofing installation and steel fiber reinforced cast-in-situ final lining construction[2] are ongoing in the North Bound Platform Tunnel, while finishing works are already progressing in the South Bound Platform Tunnel, the Concourse Cross Adits and the ventilation tunnels and egress trunnels.
The waterproofing and concrete works in the Main Shaft, the Ancillary Shaft and the East and West Headhouse are also well underway.
Conclusion
The successful completion of the SEM tunnels was a major milestone for Sound Transit, but also for the contractor, designer, the construction manager, the geotechnical engineer and the inspection team.
The multitude of technical challenges that needed to be addressed throughout design and construction of this project could be resolved by a group of respectful and dedicated professionals working closely together. Open communication, good discipline and having qualified individuals in a partnering environment on site proved to be invaluable components for success.
Fig 1 – Ultimate configuration of Beacon Hill Station Fig 1 Fig 2 – Baseline ecavation and support scenarios Fig 2 Fig 3 – SEM excavation sequence for Platform Tunnels Fig 3 Waterproofing the Platform Tunnels Waterproofing the Platform Tunnels SEM excavation of a Beacon Hill Station platform tunnel SEM excavation of a Beacon Hill Station platform tunnel The Mitsubishi EPBM waits to be pulled through the South Platform Tunnel The Mitsubishi EPBM Centre bench excavation of a Concourse Cross Adit (CCA) Centre bench excavation of a Concourse Cross Adit (CCA)
