In mid-2006, having been awarded its first road tunnel contract in Iceland, Metrostav and its local joint venture partner Hafell began to mobilise for the construction work that would call for two independent sites and offices, one for each of the separate tubes to be driven in the remote north of the country and the associated link roads.

Hedinsfjordur project would see two tunnels constructed – Siglufjordur (‘Siglu’) Tunnel and Olafsfjordur (‘Olafs’) Tunnel, each named after the nearest towns. Under the scheme developed by the Icelandic Road Administration (Vegagerdin), the towns would be directly linked once the tunnels were driven through the two mountain ranges that separated the communities.

While mild weather with long hours of daylight would be enjoyed in summer months the island, being almost on the Arctic Circle, would see harsh winters bring heavy snows, strong winds and prolonged darkness. Having such extremes of weather and also the natural barrier of the mountains and valleys there had to be two different construction site offices and depots.

In effect, it would be like two separate, but related, projects until excavation was completed and the next phases of the works took over, such as installation of waterproof and frost protection (WFP), drainage, services, and the tarmac road surface.

One site was chosen to be located not too far from the portal site near Siglufjordur, the other close to the portal site at Olafsfjordur. The plan was to drive the tunnels from each portal and it was anticipated – and turned out to be the case – that the shorter, 3650m long, mined section of Siglu tunnel would hole through first.

Then, in Hedinsfjordur valley, the activity would be joined by a carefully landed from the sea construction plant for the crossvalley earthworks that would provide the route up to the new portal opening on the opposite mountai range. Then, Siglu’s tunnellers would commence blasting at Olafs, driving from the opposite direction to that already underway.

Starting at Siglu
Blasting of Siglu tunnel began in September 2006 at Chainage 2.400km along the new route from the existing road near Siglufjordur. The open cut had been made behind where, at Chainage 2.275m, would later be built a 125m long concrete portal. The drive was to advance on an uphill Slope of 1 per cent for approximately 1.6km to crest, and then the excavation progress down on a 3 per cent slope for approximately 2km, to breakthrough at Hedinsfjordur valley.

Excavation started well. The drive passed a geothermal area that Siglufjordur draws hot water from without causing any known disruption to the flow.

Geology along the alignment comprised olivine basalt for the majority with some porphyritic basalt towards Hedinsfjordur and the final, downhill stages of the excavation. There were regular, and some large, sub-vertical dykes along the majority of the alignment, and faults.

Both the dykes and the faults were found to be more concentrated after about 700m into the drive. The faults were particularly intense around chainage 3.3km, where there were longitudinal failures and the first groundwater inflows of note began. Beyond that zone the flows stopped, arising only along relatively short sections at further faults at the middle of the tunnel.

Three-quarters along, though, there were some more longitudinal failures of the rock, again caused by sub-horizontal faults and more inflows began. It was approaching this portion of the drive that the first grouting was needed – in three sections, apart from some needed closer to the end. Overall, Siglu had proved to be a reasonable tunnel to drive by the time Metrostav holed through to Hedinsfjordur in March 2008, only to catch a surreal sight: fromunderground the tunnellers were looking out at a green play of faint, wavy bands of light in the sky – the Northern Lights.

‘Pretty good’ tunnelling
The relative ease of tunnelling in Siglu gave no hint of the massive geological challenges that would be met in Olafs. However, Iceland is known for its radical variability in geology, some road tunnels have large groundwater problems while others let tunnellers drive forward, possibly with some wetness or stability issues. Difficult to predict, although surface features such as dykes slanting up from the depths can give some potential insights, though nothing definite.

While Olafs would add its name to the annals of Icelandic tunnelling difficulties, it was Siglu – which was ‘relatively pretty good’, says Ermin Stehlik, project manager with Metrostav – that was to be the better run underground, giving no warning of what lay deep in the hillside of the other valley. Even what groundwater there was in Siglu, at between 10 and 20 degrees C, wasn’t as cold as that to be encountered in highpressure jets in Olafs.

Yet, it was fortunate, too, that as the face advanced ever closer to Hedinsfjordur and to the mountains that were to be blasted through for Olafs, there was not a general rise in significant groundwater problems; not something to be easily faced in a downhill drive. But there was some inflow, which was dealt with by a system of provisional sumps with automatic pumps.

Two grouting solutions were used to stem water inflows. A total of 460 tonnes of cement grout was used for pre-grouting on most inflows using Atlas Copco’s UniGrout E 45 two x 100 pump. In rare cases, when the water remained close to the face despite the regular probe hole drilling, polyurethane (PU) resin pre-grouting had to be used. In total, the downhill stretch called for 40 tonnes of PU chemical grout.

In a few instances where there were large inflows of groundwater from probe holes the situation did become critical, requiring the Sandvik Axera T11-315TCAD drill rig to be removed from the face. Overall, these were exceptional. The advance rates achieved in these occasional wet conditions were admirable, says Metrostav.

The biggest problem encountered in the downhill section were caused by water shooting out of fissures after blasting and at times excavation had to be suspended and the fissures sealed. It was a difficult operation using PU-soaked rags driven into the fissures by wooden wedges or bunched inside ‘cushions’ locked into the rock by bolted steel sheet plates over wider openings.

When fissures were successfully sealed, a borehole would be drilled close to the source of the now-blocked flow, then a hydraulically expanded packer would be inserted. The packer would help to fill the fissure in the face area with a PU material, waterproofing the rock wall to enable longer boreholes to be drilled farther ahead for cement pre-grouting. The approach prevented free-flow and loss of the cement suspension.

Tunnelling
The excavation sequence usually involved drilling two 51mm diameter probe holes in the range of 25-32m beyond the face, ensuring a 6m overlap between each investigation into the rock ahead. Based on volume, pressure and temperature of the groundwater in the probe holes there would be a joint decision by Metrostav and the client’s technical and contract supervision services engineer, from GeoTek – ‘The Supervision’ – on whether grouting would be required before proceeding.

A Sandvik Axera T11- 315TCAD drill rig was used for both drilling the probe holes as well as the face pattern of 48mm diameter blast holes. TCAD is a semiautomatic system using a software program to measure the position and the direction of a drill bit using laser to correlate the drilling pattern and boom positions with the tunnel alignment coordinates. The boom is then moved by the operator to ensure the drill pattern remains accurate.

In both Siglu and Olafs the rounds were usually drilled to the full length of 5.27m, giving a pull of between 4.7-5m. The round lengths were reduced to 3m when more difficult ground conditions were met.

Blast holes were usually packed with Titan 7000 blast emulsion – now renamed Civec – supplied by Orica Mining Services. A Mini Site Sensitized Emulsion (SSE) pumping unit completed the charging after being transported to the tunnel face by a small truck. The Mini SSE system meant the emulsion could be pumped into two holes at the same time. It also allowed the operator to change the charge according to the type of blast hole.

The emulsion proved effective for blasting in hard and brittle rock but when more porous patches were met the crew used classic explosives. In some cases a combination of both types of explosives was employed. The 25g Nobel Prime detonator amplifiers were chosen as boosters and inserted into the blast holes together with non-electric detonators during the emulsion pumping.

Mucking out was done by Komatsu dumpers, mostly with 35 tonne capacity, which were rented from local company Kraftvelar. Spoil was removed to temporary and permanent dumping sites located close to the portals, while suitable material was put in a crusher for rock fill to await the later road works. The dumpers were loaded by Broyt C600Ws with 3.4m3 shovels. As a back-up, Volvo 180E wheel loaders were available onsite.

Successful completion
The drill and blast work in Siglu progressed with relatively few difficulties. Like on lafs, the site personnel and crews – those in charge, the foremen, the miners were all Czech and lived in the adjacent town, Siglufjordur. They worked two 12-hour shifts on a six-day week. Catering was provided on site The week ran from Monday morning to Sunday morning.

In almost 20 months of tunnelling they had achieved an average monthly advance rate of 200m. The record month on Siglu was 302m.


Siglufjordur (‘Siglu’) Tunnel proved to be the easier of the two tunnels to construct on Hedinsfjordur road project in north Iceland Excavation of Siglu began in September 2006 at the portal near the town of Siglufjordur The best monthly advance on the 3,650m long, mined section of Siglu was 302m The average progress rate on Siglu in 20 calendar months was 200m, allowing for planned stoppages There were some groundwater problems for Siglu but nothing like on its sister project, Olafsfjordur (‘Olafs’) Tunnel At times extra measures were required to stem ingress, such as plates to lock in chemical resin soaked rags in fissures Loading and mucking out worked well A well earned celebration at Siglu on holing through, in March 2008, to Hedinsfjordur valley between the tunnels under construction