Support Crosscut

Digging It: As Bertha bores, tunnel geeks everywhere will be watching

Seattle's tunneling machine is state of the art Credit: Photo: Doug MacDonald

I’ve had personal acquaintance with half a dozen or more tunnel boring machines (TBMs). Including a 25-foot diameter cutterhead TBM I lived with – or suffered with, more accurately – for nine long, cold, wet, hard rock miles below the floor of Massachusetts Bay. In its time, the 1990s, that machine was one of the largest TBMs the world had yet seen.  

On Tuesday, I got a good look inside the new Seattle TBM, the one that will bore the new, Viaduct replacement tunnel. I clambered up and down ladders, inched along catwalks, balanced by holding tight to bronze hydraulic fittings and finally crawled on my knees past a steel bulkhead right to the center core of the monster machine now poised in a pit west of Qwest Field: the green, for now, “cutterhead” that will rotate and scrape and grind its way under Seattle. Jammed into quarters made tight with hydraulics and electronics, I marveled to my guide Greg Hauser, Deputy Project Manager for Seattle Tunnel Partners, “I’ve never seen anything like this.”

“That’s because there’s never been anything like this,” replied Hauser (below), a lifelong project-hardened mole – as tunneling engineers often are called – without batting an eyelash. “This, for a machine, is almost unbelievable. I call it a work of art.”   

Hauser is in charge of testing, launching and operating the TBM. No cockiness accompanies his awe at the scale, power and complexity of the machine. He’s grateful for what he calls the “shakedown cruise,” slow and cautious, that will soon begin for the first 1,500 feet underground.  He spends his time already anticipating the challenges of tuning the machine’s operation and its delivery of each daily advance.

A lot of tunneling goes on around the world, most of it unheralded by surface dwellers for its importance to modern infrastructure. At  57-feet across, Seattle’s new machine has a handful of recent forerunners that approach its scale. But this TBM is big, really big.

It weighs in at 7000 tons, the equivalent of more than 30 orange and black BNSF Dash 9 diesel locomotives. Its 24 huge electric motors — that's the #18 motor below — generate the 25,000 horsepower that will slowly rotate the cutterhead (at about one revolution per minute) as it chews and grinds the ground – TBMs don’t actually drill into the ground. The TBM’s 56 thrust jacks that push the rotating cutterhead (the business end of the boring machine) against the ground ahead exert 44,000 tons of thrust, or 13 times the thrust of the engine and booster rocket that lifted the space shuttle into orbit. The machine is powered by what is essentially a 26 kilovolt extension cord. Seattle City Light built the dedicated feeder line just for this purpose. An extension cord!  

Seattle’s press and public get a close-up look at the behemoth on Saturday before it starts to disappear underground sometime next week. It will re-emerge after its 1.75 mile tunnel drive to Aurora Avenue North near Denny Way.  But the machine is already an international rock star, engaging engineers and other experts who are now in Seattle from Spain, Japan, Germany, the United Kingdom, Ireland, Australia, Russia and France, and just about every corner of the United States. The homegrown workforce of building trades men and women, including sandhogs, operating engineers and electricians bring valuable, pertinent experience from Sound Transit and Brightwater tunnels, has never worked on a TBM like this. 

Moles everywhere will be keeping tabs on how well Seattle’s TBM performs. And how will performance be measured? There are, of course, a myriad of measures to be tracked every day, indeed every hour by command officers like Greg Hauser and their teams. But the two key parameters are utilization and production.

Utilization is the critical ratio of hours the TBM spends mining the ground compared to hours it spends in maintenance or repair or resting. TBM components and systems must hit par day after day in order for the project to meet its time and budget marks. That’s utilization. Production is the number of feet the machine advances in a day, or a week, or a month, against the targets on which the project schedule has been established. 

Lots of things can go wrong, as well as right, in such a large and complex undertaking. Hauser isn’t a man given to speculation, but here’s my own hunch about the pieces that will threaten optimal utilization and production: 

First is the all-important assembly of the tunnel liner rings.

The tunnel that will eventually encase the new roadway gets its permanent structural integrity from a stout, 360-degree concrete lining. The lining is actually 1,450 pre-cast concrete sections, or rings (below), that are assembled within the machine as it bores forward.

Each individual ring section is six feet wide, two feet thick and arrives in 10 curved, pre-cast pieces called liner segments. (A clutch of them is in full view now for anyone traveling the Viaduct, and you’ll see them along I-5 as truck after truck delivers the segments, a few at a time, to the job site from their fabrication facility in Frederickson.) Ring assembly involves fitting those 10 giant pieces into their designated places in each ring and bolting them together, ring after ring after ring.

To get a sense of the intricacy and scale of the job, consider that each curved liner segment weighs 16 tons. Once the many-step supply chain gets a set of segments inside the machine itself, a huge hydraulic arm vacuum-grabs each segment and lifts it into place just behind the cutterhead; 14,500 segments in all, 16 tons apiece and a delicate fit per lift.

The importance of the rings to the tunnel’s structural integrity is obvious, but they also play a crucial role in the construction process itself. Once all the segments for a ring have been bolted properly into place, the two-foot wide front edge of that ring, reinforced by every ring behind it, becomes the stable platform against which the TBM’s 56 hydraulic thrust jacks push to force the rotating cutterhead into the ground ahead. Exactly six fee ahead. Then the machine stops, the thrust jacks retract, the next ring is assembled, the jacks re-engage against the new ring, and the machine pushes ahead again.

Incidentally, the two-foot thick circular rings bolted in place behind the machine make the tunnel about four feet smaller than the machine itself. That’s why TBMs don’t have a reverse gear. The machine is bigger than the tunnel behind it. The only way out is forward!

Since a new ring has to be erected every six feet, the efficiency of ring erection is crucial to the machine’s overall production rate. The cycle of moving the machine ahead can’t go any faster than the cycle of erecting the rings. And the ring assembly, from fabrication in Frederickson to transport into the narrow confines of the machine to the final lifting and bolting into place, is an extended heavyweight do-si-do. Until that routine is perfected, by practice, into a smoothly synchronized muscle ballet of construction workers and TBM equipment, the tunnel construction will move in fits and starts rather than at the steady pace required to meet production expectations.   

The red erector arms that heave the ring segments and the 56 jacks that push off the ring edge to move the cutterhead forward are hidden even now beneath the machine’s bright white shield. That’s a pity, because the efficient sequence of ring erection and each fresh push – sorry, I didn’t make this language up and I’ve never heard a bowdlerized version, even in polite company – six-foot cycle by six-foot cycle, is at the heart of the project’s success. We’ll simply have to await YouTube video of the liner assembly process, which one hopes will become available as the shakedown cruise advances.

Another part of tunnel construction that could delay the process is muck conveyance. “Muck” is tunnel-ese for everything that gets mined out of any tunnel. Hard rock, wet mud, gravel. All muck.

Most of the Alaskan Way tunneling muck will be old glacial deposits. The tricky part is that it will be under pressure as it is loosened and taken into the machine. (The zone in front of the TBM is artificially pressurized. Another story. Keep reading.) The muck stream has to be depressurized so that it drops benignly onto a conveyor, rather than blowing out the back of the machine, fire hose-style. The slow depressurization happens in a special chambered, enclosed “screw conveyor” that is a major mechanical TBM component.

After getting through the screw conveyor the tunnel muck, 850,000 cubic yards in all, will travel on a long continuous conveyor belt to the disposal barge at the north end of Terminal 46. (First stop on its way to the old Mats Mats quarry in Port Ludlow.) The conveyor system will run hour after hour while the machine is mining. Any conveyor glitch shuts the mining operation down.

The six-foot-wide conveyor belt will get longer and longer as the machine goes farther into the drive, eventually stretching to two miles. One point of local pride: While the TBM boring machine itself was built in Japan – and shipped here in 41 parts – the long, long conveyor belt assembly, easily visible at the job site, was made in Auburn at the stateside manufacturing facility of Germany’s Herrenknect tunneling equipment firm. 

That brings us to what is, perhaps, the most important, muck-related question of all: Precisely, very precisely, how much muck are we talking about? Not in total overall cubic yards, but six-foot push by six-foot push. This is where Seattle’s tunnel operation meets a basic imperative for this kind of tunneling: Always excavate just the precise tunnel diameter and no more. Unintended over-excavation can produce voids where they shouldn’t be, or gaps in loose or collapsed material. That could cause the surface above the machine or sink or settle, which really would present a problem for production.

Engineering and operating the TBM to forestall those kinds of problems is a big challenge. A detailed geotechnical survey with dozens of sample cores taken along the tunnel’s route, begun years ago, provides at least intermittent glimpses into the nature of the ground ahead of and above the machine. Yet the very heart of the challenge is an engineering computation familiar to any student of high school level geometry.

The volume of a cylinder: length times pi times radius-squared.

Think of the TBM as an empty steel can, the cylinder, being pushed through the ground. There can be no empty spaces in front of, beside or atop the can (the TBM) lest they trigger the ground surface above or around the can to settle so much as an iota. The solution is simple: make sure that the exact correct volume of material consumed by that imaginary steel can exits on that conveyor belt for every advance of the machine.  

Simple solution; devilish execution, because the exact volume of muck on the conveyor belt can only be inexactly estimated. Success will rely on the quality of the pre-job geotechnical intel, the machine’s own sophisticated sensors, the experience and intuition of machine engineers and operators and the favor of good luck.

The earthen material the machine will encounter is enormously varied, and a lot of the ground the machine pushes through lies within the pressure of the water table. Ground under pressure, like nature generally, abhors a vacuum. Operators must balance the pressure, using compressed air and other techniques, to keep the hole from collapsing.

If they under balance, material might “run,” which you don’t want. If they over-balance, especially when the machine is still close to the surface as it will be in the early going, they could blow a hole right up through the soil. That’s unlikely, but not unthinkable. A famous blowout in 1916 during construction of New York’s subway tunnel under the East River actually shot a worker right through a hole in the river bottom and up to the surface. Amazingly, he lived, and the tunnel was fixed and finished.    

Avoiding even a modest version of that scenario is one reason Seattle’s machine is equipped with arrays of sensors. I saw some of the cutterhead sensor’s black boxes when I toured the TBM. The sensors will constantly measure conditions at the mining face as the machine advances, and also monitor the conveyor system to learn everything possible about the ground around the TBM. And should they detect vulnerability to misadventure – say, at one of the critical points highlighted by the pre-construction geo-technical survey – tanks stand ready to pump sophisticated, stabilizing chemical polymers into the soil. There’s also a big network of above-ground sensors.  

Ground management is also why the cutterhead face comes with 20, platter-shaped, carbide-tipped “disc cutters” (at right) that can fracture the occasional boulder into machine-chewable chunks. (Those boulder-busting disc cutters are manufactured by Seattle’s own Robbins Company, the firm that in the 1950s and 1960s basically invented the modern TBM and made Seattle a long-time center of world tunneling technology.) One other stunning innovation: If one of the cutterhead’s rippers or discs needs replacing, a new one can usually be safely installed from inside the TBM – through ingenious little pressure-tight compartments behind the cutterhead. That is hugely faster and safer than the old way of sending divers down through a big airlock to perform the arduous replacement from in front of the machine. 

So, plenty for Hauser and his mates to think about when they start the machine inching carefully ahead in the early weeks, testing systems, developing operating routines, wrestling with the TBM’s idiosyncrasies and working out its bugs. Moreover, some of the project’s trickiest geology lies in the early going.

This is tunneling at the frontier of the art. It’s not an experiment:  It builds upon, not leaps beyond, proven practice. It’s not an adventure. Adventure denotes a measure of bravado and ego: That’s not this culture. For the mole crowd, this is just another project, albeit a giant one, where they fully expect to encounter and solve the unexpected. I’ll never forget when a 14- foot diameter machine boring a 4.7-mile tunnel in solid rock for the Boston agency I led ran into an unexpected 100-foot seam of soft clay. There was nothing to do but shut down the machine and eke out the 100-foot advance, over many weeks, with picks, shovels, a pint-sized excavator and old-fashioned timber supports. That tunnel was completed and operates for its intended public benefit today.

World-class is a term horrendously over-used. But it is appropriate for describing Seattle’s Alaskan Way tunnel project. The entire underground construction world is watching. And we have the grandstand seat here in Seattle.

All photos courtesy of Doug MacDonald.

Support Crosscut