Shani Wallis, TunnelTalk

As part of an on-going programme to improve national and international railway links for the year 2000 and beyond, Austria embarked on excavation of a 9.8km-long pilot tunnel ahead of full construction of the planned 22km-long Semmering base line tunnel through the Alps. The new tunnel is on the domestic route between Vienna and Villach, which is on the main Trans-European railway route between the states of middle and eastern Europe and the Mediterranean harbours in Italy. The new alignment will supplement the existing 41km-long route, which was built more than 100 years ago and winds slowly and steeply up and over the Semmering Pass. At the lower elevation the new tunnel will allow for higher train speeds, ensure continued services through severe weather conditions and reduce travel times substantially. When complete, the new 'fast' track will carry high-speed passenger services and heavy freight trains while the existing mountain pass railway will continue as a local community service and as a tourist attraction through the spectacular Alpine landscape.

Mürzzuschlag portal from which drill+blast advanced

The 22km-long base line tunnel through Semmering was designed as a 72m² single tube, twin track tunnel. The 9.8km long x 16m² pilot is being excavated about 30m to 40m parallel to the main tunnel. With a cross passage every 1.5km, it will serve as an emergency escape for the southern part of the main tunnel where construction of intermediate escape adits is impossible.

Work on the pilot began in August 1994 when Austrian contractor Porr was awarded the five-year contract for a tender value of Austrian Shillings 500 million. By late December 1995, Porr had completed 1,100m using drill+blast and working on a 24h/day (three 8h shifts/day), 7 day/week production schedule. Immediate support and lining was designed by Austrian consultant Geoconsult on NATM principals and comprises shotcrete, mesh, arches, and rockbolts. The face progresses on a full-face operation using a two-boom, track-mounted Atlas Copco jumbo. A track-mounted Schaeff ITC excavator mucked out the spoil and is also used to excavate very weak rock. The conveyor of the Schaeff lifts the muck onto a 300m-long transfer conveyor held on brackets in the crown. This loads trains of 12m³ Mühlhäuser Rota-dump muck cars, which are pulled by diesel-drive Schöma locos.

The equipment inventory also includes two concreting trains. The first carries the dry mix shotcrete material while the second is equipped to place wet mix concrete for the cast in-situ invert slab. Tunnelling started with a precast concrete invert segment but this has since changed to a more efficient in-situ concreting operation cast in 7m lengths. Both concrete trains are based on Mühlhäuser equipment with the wet and dry mix concrete being transported into the tunnel in Mühlhäuser remixer cars. The dry mix shotcrete is applied with an Aliva Duplo shotcreting machine while the wet mix concrete is placed using an Italian concrete pump.

Geology
The geology at Semmering is highly tectonically disturbed and comprises three principal rock types of phylite, limestone/marble and quartz. Water ingress, although not usually under pressure, has been and is predicted to be high in the frequent fault zones and large karst features are common in the limestone. Probing ahead, with a minimum 40m long probe and a minimum 10m overlap, is a contract specification which increases to a 100m long probe in zones of very poor rock. "Rock conditions are particularly bad at fault interfaces," explained Dr Josef Kaiser, the project's geologist for Eisenbahn-Hochleistungsstrecken AG (HL-AG), the client organisation established by the Austrian government to implement new railway improvement projects. "We are currently in quite a disturbed zone of faulted rock and although the cover is only about 150m, the deformation is high with up to 4cm on the vertical and 17.5cm on the horizontal," said Kaiser. "We have been in better rock and the long length of the tunnel in the phylite is expected to be quite favourable for tunnelling. There is however another bad fault zone ahead of us in which water ingress is expected to be as high as 4-5 litres/sec and deformation is also expected to be high."

A combination of Swellex rockbolts of 1.5-2m long and capable of withstanding 100kN pull tests and fully grouted SN rebar bolts of up to 4m long and tensioned to 250kN are the principal anchors used in the immediate support regime. The Atlas Copco boomer drills the bolt holes and the SN anchors are tensioned after 4h rather than the more usual 24h.

Plan of the Semmering main tunnel and its parallel pilot tunnel

When TunnelTalk visited the project in late December 1995, the trip into the face coincided with a rockbolting operation in a very poor rock zone. The face, which had been shotcreted and rockbolted, was dry, but the high deformation was evident in the flaking shotcrete and the squeezing off of the rockbolt face plates. Measuring stations are placed at 20m to 50m intervals and measurements are generally taken on a daily basis. However, in zones of high deformation instrumentation stations are installed every 10m and readings are taken twice a day for the first 2-3 days, once before and again after the blast.

"Although deformation in this particular zone is high, at 40mm on the vertical 175 mm on the horizontal, the movement generally ceases by the fifth day," explained Josef Schöggl, Project Engineer for HL-AG. "As the face progresses, instrumentation readings are reduced to weekly and monthly once the face has advanced more 500m ahead," he said.

In this highly active zone, the tunnelling crew was excavating short 1m rounds of the full face, including the invert, and installing full round expanding arches on 1m centres. These arches had four 20cm expansion joints in their perimeter, which allow the rock to deform without buckling the arches. Shotcrete in poor rock zones is up to 15cm thick and spiling and forepoling was used also to presupport the face in bad ground. In better rock, the Atlas Copco jumbo drills up to 80 charge holes of 1.3m to 2.5m-long in the 16m² face and uses about 1kg/m³ to achieve a near 100% pull.

At the time, the plan was to leave the pilot tunnel lined with its primary shotcrete lining. The dry mix shotcrete applied by the Aliva Duplo had a maximum 8mm aggregate and a minimum 380kg of cement/m³. With the appropriate accelerators and additives this shotcrete has exhibited core test strengths of up to 40N/mm² after 30 days and up to 1.5N/mm² after 12h.

The contract is based on the new ÖNORM B2203 system in Austria

Pilot experience
As well as providing a necessary emergency escape route for the adjacent main tunnel, the 9.8km-long pilot serves three other vital functions. By actually passing through the various rock types, it not only provides valuable geological information for the large tunnel project, but also provides information on which actual time and costs of the large tunnel operation can more accurately be estimated. Where the pilot tunnel has a tender value of AST500 million, the 22km long single tube double track tunnel of the main project is estimated at some AST6.1 milliard (thousand million or billion) in January 1993 figures. Comments at the time were that, the more accurate the time and cost estimates for the larger tunnel, the more likely it is to receive government funds. The 22km-long main tunnel is expected to take about eight years to construct and was planned for a 2005 opening date. To meet this schedule, detailed design would have to start in 1996 with construction starting within two years of that. Funding for the main tunnel however had not yet been approved. The 9.8km-long pilot tunnel was expected to be complete by the end of 1999.

The pilot tunnel contract was tendered for and progressed on a civil contract based on the new ÖNORM B2203 system of building tunnels in Austria which was introduced officially in October 1994. Although still based on rock classes and appropriate categories of support, this is quite a different system from the usual in that the rock classes, as identified by the geological investigations, and the categories of support for each, with their corresponding cost, are designed to be flexible rather than rigid. The quality of the rock at the face is agreed between the contractor and the client's representatives and the elements required to safely support and line the rock are selected, also following mutual agreement. The method of payment is based on tendered rates of advance in the predicted rock conditions and on a tendered schedule of rates for each cubic metre excavated in the rock quality categories. Support is paid for on a unit price per unit installed.

ALIVA Duplo dry mix shotcreting machine.

For example, at Semmering, Porr was said to have based its tender on advance rates of up to 14m/24hr day in the best quality rock (1-1, 1-2) to as little as 4m/day for the worst expected conditions (9.7). Against such estimated advance rates, the contractor entered a cost/m³ to excavate the various rock types. Without having seen this type of confidential information for the Semmering pilot, it was suggested that the rate for a similar sized tunnel in similar types of rock might range from about AST550/m³ for the best quality rock (1-1), to AST1,200/m³ or more for the worst situation (9-7) in very poor rock. Although it was still too early to judge the effectiveness of the new ÖNORM B2203 form of contract, it was said that it has the potential to rule out the difficulties experienced under the former more rigid, method of contract management which controlled the moving from one rock class/support category to the next. Early experience with the new form was said to be positive but always with room for disagreements and differences of opinion.

As part of the pilot tunnel exercise, Project Geologist Dr Kaiser keeps a log of geological face maps as well as records of the water ingress encountered, the rates of advance and the support elements installed. "Through this data, we will be better prepared to estimate the cost and construction time for the larger diameter tunnel," he explained. "A TBM operation for this Semmering pilot was considered but was discounted because of the high deformations and the bad quality rock conditions anticipated in the fault zones. A TBM might well achieve up to 40m to 50m/day in the phylite, but the risks of squeezing ground and of high water ingress potential were considered too high for current TBM technology for the pilot."

Although on a slight uphill gradient of 2% over the first 750m and 1% thereafter, the pilot tunnel is being fitted with a comprehensive pumping system with a capacity of 150litres/sec over a long distance and of up to 300litres/sec on shorter lengths. The pilot tunnel contract also includes the raising of two ventilation shafts of some 2.4m in diameter and about 270m deep. These are located at about chainage 3,500m and 9,000m in the 9.8km long pilot.

The portals for the pilot and for the adjacent main line tunnel lie alongside the tracks of the existing railway. To avoid interrupting services, all work for the pilot is advancing from a workstation excavated at the start of the pilot tunnel. This work area not only houses most of the plant and materials needed for the job, it also leads to an underground chamber which has been excavated specifically for the safe and secure storage of the excavation explosives. The 60m² x 230m-long enlargement at the start of the pilot will also provide a working station for the main tunnel advance from this same portal location. To respect the local environment, it was suggested that the main tunnel should advance from the pilot access within the mountain with main tunnel portal being created from the inside out and as one of the last pieces of tunnel to be excavated.

As might be expected, news on when the main Semmering tunnel project might advance to the detailed design and construction phases was being awaited eagerly. However, Austria, as a new member of the European Union, is feeling the squeeze on government spending potential as it tries, like several other countries, to meet the criteria of the Maastricht Treaty for a single currency for EU countries. As a government funded project, the Semmering main line tunnel may have to wait several years longer than many might have hoped.

Planning for the Tunnel Tyrol beneath the Brenner Pass
One of the most ambitious ideas for improved rail transportation on the Continent for the 21^st Century is the Tyrol Tunnel project which promotes the construction of more than 500km of tunnel or some 83% of a 350km-long twin-tube base-line route beneath the Brenner Pass in the Alps linking Rosenheim in Germany to Verona in Italy and with intermediate interconnections to existing rail networks at Wörgl, Austria and Bolzano, Italy. The feasibility of the plan is based on using the system for freight trains only and with trains that are totally automatically operated, the basic principle being to provide totally separate passenger and freight rail services since the requirements for each are so substantially different.

Schematic illustration of the Brenner Baseline Tunnel

High speed for freight traffic is not a priority nor is ventilation, providing the trains are automatic and un-manned. Also, where long periods of time spent travelling through long tunnels would be unacceptable for passenger trains, it is not a concern for automatic freight trains.

Passenger rail networks, which require high standards of ventilation and public safety facilities in long tunnels, already exist, and, with some upgrading, high speeds of 25km/hr could be reduced to an acceptable minimum of 60km/hr in difficult alpine terrain. An improved alignment for passenger traffic through the Tyrol proposes a route of about 424km of which about 5% would be in tunnel, giving passengers the high speeds they demand as well as the opportunity to admire the fantastic scenery of the Alps.

Promoted by the International Planning Group (IPG), which includes Austrian geotechnical consulting firm Geoconsult; D2 Consult (formerly Mayreder Consult) of Linz, responsible for tunnel design; Tunnel AG of Vienna, acting as project co-ordinator with responsibilities for cost analysis; Intrsys of Munich, responsible for traction technology; and Robbins of Seattle in the USA, studying necessary TBM technology development; the long tunnel plan is designed to protect the environment and the quality of life in the alpine regions and resist pressure to increase road traffic facilities.

More than 30,000 cars, 4,000 trucks and 130 trains per day are said to use the Brenner corridor through the Tyrol with truck traffic peaking at 6,000/day on occasions. By 2010, these figures are expected to double. The separate passenger and freight networks, it is suggested, is more efficient, more cost effective, less limiting in capacity, and more appropriate to the specific needs of both than the current mixed service networks.

In his paper and presentation to the series of lectures on 'TBM tunnelling trends' in Linz on 14 and 15 December 1995, Professor Johann Golser of Montan University of Leoben and a Director of Geoconsult, suggested that the 500km of single-track, twin-tube tunnels for the base-line rail freight proposal would have a 5.55m internal diameter and would be excavated in about five years using more than 20 TBMs working simultaneously and achieving an average progress rate of 700m/month. The proposed alignment allows for more than 30 tunnel headings from 15 access points with the longest single drive being about 25km long.

To maintain a consistent average of 700m/month/machine, the TBMs would have to be designed for continuous mining and with the lining or support installed immediately behind the advancing cutterheads. The TBMs would also need to be designed to cope with the prospect of high deformations and squeezing rock conditions particularly under the high overburdens of 1,200m under the northern and southern Alps and up to 2,400m in the central Alps.

In geological terms, the alignment is predicted to pass through:
• 2-5% in stable to slightly friable rock in which the rock mass behaves elastically and in which minor local support would be required;
• about 70% in friable rock in which the failure phenomena are shallow and equilibrium of movement would be reached relatively quickly;
• about 10% in a category of rock which includes Quaternary valley fills and soft rocks which are under pressure exerted by low cover and in which failure mechanisms may reach more deeply into the rock; and
• about 10-15% in rocks under pressure exerted by high cover which would include greywackes and phyllites in the Kitzbühler Alps and in the mica schists and possibly the gneisses which are under the highest overburden.

Radial deformation due to high primary stresses is expected to be in the range of 10cm-20cm, and up to 30cm and more in extreme situations. Under such conditions, the TBMs would be designed to cut variable diameters with an overcut capacity of up to 30-40cm on diameter. Open TBMs with a single-shell high-strength shotcrete lining in conjunction with systematic rockbolting and with special deformable elements to provide longitudinal deformation slits in the shotcrete shell are suggested as best suited for highly squeezing rock.

Elsewhere, double shielded TBMs erecting a final single-shell precast concrete segmental lining would be expected to provide the highest advance rates. In addition to the overcutting facility, the shield of the TBM would also need to be variable in diameter to prevent becoming trapped in potentially squeezing ground. Where the segmental lining would accommodate radial deformation in the rock of up to 5cm, depending on the joint detail and the annular backfillng grout, special yielding elements in longitudinal joints of the segmental lining would be required to cope with high deformations or squeezing ground.

Probe drilling and dewatering holes drilled parallel to the continuous excavation process would be additional requirements for the excavation operations and continuous conveyor belts could provide advantages over conventional muck trains for hauling the many millions of cubic metres of rock to create the tunnel tubes.

Although still in the early stages of development, the idea of very long, freight-only rail tunnels, using fully automatically operated trains, is an exciting proposal and one which could well provide a more reliable, more environmentally friendly, safer and most cost effective alternative to trucks for transporting freight across the great natural barriers which still divide an ever more commercially integrated Europe. Funding remains a constant limiting factor and proposals for a private enterprise venture along the lines of Eurotunnel were being examined.

References

Brenner Base Tunnel - let the works begin! - TunnelTalk, April 2011
Design considerations on approach to Brenner Baseline high-speed rail connection in Austria - TunnelTalk, February 2008

Search

TunnelTalk Articles TunnelCast Videos Archive Feedback Current Diary Dates World News Briefs Career Moves Accolades & Tributes Stay Connected Procurement Job Opportunities Equipment Sales DIRECTORY Marketplace About Us Advertising Contact Us Legal Notices Home Office     Latest Alert Alert Signup Alert