Published on 22 January 2013
The safety fences installed in the central reservations of most motorways are a familiar sight to drivers. They are designed to absorb the impact of a 1.5t vehicle travelling at 70mph hitting the fence at an angle of 20◦ and to deflect the vehicle back on to the carriageway. Motorists will also have noted that, at places where there are gaps in the fence to make provision for emergency vehicles to cross at the central reservation, the longitudinal members are angled downwards and anchored to the ground. The change in direction at the downwards ramp of the fence is achieved using a special beam section made by cutting a standard beam in half using a mitred cut and butt welding the halves together at the required 6◦ angle.
The Department of Transport approached TWI after a number of cases of failure of this butt weld. The failures occurred instantaneously as a result of impacts by cars several metres away from the welds. The downstream half of the beam then sprang out in front of the upstream half impaling the car as it slid along the fence, often with appalling consequences.
The problem posed to TWI related to the quality of the butt welds. In early batches, these had been made from one side only and it was decided that these should be replaced urgently. However, this left 20,000 double sided welds in service up and down the country.
Initial studies showed that there was no good correlation between the results of non-destructive tests on the welds and their performance. It was therefore decided that there was little to be gained by a rush programme of inspection. Instead, a statistical approach was adopted. In a period of only two weeks TWI carried out over 50 full-scale tests on welded beams, taken from motorway sites all over the country. The vehicle impact event was simulated in the laboratory using a purpose built test rig.
Based on the results of these tests the performance of the whole population was assessed so that a planned programme of replacement could be conducted based on lane closures when scheduled for other purposes. This was greatly preferable to the alternative of rapid replacement, involving unscheduled lane closures, which themselves could lead to serious accidents. The resultant saving in terms of cost of lane closures (including lost journey times and resulting accidents) was estimated to be about £2M.
For information about TWI's capabilities please contact us.