THE EUCLID CREEK TUNNEL (ECT) is the first of seven planned storage tunnels under "Project Clean Lake" that will relieve over flows in the Cleveland-area sewer system. This 18,000ft-long (5.5km) 24ft-diameter (7.3m) tunnel is roughly 200ft (61m) beneath Cleveland’s streets, with 3,000ft (914m) traversing beneath Lake Erie. When operational, the tunnel will hold approximately 60 million gallons (228 million liters) of combined stormwater and wastewater. After heavy rainfall, the combined sewage will be pumped via the adjacent Tunnel Dewatering Pump Station to the Easterly Wastewater Treatment Plant, protecting the valuable freshwater resource of Lake Erie and its tributaries. The annual one billion gallons of storage is enough to fill more than 1,515 Olympic-sized swimming pools. In American Football terms, it would overflow the Cleveland Browns Stadium every year.
Hatch Mott MacDonald (HMM) was appointed lead consultant for ECT back in 2006. The firm was responsible for overall design of the project, geotechnical oversight including the geotechnical baseline report preparation, cost estimation, risk analysis and preparation of the plans and specifications (tender package). The McNally-Kiewit Joint Venture (MKJV) secured the USD 198.6M contract to build the project in December 2010, with notice to proceed given in April 2011. This highly successful project has experienced no significant claims or delays and is now completed and ready to accept flow later this year.
PROJECT GEOLOGY AND ALIGNMENT
The ECT traverses through predominantly residential neighbourhoods from Bratenahl south of Interstate 90, northeast through Cleveland to the NEORSD’s Easterly Wastewater Treatment Plant. There the tunnel swings under Lake Erie for nearly 3,000ft (914m), and terminates near Euclid, Ohio, at the intersection of Saint Clair Avenue and 185th Street.
The TBM had a 27ft (8.2m) diameter cutter head that was affixed with 44 single-cutter discs and eight dual-cutter discs. The TBM was rotated by nine hydraulically-operated torque hubs, and had 32 thrust cylinders as well as 16 thrust pads (two cylinders per pad). Each pad had a bearing area of 159,952sq.mm (approximately 0.16m2 or 247.9in2).
Like most of the NEORSD tunnels before it, the ECT mined through the Chagrin Shale formation. This formation is massive and relatively stable, so earlier tunnels in the NEORSD’s program, especially of smaller, 8 to 13ft (2.4 to 4m) diameter, were typically successful. However as the NEORSD required larger and larger tunnels to meet their Consent-Decree-mandated storage volumes, some of the Chagrin Shale challenges and risks became more apparent, leading to past project delays and ultimately litigation. The two main risks included excessive H2S and methane infiltration, and deterioration of the crown and invert. This deterioration was caused by thin bedding coupled with relatively high stress, and was exacerbated in the crown by gravity and in the invert by water and train traffic. The traditional method of support had been a two-pass system using steel ribs and timber lagging, followed by cast-in-place concrete. Control of the rock overbreak proved difficult with largediameter two-pass operations, and invert degradation led to train derailments. The NEORSD and HMM set out to mitigate these risks so that there would not be a repeat during the Euclid Creek Project.
TUNNEL INNOVATIONS
During the course of the design, HMM introduced several key innovations. For the first time, a NEORSD tunnel would use one-pass, precast, bolted, gasketed concrete segments reinforced with steel fiber alone (no rebar) to line the tunnel. This improved the overall quality of the tunnel and saved several months from the construction schedule, while mitigating scores of potential risks such as overbreak, gas, derailments, and work stoppages. However, several residual risks were created by the use of segments in conjunction with an open-faced TBM in the shale bedrock.
First and foremost, the TBM would cut a void that would need to be filled immediately behind the TBM shield to provide segment support. This would be difficult to do with an open-faced TBM, but no one wanted to specify a closed-face TBM just for this eventuality since the rock was generally excellent and little to no groundwater was expected. In addition, the one-pass system meant that overbreak above the TBM or lining could not be easily seen, and more importantly, any overbreak ingested at the TBM face would result in voids behind the lining. Such voids could lead to lining instability in general and operational requirements also dictated that these voids be filled so that the lining could resist internal surge pressures during tunnel filling in a storm event. To fill the tail void behind the segments, HMM specified a twopart sodium-silicate-accelerated grout system, injected through the tail of the TBM. However, this met resistance as many in the industry were quick to point out that nowhere in the world had this been attempted in conjunction with an open-faced TBM.
While this was true, the design team felt that the risks mentioned above, and additional residual risks from the tail void injection, could be mitigated in the design and construction. The required residual risk mitigation included: keeping grout from flowing forward around the shield, potentially locking the TBM in place and or flowing into the open TBM face; avoiding ‘ring squat’ into the annulus overcut void around the lining while the grout flowed and set; detecting and grouting overhead voids as the TBM progressed.
To provide immediate support to the segments, the two-part sodium-silicate-accelerated grout was designed to gel in 30 seconds or less to limit forward flow and provide resistance to segment movement. Additionally, the gel strength was specified to be sufficient to provide a firm foundation for the TBM trailing gear and other loads. During construction, the MKJV developed and tested more than 60 mix designs until a working annular grout mix was created and proven through full-scale field trials. It is believed that this system of tail void grouting behind an open TBM in rock was indeed a world first and an innovation that provided great benefit to the NEORSD and their rate-payers.
Although this annular grout may have been the most critical project risk-mitigation element, probing through the lining (and 3ft into the rock above) was also required to check for overbreak in the rock and resulting voids. Contact grouting was required to extend into any overbreak areas or voids to function both as consolidation grout as well as contact grout.
Several other innovations were introduced on the ECT project, including:
¦ The successful fabrication, transport and testing of full-scale plastic-fiber-reinforced concrete segments. This exercise, which to our knowledge had not been done before, will pave the way for potential use of plastic fiber on future segmentally-lined tunnels.
¦ Physical modeling of baffle drop structures. While these had been used successfully in Cleveland for 100 years, the design was strictly empirical – old working designs were copied and there were no reliable design algorithms to upsize the flow beyond what had been shown to work in existing structures. The solution, which was approved and funded by NEORSD, was to partner with the Iowa Institute of Hydraulics Research to build a scale model of the baffle structure. The data was evaluated and the results were fed back into the design process. This allowed the design of larger, higher-capacity baffle structures and the data from this project is now being used all over the world, on projects from the United Kingdom to New Zealand.
¦ While it does not qualify as an innovation as such, the starter and tail tunnel final linings used plastic fiber as the sole reinforcement. According to the fiber manufacturer, no one had ever used plastic fiber as the sole reinforcement for a cast-in-place tunnel lining prior to the ECT project.
PROJECT RESULTS
After some early learning curve issues with accelerators and other modifications, the MKJV ultimately averaged 17 rings per day with best days in excess of 30 rings per day (150ft or 46m) using three shifts. The vast majority of the tunnel was installed with insignificant stepping, lipping or other offsets in the segments. Where segments had squatted or moved prior to grout gelling or other reasons, hydrophyllic polyurethane grout was injected behind the segment gaskets. However, this was only required if the movement exceeded the tolerance on required gasket pressure resistance, and very few such instances occurred.
Gas was mitigated effectively via a combination of the segmental lining and a robust ventilation system. Slightly elevated levels were noted occasionally, yet there were no shutdowns of note. Invert degradation was mitigated by the concrete segments, and the MKJV was able to build the tunnel under factory-like conditions rather than fight overbreak and other risks for past tunnel projects in the area.
Check-hole drilling was routinely performed several feet through and behind the lining at the crown, using dedicated drilling equipment on the trailing gear that was located five rings behind the TBM shield. Contact grouting was immediately injected as necessary through these same check holes.
Very little overbreak was observed using a bore scope through the check-holes, and typically a fully grouted void was present. This was confirmed when the horizontal adits were excavated at the connection points to the tunnel. The shale was hand-mined revealing the annulus grout in nearly perfect encapsulation of the segmental ring. The two-part accelerated grout method has proven to be very successful, and the NEORSD is using the same system on the Dugway Storage Tunnel (DST), which will connect directly to the ECT tunnel at Shaft 1 near the Pump Station (see page 24). The Salini Impregilo/Healy JV recently started construction on the DST project, (approximate USD 153M Contract Value).
This tunnel, designed by the HMM/MWH Americas joint venture, is of the exact same diameter as the ECT project, albeit significantly shorter. However future designers should remain cautious, especially if using this method in rock with potential for large groundwater inflows. The evaluation of the plastic fibers was promising, particularly for the concrete segments, and the NEORSD is evaluating the use of plastic-fiber-reinforced segments in an upcoming tunnel to further prove the viability of these fibers. The physical baffle-structure modeling allowed efficient design of structures many times larger than what has been done in the past. In fact, additional physical modeling on subsequent projects has yet to indicate a size limit. Furthermore, by offsetting the dividing wall from the center of the shaft into the area traditionally reserved for worker access, the flow capacity can increase without increasing the shaft diameter.
CONCLUSIONS
By any metric, the ECT project was extremely successful, and has won the International Project of the Year Award from the Tunnelling Association of Canada and a High Commendation from the NEC/ITA. However, it must be mentioned that the project’s success could only have been achieved via a constant and open collaboration between the NEORSD, HMM, MKJV, Herrenknecht, segment supplier CSI/Hansen, grouting specialists at BASF, and others. The staff and expertise provided by all parties was top-notch and all knew that if proper attention was not given to key, critical issues such as the grout make up and injection, this project could have had a very different ending
