Bridge piers and columns rarely have to resist the impact of a fully-loaded HGV travelling at 110km/h or more. In fact the Highways Agency has records of only 'one or two' maximum impacts during the last 20 years or so.
This is just as well, according to Agency principal structural adviser Martin Lynch. The incident last June, when the central piers of the thaumasite- hit Tredington-Aschurch overbridge on the M50 was struck, involved an unloaded vehicle and an impact speed of less than 65km/h, and damage was not severe (NCE 11 June). But, says Lynch: 'Undoubtedly there are bridges which would collapse if struck.
'A large proportion of piers and columns have less than 30% of the capacity needed to resist HGV impact. And although a lot of bridges could survive the loss of one column without collapsing completely, the potential consequences of a catastrophic collapse are so significant we have to take the problem seriously.'
Main worries centre on heavily trafficked interchanges and rail overbridges, Lynch says. 'If an overbridge carrying a minor local road was demolished, loss of life is likely to be low, and we calculate the debris would be cleared in 24 hours or less. It would be very different with major structures.'
Strengthening bridges against impact began about five years ago, mainly on a 'pragmatic' basis. Now the Agency is engaged on a major risk assessment programme to enable it to target the most vulnerable bridges, and while it does so it has suspended impact strengthening operations on all but obviously high risk structures.
This is due partly to dissatisfaction with current techniques. 'The main method has been adding reinforced concrete jackets around piers,' Lynch reports. 'But this usually looks awful afterwards - and the work causes unacceptable levels of traffic disruption.'
Steel brackets reinforcing the connection between the tops of columns and the decks above have also been tried. A few years ago, however, reports from the US and Japan began to suggest there might be a better, less obtrusive alternative.
In essence, this was the use of various types of high strength fibre - glass, carbon, even aramid - bonded to the surface of concrete elements by long life epoxy resins. Developed after a series of earthquakes in California and Japan brought down an alarming number of concrete highway structures, the new technique appeared to offer many advantages over the traditional alternatives.
Fibre resin composite materials have been around long enough to demonstrate more than adequate durability compared to steel, while being much lighter for the same strength. And structures strengthened by composite materials tend to retain their original appearance. Best of all, the new technique promised significant cost and time savings and massive reductions in traffic disruption.
Bridge wrapping, as the technique became known, differed in one important respect from the unconnected proposal to use pultruded composite materials to strengthen bridge beams. Calculating the beneficial effects of attaching a high strength ribbon of carbon fibre reinforced epoxy to the underside of a prestressed concrete beam is a straightforward exercise in structural analysis. Trying to predict the behaviour of a wrapped concrete column under the influence of the transient horizontal and vertical accelerations of an earthquake is another thing altogether.
The same goes for impact loads. 'We know they will be different from seismic loading, but we don't know how different,' says Lynch. 'We're just beginning an 18 month programme of static and dynamic tests on model piers at the Transport Research Laboratory to check out initial design rules derived from US seismic engineering guidelines.'
These rules were the basis for a key £50,000 full scale trial programme earlier this year. Three companies were invited to demonstrate their products on three 'typically weak' 6m high columns of the Bible Christian overbridge - so-called, apparently, because the road it carries leads to a nearby chapel. Located on the A30 in Cornwall, the bridge was chosen because of the relatively low traffic on the dual carriageway road before the main holiday season. Structural design for the strengthening operation was entrusted to Maunsell Structural Plastics.
Of the three systems one, from Dupont of Switzerland was based on aramid (Kevlar) fibres, one, the US-developed Sika Wrap Hexcel, on glass fibres, the third, from Calfornia's XXsys Technologies used carbon fibres. All fibres came in the form of unidirectional fabrics, with more than 70% of fibres aligned in one direction. They were either sheets or ribbons, with the Kevlar and carbon materials already pre-impregnated with epoxy.
Concrete surfaces were cleaned and repaired in advance. Generally, an epoxy primer coat was applied first, followed in the case of the carbon system by a first coat of resin into which the fabric was rolled then coated with a lower viscosity resin. Up to five more layers of fabric were added before a cosmetic grey polyurethane topcoat was brushed on.
The aramid tape used for the trials was coated with resin before it was offered up to the test column. A total of 12 layers were applied before the finishing coat.
Contractor for the aramid and carbon systems was Balvac Whitley Moran. Makers Industrial applied the Sika Wrap Hex system, and for these trials opted to pre-impregnate the glass fibre cloth immediately before applying it to the column using a specially-developed 'mangle' from the US.
Sika civil engineering marketing manager Richard Barton says: 'This gives a very constant resin thickness and hence a very predictable section modulus.'
After emerging from the mangle, the material was wound on to a detachable roller then wound off again onto the column. After 17 layers had been applied the same cosmetic top coat ensured consistency of appearance as well as providing essential protection against ultra-violet degradation.
Lynch reports that all three systems were applied in much the same time, but draws no particular conclusions from this - 'trials conditions are not completely realistic, and weather conditions varied for each trial'. But he does have some preliminary reactions on ease of use.
'The aramid system seemed to be the easiest to apply, the fibre being very soft and pliable and kind to hands. The glass cloth was also pretty pliable, but I have some reservations over the carbon fibre material.
'It was quite stiff, and I wonder how easy it would be to use on rectangular columns with sharp arrisses.'
Sika's Barton believes the very high cost of aramid systems would rule them out for anything but the most demanding applications. 'Kevlar is used in bullet-proof vests, so it's obviously very good at stopping impacts from bullets.
'But an HGV is no bullet.' He adds: 'The choice between glass and carbon fibre really comes down to balancing the extra cost of the carbon fibre itself against the savings in time against glass fibre.'
Other systems currently on the market in the UK include MBrace from Feb-MBT and Replark from the Sumitomo corporation. Replark, a carbon fibre- based system, is claimed to be the only one to have demonstrated its effectiveness during actual earthquakes - in Japan and Italy.
Two grades of 'pre-preg' unidirectional carbon fibre and unidirectional glass fibre are available as part of the MBrace system. The Highways Agency hopes that manufacturers will offer a wide range of alternative fabric weaves in the longer term, allowing each application to be tailored to the structure concerned.
'At this stage it looks as though this technique will cost less than half as much as the alternatives,' Lynch reports. 'Disruption will be much less. And there will be no obvious change in appearance afterwards.'