Support Crosscut

Why is Bertha stuck? WSDOT’s 5 levels of tunnel-drilling hell

Tunneling crews discuss their progress as they operate Bertha, the world’s largest tunneling machine, in November 2013. Credit: Photo: WSDOT

The world's largest tunnel machine, Bertha, is stuck 60 feet beneath Downtown Seattle. The tunnel contractor has drilled down, pumped out and pressurized a gap in front of the machine and is sending divers down 24/7 to try to figure out what brought the machine to a standstill nearly seven weeks ago.

According to the Washington State Department of Transportation (WSDOT), diver teams are working in shifts examining the cutterhead from an approximately 14-inch vertical crawlspace created between a claylike Bentonite barrier sprayed into the ground ahead of Bertha and the top half of the towering, five-story tall behemoth.

Depending on what they find, here are five potential scenarios for Bertha's future, ranging from inconvenient to disaster:

Scenario One: Pipe Gone Wild.

Divers may find more remnants of the 8-inch diameter steel pipe tangled within the cutterhead of the machine. If so, they'll need to clear it out and determine how badly the cutterhead may have been damaged by the encounter. That will take time, but the bigger question will be how much metal junk remains in the 9,000 foot path ahead of the machine. If there is more, then finding it, removing it — and being certain about it — will take time and resources.

Scenario Two: Big Rock Indigestion

Divers may find large rocks or boulders jammed into the machine. In that case, the fix may be to change some of the soil cutting teeth back to disc-style rock cutters like those used when the machine launched from its concrete starting pit. Disc cutters can break rocks, provided they can get a purchase on and exert enough pressure to do it. If not, then some other plan will be needed to remove large rocks. Bertha does not have a secondary rock crusher mechanism as do some other TBMs.

The tunnel route is a diverse patchwork of ground conditions including "non-cohesive" sandy, gravelly, clayish earth, all below the water table. Similar to the metal pipe, it is hard to know beforehand how many rocky boulders lurk in the mix along Bertha's path.

Scenario Three: A Bearing Surprise

At the core of any TBM is the vital main bearing. There is a small chance, observers worry, of problems related to those that Bertha's builder, Hitachi Zosen, had and fixed during testing of the machine at the factory. Bertha is a giant, and the pressures at the bottom of the machine vary significantly from those at the top of the machine, but TBMs are routinely built to cope with that fact.

It would take something truly extraordinary to prang the main bearing, but if it were damaged, it would be a major showstopper. UK Tunnel expert Alastair Biggart: "The heart of… any TBM, is the main bearing. If this should be damaged during tunnelling it is a major problem for the project. It is possible to change a main bearing within the tunnel, but extremely time consuming and expensive."

Scenario Four: The Biggest Problem

If the divers take a good hard look and don't find any obvious problems, that could mean the problem is Bertha's design itself. This would be the biggest problem of all. TBM selection is often the most crucial decision of a tunnel job. The situation in Seattle may turn out to settle a contentious technical question: Which of two competing types of machines was best for this job?

Tunneling here is known to be tricky. "We can never take for granted the tendency of the soils in Seattle to misbehave if given the opportunity," consulting project managers from the firm Hatch Mott MacDonald wrote in a report to WSDOT.

To this day, there is industry disagreement about what machine design is best for Seattle's conditions. In its "design-build" process, WSDOT specified a category of machine (closed face pressurized), but left the design choice up to the main contractor. Two different types of machines were contenders for the work: An Earth Pressure Balance machine (which Bertha is) or a Slurry/Slurry Mixshield machine. There are important technical differences between the two and each has vocal proponents.

Four firms made proposals to build machines for Seattle Tunnel Partners, which picked Hitachi-Zosen's Earth Pressure Balance concept.

One losing bidder (and former holder of the World's Largest Tunnel Machine record) has been highly critical of this decision. The founder of German firm Herrenknecht, Dr. Martin Herrenknecht, has vigorously questioned the decision to use an EPB machine in Seattle. At last year's World Tunneling Congress in Geneva, industry journal reported:

The world will be watching this project closely, and none more so than Dr. Martin Herrenknecht, who could not resist questioning the choice of machine during a lively question and answer session. "Why did you choose an EPBM? I would have selected a Mixshield slurry system," he said. "I wish you good luck, but you will have problems," he warned…

Dr. Herrenknecht added in reference to anticipated boulders in the glacial deposits of the drive in Seattle: "You would be much better cutting the boulders. Settlements will be a problem and how will you get in under 5.6 bar pressure? With this high water pressure I am quite afraid. I think that finally you will end up with a slurry machine on a screw conveyor." In response, an unruffled [STP representative] Sanz said that Dragados and Seattle Tunnel Partners were happy with their choice of machine. "We have analysed the situation carefully and for us an EPBM is good enough."

What happens with Bertha may well settle who is right.

Bertha is due to encounter far higher underground pressures along her route than she currently is. The need for major design changes would necessitate extremely costly work. WSDOT appears aware of this, as the Seattle Times reported last week that Todd Trepanier, WSDOT administrator for Highway 99, said that it was entirely possible that Bertha might need to be modified.

A Fifth Factor in Everything: The Human Factor

TBM operators have a unique learning curve on every project. Most machines are one-off devices. Both machine and operators must learn how everything, especially the ground being tunneled through, behaves. It takes time and technical discipline to find the precise way to run a machine to achieve optimum utilization and production. What drive torques, pressures or advance rates to run, how many and what type of soil conditioning agents to use, how to have round-the-clock crews (Bertha's crew numbers about 28 people during operation) synchronize their practices and how hard the whole system can be pushed are all vital questions.

Concerned about operational factors, WSDOT demanded contractor STP answer key operational questions, including:

  • "Information about possible deviations from planned operating procedures
  • "Details about changes to soil conditioners prior to stoppage
  • "Soil conditioning plans moving forward…
  • "List of factors and decisions that led to operating machine at extremely high temperatures prior to stoppage on Dec. 6"

It may be that in the process of learning how to optimize Bertha's operations, she was pushed too far and got all gummed up. To overwork WSDOT's odd anthropomorphic Bertha PR-shtick: Bertha may need better coddling and swaddling. It may prove essential to stay within a tighter band of operational limits that are still to be determined. If that's the case, rather than one of the worse scenarios, it will be a relief.

The cutting-edge technical challenges of the Seattle tunnel put it closer to a moonshot than a freeway job. It is ambitious and technically thrilling and the teams are packed with global talent. Still, important unknowns are inherent in tunneling work and this job is fiercely complex.

What the divers find out in the hours and days ahead is likely to reveal whether Bertha's troubles are ordinary, major or will be downright disastrous to the budget and schedule. No matter which, an aggressive recovery plan — technical, financial and political — will need to emerge quickly from somewhere.

Share On:
Support Crosscut