The Lenihan Dam Outlet Modification Project includes the construction of a new tunnel outlet at the Lexington Reservoir, in Los Gatos, California. Owned and operated by the Santa Clara Valley Water District, the Lenihan Dam’s original steel lined outlet has buckled over the years, requiring numerous repairs. This led the California Department of Water Resources, Division of Safety of Dams (DSOD) to restrict peak discharge to 2m3/sec, compared to the previous 11.5m3/sec. The new conduit will consist of a 1.37m (54”) concrete lined steel pipe, sited inside a 610m long maintenance tunnel running through the eastern abutment of Lenihan Dam.

The project also includes a multi-port intake structure, a shaft located in Lexington Reservoir and an outlet structure adjacent to Los Gatos Creek. The existing outlet conduit beneath the dam will be abandoned in place after completion of the new outlet.

Contract award

The Lenihan Dam Outlet Modification Project was awarded to FCI Constructors, with Drill Tech Drilling & Shoring as its tunnelling subcontractor, with a lowest bid of US$39,173,160. Notice to Proceed was issued on 10 September 2007 with a completion date of 03 September 2009.

The District normally oversees its own construction projects, but in this case (due to the tunnelling works) it appointed Hatch Mott MacDonald (HMM) the Construction Manager. The project was designed by Jacobs Associates (JA) in association with Montgomery Watson Harza (MWH).

Project Schedule

The two-year construction contract includes three project milestones:

  • Extend existing outlet at the downstream end of the dam to allow flows through the existing pipeline by 15 November 2007
  • Complete intake shaft excavation and all other work to allow the reservoir water elevation to be increased above El.576 no later than 15 November 2008
  • Complete all work on the project within 725 days from Notice to Proceed

In order for FCI to complete the intake shaft excavation, the water elevation in the reservoir must be lowered. FCI is allowed, as per the contract, a period of five months from 15 June 2008 until 15 November 2008 to work within the banks of the reservoir. During this time the reservoir level will be kept between El.576 and El.557. If the District had been unable to guarantee drawdown of the reservoir the contract allowed for postponing construction one year until summer 2009; fortunately, this has not been necessary.

Communication structure

The project offices are located in a lot adjacent to the dam. FCI and Drill Tech each have an office and the project owner, construction management team and design support representative share an office. This arrangement provides the opportunity for daily face-to-face interaction between those involved in the project. General progress meetings are held each week with other specific issue oriented meetings (i.e. technical or environmental) held as needed. Major correspondence including submittals and requests for information are exchanged electronically and in hard copy when requested, streamlining communications.

The partnering process is in use on this project. Quarterly general partnering meetings are held off-site and facilitated by a third party partnering consultant as a means of openly discussing the successes and challenges of the project and ways that challenges can be addressed. Online partnering surveys and management level partnering meetings are also held monthly.

Local geology

The project is located in the foothills of the Santa Cruz Mountains, approximately 10 miles from the epicenter of the 1989 Loma Prieta Earthquake. The geology along the tunnel alignment is mostly Franciscan Mélange, characterised by relatively competent blocks of varied rock types embedded in sheared matrices of weaker soil-like or rock material. The matrix material is often shale, sandstone, serpentinite, or clay. Blocks can range in size from sands to boulders or larger. Typical rock types for the blocks in the mélange are sandstone, greenstone, serpentinite, chert, shale, and argillite[1].

The ground encountered in the tunnel so far has been classified using the Terzaghi system as moderately jointed to blocky and seamy when the heading is within large rock blocks, to crushed when the weaker sheared matrices are dominant.

The potential challenges for excavation of a tunnel in Franciscan Mélange include the heterogeneity of the face, the potential for squeezing ground, and the durability of the weaker matrix as an invert material. The presence of a large hard block in the face that is otherwise comprised of weaker material can cause crews to switch excavation methods in the middle of a round. When Mélange is located at an area with high cover there is a higher potential for tunnel convergence, or squeezing ground.

The weaker matrix material is subject to degradation and is easily ground up by repeated vehicle traffic. Due to this, the invert had to be protected during excavation and could not be left as exposed ground. An initial invert, or mudslab, of cast-in-place reinforced concrete was required in the contract. The mudslab was placed approximately once every two weeks prior to weekend closure. In areas where squeezing ground was experienced the mudslab was placed more frequently. This served not only to protect against degradation but also as a means to control convergence by acting as an invert strut.

Mobilisation

The first activity in the field was extending the existing outlet pipe approximately 30m downstream in the Los Gatos Creek. This was necessary in order to prepare the site for tunnel excavation. Portal development was the next activity. The portal is located to the east of Los Gatos Creek in St. Joseph’s Hill and is the future location of the outlet structure and the point from which tunnelling originated.

Excavation of the colluvium and weathered greenstone in the portal cut was performed with a Caterpillar 330CL Excavator and took about two weeks to complete. The portal consists of an approximately 6m high shotcrete and soil nail wall set back into St Joseph’s Hill and was constructed in four lifts.

Initial tunnel ground support

The primary method of ground support in the tunnel is 150mm (6”) steel sets (W6x25) at a typical spacing of 1.2m with varying amounts of fibre-reinforced shotcrete. Timber is used for cribbing and blocking, with a typical blocking point spacing of 900mm. When necessary, in areas of low stand-up time to limit overbreak, No. 9 rebar spiles are used for pre-excavation ground support in the crown of the tunnel. When used, spiles are spaced approximately 300-400mm apart and generally 3m-3.5m deep, typically from quarter arch to quarter arch. Spiles are installed with a lookout angle of approximately 12° to 15°.

Three ground support types are used in the tunnel. The difference in support types is the thickness of shotcrete applied between the steel sets. A description of the ground support types used follows below:

  • Type 1: A minimum of 75mm of fibre reinforced shotcrete, where required
  • Type 2: A minimum of 150mm of fibre reinforced shotcrete
  • Type 3: A minimum of 230mm of fibre reinforced shotcrete

After excavation and mucking has occurred the steel set is installed. First, the leg pieces of the horseshoe shaped steel set are installed at the proper spacing with the use of collar pipes and the proper location with the use of a laser guide. After the legs are in place the arch pieces of the steel set are installed. The arch is lifted and held in place with the roadheader until it has been bolted to the leg pieces and additional collar pipes in the arch are installed. Once the steel set is in the proper location timber cribbing and blocking takes place. Several rounds are often completed and steel sets installed before shotcrete is applied. The shotcrete thickness is completed generally within two tunnel diameters (8.5m) of the face.

Determination of the required initial support is made by the design representative and documented on a Required Excavation Support Sheet (RESS). When specified on the RESS, fibre reinforced shotcrete is applied for Type 1 ground support. If fibre reinforced shotcrete is not specified for Type 1 ground support Drill Tech has the option of applying non-fibre reinforced shotcrete. In those situations Drill Tech has typically elected to do so. In the first 386m of the tunnel Type 1 ground support has been used for approximately 80% of the excavation, Type 2 ground support has been used for approximately 17% of the excavation, and Type 3 ground support has been used for approximately 3% of the excavation.

Tunnelling

The tunnel is horseshoe shaped with an excavated width of 4.3m and an excavated height of 3.8m (figure 1). The tunnel is approximately 610m long with two curves and slopes at approximately 3% upstream (Figure 2). Excavation was completed using two methods: Mechanical excavation with a roadheader and drill and blast excavation. The primary equipment used for excavation included a roadheader, a scooptram, a drill jumbo, and jackleg rock drills. The roadheader selected was an Alpine AM50 reconditioned by Antraquip Corporation. The drill jumbo was used for probe hole drilling and drill and blast excavation. The scooptram used to remove muck from the tunnel was a Wagner ST3 scooptram with a 2.3m3 bucket.

During roadheader excavation the scooptram bucket is filled with muck while the roadheader operates. The excavation typically pauses while the scooptram removes muck from the tunnel. The picks on the cutterhead of the roadheader are replaced approximately every 30m.

During drill and blast excavation the holes are drilled approximately 600mm longer than the round length (1.8m holes for a 1.2m round, 2.1m holes for a 1.5m round). Perimeter holes are drilled approximately 600mm apart and all other holes are drilled approximately 750mm apart. A full face of drill and blast typically requires between 60 and 70 holes. Drill Tech is using both nitroglycerine and non-nitroglycerine based explosives. Standard non-electric tunnel delays 0-19 are used. Each shot is initiated with non-electric shock tube or safety fuse.

The ground conditions present at the face determine the excavation method used. The roadheader is used in softer ground and drill and blast excavation is used in harder ground and to remove hard blocks in a softer face. Ground conditions beyond the excavated face are investigated via probe holes drilled into the face. During probing, two holes are drilled, each 15m long. Subsequent probing sessions are required to overlap by 6m. Locations of the probe holes in the face are in the left quarter arch and in the lower right of the face. Lookout on probe holes is typically 5° to 7°.

Average cycle time statistics have been generated from data collected by inspectors from HMM for each excavation method. Cycle time for each round includes the time required to install spiling (if used), perform excavation by either method, remove tunnel muck, and erect the steel set. Shotcrete application is not included in cycle time as it is typically applied after several rounds are completed. Figure 3 shows the breakdown of time required for each activity in the different cycles based on the average cycle length of a round. The average time to date to excavate and place a steel set with the roadheader is 3.4 hours compared to 6.4 hours for drill and blast with a 1.2m steel set spacing and 7.3 hours for 1.5m spacing.

The largest factor in cycle time is the time required to remove the muck from the tunnel. One benefit of roadheader excavation is that it allows for concurrent excavation and mucking, whereas mucking must take place after detonation of explosives and an appropriate ventilation time when drill and blast is used. Time required to remove muck from the tunnel was seen to increase with tunnel advancement because of the travel time required for the scooptram to discard muck with each trip.

Steel sets were placed at 1.5m spacing for 30m of the tunnel, as per the RESS, because ground conditions were favourable. The increase in spacing allowed for an increased rate of tunnel advancement. When ground conditions changed excavation returned to the 1.2m spacing typically required by the contract.

Tunnel excavation began on 07 November 2007 and was expected to be complete by the time of this publication. Excavation was performed 24 hours a day with two 12 hour shifts Monday through Friday each week with mudslab placement approximately every other Saturday. At the time of writing this article, the average production rate was 3.4m/day with the highest production rate for a single shift of 6m, the highest production rate for a single day of 7.3m and the highest production rate for a single week of 28m.

Convergence monitoring

In the first 385m of the tunnel, Drill Tech installed convergence monitoring points at seven locations determined by the designer. Monitoring points are typically installed within 1.8m of the face. Measurements are taken with a tape extensometer. The first measurement was generally taken within 48 hours of installation of the monitoring point. Subsequent measurements are generally taken every other day for the first two weeks and weekly thereafter. The rate of convergence is seen to slow over time. To date, maximum convergence measured in the tunnel is 0.80%.

Pre-excavation grouting

The first time that groundwater inflows reached the pre-excavation grouting criteria of 38l/m from any single feature or 0.38l/m per 300mm of probe hole was on 11 March 2008 at tunnel station T9+80. Flows in excess of 410l/m were measured emanating from 4 probe holes in the face. The face at the time was covered with shotcrete and consisted of crushed shale. After 12 shifts of grouting that pumped 100 bags of Type III cement, tunneling resumed.

Site restrictions

The footprint of the site at the portal that is available for staging and laydown is limited. The portal site is located at the base of Lenihan Dam. This requires vehicles and equipment to travel down a 20% grade and around sharp turns in order to transport material into and out of the site. The site for the future intake structure lies within the banks of Lexington Reservoir. Work within the banks of the reservoir is only allowed to take place between 15 June 2008 and 30 November, due to environmental concerns and permit conditions. There is also limited laydown and staging area near the location of the future intake structure. The limitations imposed by the overall project site require special considerations for material handling and scheduling work.

Safety

The project currently has an excellent safety record with over 30,000 man hours worked and no reportable incidents. All personnel who work on-site at the portal or in the tunnel go through a safety orientation before they are permitted access to the site. Personnel who go through the safety orientation are given a sticker to place on their hardhat to show that they are eligible for site access. Safety measures include a mine rescue team and half face respirators during roadheader excavation.

A five person mine rescue team was trained. Because of the dangers associated with using heavy equipment in the confined tunnel size, the number of people permitted in the tunnel at any one time is kept to a maximum of nine and foot traffic is kept to a minimum while the scooptram is operating.

Half face respirators are required to be worn during roadheader excavation due to the amount of dust generated even with the ventilation system set to exhaust and the use of water from a hose and misters on the roadheader to wet the face.

Environmental

The location of the project site led to a number of environmental challenges. Water quality was an issue of particular concern because of the proximity of the construction site to Los Gatos Creek. Strict water quality standards are enforced for any discharge into the creek and require the dissolved oxygen, temperature, and the change in pH and turbidity to be recorded. The project site is also a potential habitat for a number of protected species in California. Special care had to be taken when dusky footed woodrat nests were relocated out of the limits of disturbance for the project.

The contract requires that noise pollution be monitored during construction activities. Noise levels from construction are not permitted to exceed county limits of 60dBA daily, except Sundays, between 7.00am and 7.00pm, and 50dBA daily from 7.00pm to 7.00am. Noise monitoring is required when a new construction activity begins and measurements are taken from a nearby residence, half a mile from the construction site on the other side of State Route 17. To date, noise from construction activities has not been seen to exceed the county limits with noise from Route 17 registering higher than construction noise.

When drill and blast excavation is used FCI is required to perform vibration monitoring with maximum allowable peak particle velocities of 12.5mm/sec at frequencies of 10 hertz or less and progresses linearly to 50.5mm/sec at a frequency of 40 hertz or above. Vibration monitoring occurs approximately 75m from the portal area at the abutment of an existing pedestrian bridge and at the closest abutment of a bridge across the dam approximately 290m from the portal.

To date, vibration monitoring during drill and blast excavation has not caused peak particle velocities above the maximum allowable levels. The trigger level of the seismograph is set at 60mm per second and in most cases was not met.

Current progress

The tunnel will hole-through at the base of the intake shaft. The excavation and support of the intake shaft will be conducted between 15 June 2008 and 15 November 2008 during reservoir drawdown. The shaft will be a minimum of 4.5m in diameter and approximately 10.6m deep. The ground support for the shaft is designed to consist of steel liner plate and steel ring beams.

Final lining of the tunnel will be cast-in-place reinforced concrete. The walls and arch will be 300mm thick and the final invert will be 450mm thick. FCI plans to use two 9m traveling forms. The estimated rate of concrete lining placement is 18m per day.


ACKNOWLEDGMENTS

Special thanks are due to the Santa Clara Valley Water District and its staff for permission to publish this paper as well as colleagues from Hatch Mott MacDonald including Dave Young, Mike Murray, and Ronnie Strasser for their mentoring and encouragement

Roadheader excavation of the new outlet Portal construction underway Installation of spiling Standing a steel set with the roadheader Application of shotcrete to blocked steel set Figure 1 – typical tunnel cross section Figure 2 – Longitudinal section Drill and blast excavation Use of face respirators Figure 3 – Average cycle times for each excavation method