Repairing the Montblanc Tunnels on the Lleida-Barcelona High Speed LinkThe Camp Magre, Lilla and Puig Cabrer tunnels are situated on subsection lV-b (for the first two) and V on the Madrid-Saragossa-Barcelona-French border high speed link, section Lleida-Martorell, in Montblanc and La Riba municipalities in Tarragona province.The Camp Magre and Lilla tunnel alignments crosses Tertiary formations (Eocene-Oligocene), consisting of limestone, marls and reddish gypsum-bearing argillites of variable strength whereas the Puig Cabrer tunnel alignment crosses Triassic formations (Mucheskalk) and has been driven in materials predominantly comprising of limestone, dolomites, marls and argillites of various colours.The three twin-track Montblanc tunnels have a total length of 3,753 metres, divided as follows: 984 m (Camp Magre), 704 m (Puig Cabrer) and 2,065 m (Lilla). The cross sections are 110 sq m (Camp Magre), 113 sq m (Puig Cabrer) and 90 sq m (Lilla). The outer diameters are 14.52 m (Camp Magre), 15.52 m (Puig Cabrer) and 12.92/13.52 m (Lilla) and the inner diameters are 13 m (Camp Magre), 14 m (Puig Cabrer) and 12 m (Lilla).The three tunnels have been affected by swelling of the subsoil that lifted the floor of the tunnels and caused cave-ins. The subsoil lifted as much as 80 cm due to the presence of anhydrites, a material derived from clays, characterised by a high expansion power when in contact with humidity. The strength with which this material pushed the ground equalled 450 tonnes per sq m. Read E-News Weekly 47/2003 & 11/2003. On the sides of the Camp Magre tunnel, pressure provoked cracks in the floor. The result was that the distance separating the sidewalks built at each side of the tunnel was uneven. There was a difference of 60 cm at some places and 20 cm at other places. The floor was not uniform and bad ground undulation had been detected. Inside the tunnel, pressure was such that the trackbed divided in two parts not fitting together in the middle. Repairs consisted in demolishing the concrete bed to dig up to a depth of 4.5 metres to create a circumference. The lower part has been filled with concrete and made waterproof to avoid that even a water drop penetrate. The upper part, or vault, of the tunnel has been reinforced by building an additional concrete wall cast on thick welded wire fabric (10 tonnes per metre of tunnel). Adjusting the welded wire fabric to reinforce the Montblanc tunnels Three different types of sensors have been installed to monitor systematically any movement of the subsoil. These sensors were extensometers in the floor and in the welded mesh and hydraulic units to measure the ground pressure.To know the strength and deformability parameters and establish thereby the ground behaviour, the RMR of the excavated grounds has been determined during tunnelling. The values were variable both in the Lilla and Camp Magre tunnels, with grounds showing low RMRs (20) corresponding to clays in the areas close to the tunnel entrances or RMR measurements of 45-50 in more compact limestones and argillites. In the Puig Cabrer tunnel, the measured RMRs exceeded those observed in the other two tunnels, varying from 75 in limestone to 45 in marly clays. Excavation was carried out using the drill/blast method due to the hardness of the ground, in top heading / bench sequence as per NATM. The typical cladding consisted of shotcrete with Degussa's alkali-free MEYCO SA 172 and bolts, while steel arches were limited to some sections where the ground quality was worse. Visit www.degussa-cc.esTo be able to repair the tunnels, the strategy demanded four concreting plants installed by Readymix Asland (Hormitaver), with four Degussa Construction Chemicals deposits including the POZZOLITH 390N , MEYCO TCC 791 and GLENIUM T-802 admixtures at each plant. Construction took place in two well separated phases: excavation of the deep invert and construction of the vault cladding. Visit www.degussa.com and www.readymixasland.esThe start situation prior to excavating the invert in the three tunnels was a flat reinforced concrete floor. The first necessary step was then to demolish the concrete bed to allow crews to excavate the deep invert. Before the invert excavation, it was necessary to secure the existing supports and cladding to make sure they would not be affected by construction of the tunnel invert. Therefore, the supports have been anchored to the ground by means of bolts installed into the walls.Once the supports had been safely secured, excavation of the invert commenced, always using a jack hammer or a road header despite the hard ground, in order to avoid further fracturing in the massif. Excavation of a circular section was achieved in stages, with the objective of minimizing alterations to the rock in contact with the invert. During the first stage, the ground was digged and mucked away so as to shape a kind of trough. In this manner, it was possible to do part of the works without discovering the materials that form the bottom of the final excavated profile. Of course, drying up and subsequent alterations of the soil were avoided and changes in the strain condition were minimized.The first phase was followed, with minimum delay, by the refining tasks to excavate the ground until its ultimate rounded-shaped section. Workers used jack hammers or road headers, attempting to shape the adecuate profile with minimum adverse effects. Immediately after the refining phase, the excavated portion was supported spraying fibre-reinforced concrete. Spraying this first layer had two objectives: avoiding changes of the rock mass humidity, that could have caused rock alterations, and supporting the tunnel invert just excavated.Concreting of the invert section was undertaken in two phases. During the first phase, concreting of the bottom of the invert was achieved in sequences of 24 linear metres. After hardening of this layer, concreting of the sidewalls took place by means of tunnel formwork units, in cycles of 12 metres. This phase concluded the concreting of the invert, the remaining task being the pouring of concrete in the trough to fill it with conventional HM-15 concrete up to the level of the trackbed.The steel reinforcements have been placed on the vault in two ways, according to the specific circumstances encountered in each tunnel. With the first method, full section reinforcements were prefabricated outside in 2.4 m-long pieces, that were introduced afterwards in the tunnel by means of a special machinery. The second method consisted in placing the steel reinforcements directly in the tunnel by means of a carrier specially designed for that purpose. Concreting of the vault necessitated the use of formwork units and reinforced formworks, of 12 m of length, able to resist to the pressure that such a fluid concrete produces, which is much greater than the usual thrust exerted by tunnel lining concretes. When concreting of the vault was completed, crews installed the culverts, rails, earth plugs, track crossings and all other components of the track base.The Polytechnic University of Catalonia (UPC) designed special specifications for self-compacting concrete, stipulating the characteristics of tests and control values, as well as the required constructional prescriptions.To achieve the required fluidity and fast setting, it was necessary to add admixtures like GLENIUM T-802 and GLENIUM TC 1325 and incorporate also silica fumes, needed to attain strengths higher than 80 N/sq mm. The action of silica fumes is to provide a better resistance due to lower porosity and puzzolanic action and to supply the required filler to achieve a self-compacting concrete and consequently reduce permeability. Dosing of the silica fumes equals to 10% of the cement weight.The admixtures that Degussa Construction Chemicals incorporated to the concrete were GLENIUM T-802, POZZOLITH 390 N and MEYCO TCC 791. In a second stage, GLENIUM T-802 was replaced by GLENIUM TC 1325 in the Lilla tunnel.The Getinsa-Geocontrol JV was the construction manager during the strenghtening works of the tunnels and also carried out a series of studies on the concrete behaviour to analyze the correlation between the different variables according to dosing. Not only the goal was to know how the water-cement relation and the superplastifier percentage and rate affected the concrete properties but also to study and propose a new dosing.Eventually, repairs of the tunnels have costed EUR70.7 million (47.8 millions for the Lilla tunnel and 22.8 millions for the Camp Magre and Puig Cabrer tunnels). Click es/20. 30/06.