The cementing of Pike River

The supply of high performance concrete for the isolated Pike River coal tunnel project had its challenges, says Concrete Society president Chris Munn.

The Pike River Coal Tunnel Project is very unusual in that it is a major civil engineering project located deep inside Paparoa National Park with rigorous environmental restrictions.

Developing the access tunnel to a seam of high quality coking coal located under Department of Conservation land, 50 kilometres inland from Greymouth, has been a two year project that is due to finish this month (December 2008). The export resource is 58 million tonnes, and almost 20 million tonnes of this is recoverable.

The access tunnel is 2.3 kilometres long and inclines at a grade of one in 11, and is 5.5 wide and 4.5 metres high. At the time of Chris Munn’s presentation at this year’s concrete conference, the tunnel project had reached over 2.1 kilometres in length and involved over 5000 cubic metres of special concrete supplied by Allied Concrete from its batching plant at South Beach near Greymouth, some 75 kilometres away.

The environmentally sensitive area and the special concrete mix for lining the tunnel, had its unique challenges, says Chris Munn, who is the national technical manager at Allied Concrete. 

“Concrete was supplied on a 24 hour, seven-day basis using three shifts,” he says. “Site placement procedures required the concrete to be retarded for up to 13 hours prior to use at the tunnel face.”

Because the site is in a mountainous national park, it wasn’t possible to conduct any exploratory drilling along the path of the proposed tunnel route. The coal seam is located 2.3 kilometres from the tunnel portal, and is about 360 metres below the surface at the point of intersection. Geological data had to be gathered by mapping stream beds above the proposed drive, and exposed cliff faces in the area. Investigations indicated that most of the rock was expected to be of high quality with minimal fracture planes, says Munn, and likely to be self supporting, with relatively minor areas (less than 10 percent) requiring supplemental supporting systems. Hence, fairly modest volumes of concrete were expected to be used (between 200 -300 cubic metres). That wasn’t to be.

“Immediately after the first full face blast in mid September 2006, it became apparent that there would need to be a rapid re-evaluation of the tunnelling methodology.”

Because of the poorer rock conditions one to two metres beyond the “outer shell” of the parent rock, the entire project’s allocation of rock bolts and galvanised mesh were consumed in the first three month’s work.

Instead of a largely self-supported tunnel, the rock’s properties and dense fracture patterns in the gneiss were such that large amounts of shotcrete were required for the long term stability of the tunnel. Initial shotcreting was by the dry process, supplied in bulk loads from Allied’s batching plant at Greymouth and applied over weakened areas of rock that was reinforced with galvanised mesh.

“This operation was slow and also required significant volumes of shotcrete to be applied due to over-break.”

It became obvious that hand spraying was not going to be sustainable given the likely volumes involved for the job and the cost, recalls Munn, and the decision was made to turn to robotic wet shotcreting. This involved a new shotcreting machine brought in from Australia by McDonnell Construction, and concrete from Allied’s batch plant that could be carted the one and a half hours to site, then held on site for a period of up to 13 hours at a slump of 180mm.

“This would enable concrete to be used at short notice if the tunneling operations encountered soft rock at any stage during the 24 hour operations.”

The material was also to be fibre reinforced using polyolefin type structural fibres, as used on the Northern Gate Project in Auckland where Allied Concrete was also involved. The concrete was to have a nominal strength of 40 MPa based on cores taken from test panels made on site, or taken from selected locations within the tunnel, rather than based on traditional QA systems.

The remote shotcrete machine arrived in January 2007 and came straight out of the box for initial trials at the Allied concrete plant in Greymouth using existing concrete mix designs.

“In my experience, you always run into strife using equipment straight out of the box.”

The problem was the concrete mix – it had to have a very high slump of 200mm, and it also had a high air content because of the reinforcing plastic fibre, and very fine sand –  which lead to pumping issues.

“The standard spray mix would not go through the robotic pump unit unless the slump of the mix was very high, and it would not start pumping after relatively short stoppages.”

Extensive trials extended over a period of several weeks, involving consultation with Sika’s research laboratory in Zurich, resulted in a final mix design contained significantly higher proportions of aggregate than would normally be used to counter the impact of the relatively fine sand. The mix is made up of concrete, water and eight other ingredients that Allied is keeping to itself.

“We ended up with a well-constituted concrete mix that could be sprayed on the tunnel walls from between five and 10 metres away, and were presented with no problems.”

By the end of September this year, under very trying conditions, over 4000 cubic metres of concrete had been delivered to the remote site and used successfully in the tunnel. “And only one load was rejected,” says Munn. “I think that is pretty good going.”

• Pike River Coal’s capital expenditure budget was $246 million (increased by $23 million) after two years of burrowing through the Paparoa range to reach the coal seams. The additional costs relate mostly to the tunnelling, says the company, and will be partially offset by higher sales revenue than previously forecast. 

• This article is based on a presentation made by Concrete Society president Chris Munn at the recent concrete industry conference.