Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Trial and error

Northern Ireland - Four experimental prestressed concrete reservoirs have required emergency surgery to avert failure.

New technology brings with it risks, and in civil engineering testing of innovations is sometimes only finally achieved in the field.

An example of non-traditional engineering which has literally come undone can be found in prestressed concrete water reservoirs in Northern Ireland.

A nightmare scenario unfolded during routine inspections of the tanks at Conlig near Bangor and Ballycullen near Newtownards in North Down, owned by the Northern Ireland Department of Regional Development's Water Service.

'The tanks range in diameter from 30m to 48m. The ones in Northern Ireland were built in the 1970s and it seems problems were evident from the 1980s. However in detailed inspections in April 2002 it was found that prestressing strands had failed on the back of two of the reservoirs, ' says Joe McClintock, Northern Ireland area manager of Hyder Consulting, consultant for the works.

Although the tanks had not failed, there could be little engineering reassurance. And structural failure, if it were to occur, could be sudden.

'There was a health and safety risk from structural failure. But there was also a public health risk from the ingress of potential pollutants into the water, so there was a dual edge to the work, ' says Northern Ireland Water Service water procurement manager David Saville.

'The tanks have a storage capacity of 40M litres, and supply around 15,000 people in the Newtownards and Bangor areas. We had to reduce the storage capacity by half emptying the reservoirs to relieve the stresses on the walls because of concerns about stresses in the bottom strands.

While we would normally have 2.75 days' storage, that went down to 1.25 days, which is cutting things a bit fine, ' he adds.

The failures in the two sister tanks were at different locations - one in the top strand recess, the other in the bottom. Rather than demolish the tanks, a decision was taken to provide what is literally a belt and braces approach: New prestressing strands were to be installed like new girth restraint around the tanks while the original strands, suspect to corrosion, were to be removed.

All four tanks in the Water Service's North Down area are being treated in a £1.1M, 20 week contract to be undertaken by Graham Structural Repairs, a subsidiary of Northern Ireland contractor John Graham.

'There were no records of what the strand stresses were in the original construction, so we had to use our engineering judgement. During construction, the strands were wound round and round the tanks: there were no plastic sheaths or ducts, the strands were just placed against the concrete, ' says McClintock.

'We have some basic information from the original design drawings and specification, but have had to make estimates ourselves about the structural fabric, ' he adds.

The strength of the concrete in the domed roof was estimated at 20N, for example. 'We couldn't take cores because the dome is too thin, tapering to 80mm 5m away from the haunches which are 200mm thick, ' explains McClintock.

Any dramatic alteration in the stress regime of the tanks could risk catastrophic failure, so the replacement method was more akin to delicate surgery or needlework than civil engineering. But working out the appropriate stresses to be applied to the tanks, and estimating those already being applied was no easy task.

Forensic engineering gave some information but had to be allied to an observational approach in the work.

'We had to ensure that the work would have no possible adverse affect on the structure, but if there was any risk, then this would be flagged up immediately, ' says Graham Structural Repairs contracts manager Gareth Doyle.

The method devised was to install temporary strands above and below each existing strand, which is then destressed and removed. The tanks were also propped with scaffolding at points around the circumference.

New strands are then installed, and stressed with a jacking system by subcontractor Balvac. They are sheathed in PTFE plastic for corrosion resistance, before the tank walls are painted in gunite for further durability and protection.

Strain measurement was used to gauge the behaviour of the structure in the different loading conditions.

'Releasing or applying stresses means you get alterations in strain. We did one band at a time: We measured the strain in the concrete when the temporary strands were installed, when the old strands were removed, when the new strands were installed and when the temporary strands were then removed, ' says Doyle.

Constant monitoring would show if there was any significantly worrying alteration to the stress environment of the structure.

'The temporary had to reflect the permanent state, ' he adds.

The stressing forces were significant. 'The highest are at the top band level where the strands also have to take the loading springing from the roof.

The loads then decrease for the lower bands, with the exception at the bottom, ' says Doyle.

'Jacking loads were 5,450kN for the biggest 48m diameter tank at the top band, ' he explains.

The levels in the dome were also continually monitored, with changes a sensitive indicator of movement in the walls or changes in the stress environment of the structure.

'We predicted when we would apply the temporary strands and their prestressing force that the dome would rise at the centre by 14mm, and it did. But when we installed the new permanent strands, the dome didn't relax as much. The design stress in the new was probably greater than that in the old strands, probably due to slight relaxation over time, ' says McClintock.

While the original strands were wound completely around the tanks, the new strands were wound in semi-circumferences, with anchor points on opposite sides. This is more efficient in distributing the prestressing forces and overcoming friction forces generated by the walls of the tanks in prestressing.

The refurbishment project is one of which the team is proud, making the best use of the original engineering, which while lacking in robustness certainly had elegance in theory. 'It's the best environmental solution - we are in effect recycling the structure and minimising the structural intervention, ' says McClintock.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.