Train operations on the Red Line of the St Petersburg Metro ceased between Lesnaya and Ploshchad Muzhestva in 1995, when water and soil inflow through the deteriorating tunnel lining became too great to control. The twin stacked tunnels were allowed to flood behind concrete bulkheads and at the time, due to the high cost of repair, it was considered highly unlikely that this section of the railway would ever be reinstated[1&2]. The result was devastating. Not only was this important section of the line lost, but commuters now had to transfer by bus, from the station tunnels deep below the ground, considerably disrupting and lengthening journey times. The metro is an extremely popular form of transport in St Petersburg and over half a million commuters had to be transferred in this manner each day.
By 1997 the City decided that it was necessary to reconnect the Red Line by constructing two new tunnels and that this could only be done with mechanised tunnelling. In 1998 a JV of Impregilo and NCC was awarded the US$126M design and build contract by the client, the Transport Committee of the City of St Petersburg, to complete this difficult task. Lenmetroguiprotrans were the designers of the scheme for the City and Geodata were consultants to the JV.
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
Background
In 1703 Peter the Great founded St Petersburg on the wide floodplain of the Neva River to protect Russian interests in the Baltic. His insistence that the city become the capital of Russia and adopt a European outlook, much evident in its magnificent architecture, led to its rapid development in the 18th Century. Following the revolution of 1917, Russian interest turned to the newly proclaimed capital, Moscow, and it was not until the 1940’s that construction commenced on a metro network deep below the river’s flood plain.
The Second World War disrupted construction and the first section of the Red Line, from Avtovo to Ploshchad Vosstania, was not opened to the public until 15 November 1955. Further construction saw the line extended in phases, the section of the line between Lesnaya and Ploshchad Muzhestva was completed in 1975 and in 1978 the 29.6km line was completed with the extension to Devyatkino.
Construction of tunnels between Lesnaya and Ploshchad Muzhestva stations was problematic. The tunnels lay up to 80m below the ground, passing through a wide buried valley cut deep into the underlying Cambrian claystones. The buried valley is infilled with a combination of fluvio glacial and lacustrine clays, silts, sands and gravels with boulders (T&TI, Jan ’02, p30). In turn, this valley is overlain by glacial moraine and recent river flood plain deposits.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataThe location of weak, fully saturated, material at depth required extensive ground freezing to enable construction. Through the buried valley the tunnels were stacked one above the other to fully utilise and limit the extent of the freeze zone, which was progressively extended in sections up to 30m in length and 86m deep. Both tunnels were then constructed simultaneously, with the lower tunnel maintaining a 15m distance ahead of the upper tunnel. The 6m diameter tunnels were lined with bolted cast iron segments, reinforced in-situ concrete and finally with an inner welded casing, incorporating tubes in the invert to relieve the high hydrostatic pressures on the lining.
Despite the use of ground freezing there were two major inundations during construction. In April 1974, 40,000m³ of unfrozen material broke into the upper tunnel and then the lower tunnel creating a sinkhole at the surface 400m x 200m and 3.5m deep, resulting in considerable damage to buildings and the tunnel lining. Ground freezing using nitrogen was deployed to recover the situation and tunnelling was recommenced from the opposite station at Ploshchad Muzhestva.
In July of the following year, another inrush occurred. First through a sealed injection plug and then the invert of the lower tunnel. The lower tunnel was flooded with water pressurised at 5.6 bar and the upper tunnel filled with compressed air at 4.8 bar. Again nitrogen ground freezing was used to recover the situation and complete the tunnels.
Once operational, the tunnel lining was instrumented and monitored continually, as there were regular inflows of material through the pressure relief tubes installed in the lining. It was noted that the volume of material began to increase in November and December 1994 and during the first nine months of 1995 120m³ of sand was removed from the tunnels and daily water inflows topped 800m³. Incredibly, only once the upper tunnel had settled 300mm (up to 80mm in one month) and the daily sand inflow reached 30m³ was the train service stopped. A further sinkhole developed above the tunnel, this time 300m long x 200m wide and 1m deep.
Investigations showed that the joints in the cast iron tunnel linings had opened and that up to 70% of the in-situ reinforced concrete lining had cracked and spalled due to the effects of hydrostatic pressures, reported to be up to 7 bar. Once the service was suspended, bulkheads were constructed and the tunnels were allowed to flood.
Challenges
Reconnecting the Red Line presented an enormous challenge. It was necessary to connect the existing stations, whose points were essentially fixed, and excavate a tunnel through weak soils at very high hydrostatic pressures. There was also the question of how to deal with the previously disturbed ground, due to the earlier failures, and also to select an acceptable alignment within the operational limits for the railway.
Following lengthy studies, it was decided that the tunnels should be kept as far away from the disturbed ground as practicable. The most acceptable alignment was 250m to 300m to the northeast and at a shallower depth. A new alignment approximately 790m long and 60m deep was selected, avoiding as many sensitive surface structures as possible. The alignment now follows Karbysheva Street, which is bounded by a five-storey school, 15 storey residential blocks and industrial buildings and passes beneath Muzhestva Square.
There remained considerable geological uncertainty and a detailed site investigation programme was commissioned to explore the potential cause of the earlier failures and also to provide detailed information for the design and construction of the new tunnels. The investigation comprised boreholes together with in-situ testing using cone penetration methods through the weak glacial soils. Crosshole seismic tomography was also carried out from observation wells along the route.
A detailed geological model was built up of the stratigraphy, the variation in geotechnical properties and the hydrogeology, both for the design of the tunnel lining and also for the accurate prediction of tunnel face pressures, the control of which would be critical to the success of the project.
The investigations revealed the buried valley to be at least 122m deep and the proposed tunnel alignment to encounter low plasticity clays, silts and fine sands often in lenses or thin beds within the valley. Mixed face conditions would be met where the tunnel passed from the Cambrian claystones into the valley and back again. There also remained the potential for encountering boulders with a UCS up to 250MPa.
Tunnel reconstruction
It was recognised that the safest and most economical way to reconstruct these tunnels was to use a pressurised TBM, technology then neither previously used nor available in Russia. Consequently, tunnel construction was carried out jointly by the Russian main contractor Metrostroi and by Impregilo/NCC JV, who were responsible for the difficult sections of tunnel through the buried valley.
The JV proposed a 7.4m diameter Polyshield (slurry machine) from Voest Alpine, which had been successfully employed on the EOLE in Paris. Modifications were made to the TBM for the greater anticipated hydrostatic pressures, these included a fourth set of brushes in the tail seal, an additional lip seal on the main drive bearing, a new man lock for hyperbaric interventions and new heavy duty disc cutters. The machine was to be launched via a 70m deep access shaft constructed close to Lesnaya Station and drive to a reception chamber at Ploshchad Muzhestva Station where it would be turned around to complete the second drive.
The impermeable and hard Cambrian claystones are favourable for tunnelling and Metrostroi elected to use traditional hand mining methods to construct the shafts, approach tunnels and TBM chambers, which were lined with bolted cast iron segments. The JV proposed a dowelled pre-cast concrete segmental lining that, although common in the West, was something that had not been used in Russia before. Furthermore, as a result of the performance problems of the earlier tunnel lining within the buried valley, very detailed submissions had to be made, which were in turn closely scrutinised by the Russian authorities.
Indeed, detailed numerical and scale model tests were carried out by the Russians to confirm the validity of this concept. The trapezoidal lining comprised five segments and a key 1.4m wide x 350mm thick. The segments were connected using only conex dowels and featured twin gaskets with an integrated hydrophilic seal. For economy and technology transfer, the segments were made in Russia to detailed specifications prepared by Geodata for the JV.
TBM tunnelling commenced in early February 2002 and, following early trials of the slurry system, completed the first drive to Ploschad Muzhestva somewhat later than anticipated, in early May 2003. The return drive benefited greatly from the experience of the first and by contrast took just three months, breaking through on 27 November 2003.
The TBM performed extremely well in the onerous operating conditions and there were no problems with either the slurry pressure system or the tail void grouting system. The tail seal and main bearing seals all performed satisfactorily.
Monitoring
Since the lining technology was new to Russia and since there remained some understandable unease, the precast segmental lining was extensively instrumented. Ten rings in each of the two tunnels were equipped with vibrating wire strain gauges, installed during lining manufacture. The joints between the segments were also precisely measured to observe their behaviour and the effectiveness of the gaskets.
Measurements indicated that the lining remained in compression and that the stresses developed were well within design values. The joints remained closed and the tunnel lining watertight. In addition, a total of eighteen piezometers were installed to monitor pore pressure response due to slurry pressures. Results indicated that the pore pressures increased and then fell back to equilibrium values as the TBM passed. The amount of increase was dependent on, among other things, the proximity of the piezometer to the tunnel and that a response was measured in piezometers installed as far as 50m from the TBM. Deep monitoring points were installed around the tunnel to measure vertical and horizontal displacements. All measured movements were below the design threshold value of 20mm.
A 3.2m diameter major trunk sewer lay some 23m directly above the tunnel, over approximately 140m of the tunnel alignment, and the effects of tunnel construction were closely monitored. Surface settlements were generally between 5mm-10mm, smaller than the 15mm predicted.
Post construction investigations
Again due to the earlier problems experienced with the initial tunnels, extensive investigations were carried out to identify whether any voids had possibly been created behind the tunnel lining and whether the density of the ground had been altered by tunnel construction. Super-broad-band georadar testing was undertaken at fixed points from the surface down to a depth of 90m along Karbysheva Street and in Muzhestva Square.
Results indicated that no voids were present and that there was no groundwater flow and no low-density soil in close proximity to the tunnels. Ultrasonic investigations within the tunnels also revealed that the annulus behind the lining, whilst of varying thickness, was importantly always infilled with grout. Crosshole seismic tomography demonstrated that the velocities of the soils were not altered by tunnel construction and that the natural engineering geological environment had been preserved.
Conclusions
Peter the Great when founding the City turned to Europe and invited the best engineers, architects, shipbuilders, craftsmen and merchants to come to Russia. It seems appropriate that 300 years later the City turned once again to Europe, this time to find a solution to one of the world’s most unique and challenging tunnelling problems.
Related Files
Fig 2 – Longitudinal section through the new alignment
Fig 1 – Plan of the St Petersburg Metro showing the interrupted section of the Red Line, closed since 1995
