Fatal collapse on Cologne’s new metro line
Fatal collapse on Cologne’s new metro line Mar 2009
Shani Wallis, Editor
p1 Courtesy Der Spiegel

Site of the devastating collapse

Failure or ‘boiling’ of the invert in a deep shaft excavation is the probable root cause of a catastrophic and fatal disaster on the new underground tram line project in Cologne, Germany. Failure of the excavation caused complete collapse of one building last week on Tuesday 3 March and claimed the lives of two residents in the partially collapsed apartment buildings either side. The disaster comes more than one and half years after excavation of the running tunnels and during construction of a crossover link between the two.
Workers, presumably excavating the core of the large straight-sided shaft in which the crossover was to be built, evacuated the works when they sensed something as wrong and warned 10 to 15 employees and visitors in the city’s historical archive building adjacent to the shaft and others to run for their lives just minutes before the heavy concrete building and its heavy weight of historical books and papers collapsed partly into the shaft and into the gapping hole in the sidewalk and beneath the building.
p3 Courtesy CNN

Damaged apartment building

The body of one victim was found on Sunday morning (8 March) and the other is still missing and presumed dead beneath the rubble. Dangerous conditions all around are preventing any major clearing. The damaged buildings either side are severely compromised and disturbed ground rules out the use of heavy lifting equipment until the area is secured. One of the damaged buildings was demolished yesterday (11 March) and volumes of concrete have been poured into the voids but more stabilization is required before investigations and recovery can begin.
While the owner, Kölner Verkehrs-Betriebe AG, and the contractor are releasing little information into the cause of the collapse, others familiar with tunneling and deep foundation excavations in Cologne suspect an under performance of systems installed to control the high water content in the permeable soils beneath the city.
p2 Courtesy TUNNEL

New underground tram line

“Managing the area’s high ground water content together with perhaps some irregular geological conditions appears likely to be the root of this disaster,” said Dr.-Ing. Roland Leucker, Managing Director of STUVA, the Reseach Association for Underground Transportation Facilities based in Cologne. The geology of Cologne on the banks of the Rhine is notorious for its high ground water table in the sand and gravel deposits. The water table lies within meters of the surface and is influenced by fluctuations in the flow of the river. High spring melts that swell the river often cause construction of open pit foundations and basements for new buildings to be backfilled with water to counterbalance rising hydrostatic pressures outside the diaphragm walls used commonly to construct such excavations in the city.
The collapse occurred at about 2pm on Tuesday afternoon at the height of the working day. It occurred on the southern section of the new North-South metro tram line. The 4km of twin tube running tunnels and six underground stations is divided into two contracts with the Bilfinger Berger, Züblin, Wayss & Freytag JV working on the 3km southern section with five underground stations, and a consortium lead by Hochtief and including Strabag and ground engineering companies Keller, Brueckner and Bauer, completing the 1km northern section with the sixth underground station. A spokesperson for the owner, KVB, confirmed that the crossover was the last section of excavation on the project to finish. All other excavation was complete.
The new metro line runs between 100-500m parallel with the river and the collapse site is about 500m from the river bank, but according to Leucker, influences of the river are unlikely to have been a causing affect. “It is also unlikely that a broken water main or sewer could have caused the amount of water that appears to have run into the shaft so quickly.” The possibility of a failure in the diaphragm wall is said to be unlikely to have caused such a rapid sequence of events. Suggestion that the interfaces between the segmental lining of the four running tunnel junctions with the preinstalled diaphragm walls might have caused the problem is also deemed unlikely although sealing of the break-in and break-out eyes of the running tunnels in the diaphragm walls is believed to have been the annular grouting of the segmental lining alone. A jet-grouted block of treated ground outside the diaphragm walls at the eyes is believed not to have been installed.
p4 Courtesy Der Spiegel

The Rhine River lies some 500m from the collapse site

“Although still yet to be confirmed, it is almost certain that the tunnels themselves did not fail,” said Leucker.
Instead, and also yet to be thoroughly investigated, it is suspected that a breakthrough or ‘boiling’ of the shaft’s invert caused rapid high pressure inflow of ground water, bringing with it volumes of material, and creating the void outside the diaphragm wall into which the archive building collapsed. “The line runs up to 28m below the surface, and with the water table at about 8-10m below the surface, there is about 2 bar water pressure acting on the excavation at invert level,” said Leucker. “It is unknown how deep the diaphragm walls are in this case but for most diaphragm wall constructions, they are about 1m thick and extend into the more watertight tertiary sands and gravels beneath the upper, more permeable quaternary deposits of the Cologne region.” It is believed there had been some problems with dewatering the core ahead of starting shaft excavation.
p5 Courtesy STUVA

Crossover superimposed on the collapse

General construction of the new line began in early 2004 with installation of the heavy diaphragm walls around open-cut access shafts and station excavations. To excavate the 2,700m of 7.3m i.d. segmentally lined running tunnels of the southern lot, the JV selected two Herrenknecht slurry Mixshields. The machines were launched on 6 June and 12 July 2006 and completed their respective drives on 12 July and 4 August 2007. En-route the TBMs passed through the single crossover zone - the diaphragm walls of which had been installed but the core had yet to be excavated. They also mined through, or were pulled through, the open-cut station excavations and drove either side of the biggest open-cut station excavation on the line, also supported with diaphragm walls.
p6 Courtesy WDR

Running tunnel filled with inflow material

Core excavation of the crossover had waited until use of the tunnels for access purposes could be discontinued and allow for breakout of their segmental linings. Photographic evidence indicates that excavation of the core in the more than 30m long x 25-30m wide x more than 25m deep crossover excavation running diagonally across the alignment of the two running tunnels, had uncovered the tunnels and that dismantling of the segment lining waslikely finished. Gravel and sand filling about one-third of the running tunnels indicates that breakout of the segments was finished and that excavation was at or near the deepest level. It is reported that the excavation had reached the invert level in parts and that the first blinding layer of concrete was being poured ahead of installing the rebar reinforcement and casting the excavation’s base slab, which would be likely 1m or more in design thickness.
p8

Buckled steel decking over the shaft excavation

As Leucker explained, the diaphragm wall method of construction for deep excavations in Cologne is standard. The diaphragm walls are installed and the core usually dewatered ahead of dry excavation. Sometimes the core is excavated in wet conditions and water pumped out after the base slab is tremmied in. It is believed the crossover excavation had been dewatered ahead of excavation and that the base slab was in the process of being constructed. The alternative to dewatering is ground-freezing which is being used for construction of connections between the large open box excavation of the line’s big underground station and the running tunnels that pass outside the walls. Ground freezing however is expensive and alternatives are adopted where possible.
p8

Broken dewatering pipes

Dewatering of the large open-cut station boxes and shafts ahead of core excavation was a major operation on the contract and large diameter blue-painted pipes on the surface convey extracted ground water to the river for discharge. Excavation of the new line in the neighbourhood of the city’s main station and its famous cathedral, runs beneath the streets where possible but comes close to property lines in several areas. At the site of the collapse, the wall of the crossover excavation in the street is within 5m from the front of the buildings. Heavy steel girders and steel plate decking, with a small access opening incorporated, covered the large box to keep traffic moving during excavation of the crossover.
Media reports state that cracks noticed in the walls and floor of the basement of the archive building in early 2007, were investigated then and again in 2008 and one both occasions were considered non-urgent and no cause for concern in such a heavy building and its particular foundation construction and were not monitored for any further movement.
As well as the loss of two innocent victims, the loss of historic papers from the archive building is said to be tragic. Archive reaches back through the medieval period to the year 922 and is valued as a whole at an estimated €400 million. The long, complex and costly road of recovery, investigation and possible prosecution is only just beginning. Meanwhile the new underground tram line is scheduled to open in 2010.
STUVA’s offical journal TUNNEL researched and published a comprehensive supplement on the planning, design and construction to date of the new North-South tram line for the occasion of the STUVA conference in Köln in Nov 2007.
References
Köln - speculation and anger in collapse aftermath - TunnelTalk, March 2009

Gallery

     

Add your comment

Name

Date

Subject