Seattle is the founding city of The Robbins Company, and a place where I lived for nearly 15 years. I remember commuting on the SR99 viaduct highway while working at Robbins early in my career. The new SR99 Viaduct Replacement Tunnel Project is therefore of great personal interest to me.
The industry is all too familiar with Seattle’s SR99 Tunnel and its TBM Big Bertha. Much has been written with regards to the 17.5m diameter TBM needing repairs after about 300m of boring.
Robbins was a relatively new entry into the EPB/soft ground tunneling business when tenders were called for the SR99 project in 2011, and we made a concerted effort to get the order for this particular TBM. We teamed up with Japanese TBM manufacturer Mitsubishi Heavy Industries (MHI) to get the order. Robbins has had an association with MHI for more than 20 years, and there are jointly designed machines operating around the world on projects in India, China, the USA, and in more countries besides. MHI has built more than 1,000 EPB machines, and, in my opinion, the Japanese TBM manufacturers are further advanced in EPB technology than their European and American counterparts.
During the process of trying to win this order we learned a lot about the geology, as well as the contractors’ TBM specification requirements. The main contractor, Dragados [in JV with Tutor Perini] is very well experienced in soft ground tunneling technology, and it developed a high-level specification for the machine. All of the prospective TBM suppliers were required to quote and, if successful, supply to this standard. We eventually stepped out of the tendering process as the lower prices and greater assumption of contract risk offered by our competitors made the TBM supply on this occasion an impractical business option for us.
The current situation at the SR99 project is more positive than the media tends to paint it. The project design consultant performed a commendable job when laying out the tunnel route by building in a contingency plan. Boring through glacial till, even with modern TBMs, is never an easy task, as previous projects like the Brightwater Conveyance Tunnels have taught the city of Seattle. This is doubly so along the Seattle waterfront, which includes man-made fill, utilities, and buried refuse. In such ground, TBMs can encounter rapidly changing geology; pockets of groundwater; abrasive soil; and man-made objects such as unmapped disused pipes; foundation piles; and more.
Aware of the problems that can develop while using an EPBM in glacial till under a city with a lot of backfill, the SR99 designer wisely developed a contingency plan. The strategy, in addition to pre-planned safe havens, involved a ‘shake down’ stretch of tunnel running under no buildings. If problems did occur, repairs to the TBM could be made by sinking a surface access shaft at this location. Unfortunately the need for that repair event occurred shortly after the machine commenced excavation.
The reasons for the failure of the cutterhead seals, and, potentially, the cutterhead main bearing, are yet to be determined. I doubt there will be any signs of failure of the main bearing when the crews get a chance to inspect it. However, all parties involved are wisely taking precautions and installing a new main bearing in addition to the seals.
Seattle Tunnel Partners and WSDOT have in place a panel of experts to advise them on the highly technical details of the TBM design. I personally know several of these experts and they are well qualified to recommend and supervise the necessary repairs and procedures to get the TBM into a condition where it is able to finish this tunnel.
Having been in the TBM supply business for quite a few years, I unfortunately have to admit having been in a similar (fortunately not as well-published!) situation as the TBM supplier, and on more than one occasion. This situation, involving significant TBM problems at the beginning of a TBM drive, can result from many different factors and is not unique to the SR99 project. In fact, Robbins recently had a similar situation (admittedly on a smaller scale in terms of both public and financial impact) on a project in Turkey known as the Kargi HEPP. Despite extensive pre-planning, unexpected ground was encountered, which resulted in several in-tunnel stops and machine modifications in the first few hundred meters of the tunnel.
What happens in these situations is that you pull in the best minds with the most experience and immediately analyze the problem. The ultimate fix often ends up as a multi-level solution. You must ensure you have the problem under control, plus, take additional measures to monitor the vulnerable components and operating procedures. At Kargi, this process resulted in the remainder of the project being finished without significant TBM problems. Without a doubt a similar process is going on at SR99 with Hitachi Zosen engineers, the contractor’s specialists, and the city’s board of experts.
Being one who is keenly interested in this project, I believe that this TBM will soon be back to boring with a new completion date, which will be fulfilled. I am optimistic that this project will one day be seen as a positive in the tunneling industry, and one during which many lessons will have been learned and many advancements made. Such advancements will be put to use in Seattle and in other cities that will benefit greatly from the excavation of more underground infrastructure.
This article first appeared as a blog, The Light at the End of the Tunnel: The Positive Side of Seattle’s SR99 Project on the Robbins website
Personal thoughts on the application of mega-TBMs
Yes for decades TBMs, and a few digger shields, have had similar problems. That is why it is so SAD to see what highway engineers are doing for tunnels in Seattle and Los Angeles, thinking: “I can run a dozer. TBMs are not much different.” In 1985, I was assisting the first underground railway in Guangzhou, China for the heavy rail subway. French engineers wanted to give them a 25ft diameter enclosed face shield. I told them as long as it is free and to ask for 10 spares for anything made of rubber, plastic, and flexible piping. Bertha probably had an equipment and operator error. It is called torque for a 60/30ft lever arm. Probably no pressure gauges in the TBM to measure torque or the 1% change in face alignment or the operator wasn’t watching and no on-board CAO system was in place, operating, or being read. With torque, seals, gaskets, and bearing crash. This isn’t rocket science. TBMs have been operating since 1982 or before. I prefer open face mining like the 2014 completion of Caldecott fourth bore highway tunnel and the Devil’s Slide highway tunnel in the San Francisco Bay Area in California. No TBM problems. Bertha's problems are applicable for comments in California where the Los Angeles SR-710 DEIR [draft environmental impact report] is proposing four (4) 60ft diameter TBMs to meet at 2 miles in. No inventory of abandoned water wells has been conducted. Many homes in the 1900-1930 period may have had their own water wells for irrigation. Sometimes they used steel pipes for casing the wells and most drillers of the period would use standard oil field technology – up to about 300ft deep – and including 8in steel pipes. If abandoned they would normally dump cement and sand down the hole. If Big Hilda TBM ran into such a pipe without proper pressure monitoring – poor Hilda stuck again – we told you so!
Dr Tom Williams