UNEXPECTEDLY HIGH friction in pin-jointed cable anchor plates has been identified as the cause of last Friday's aborted attempt to raise London's Millennium Wheel.
The failure occurred under the glare of TV cameras during final testing prior to the erection of the 1,600t wheel. Four temporary support cables pulled out of their anchors after unplanned load eccentricity in the cables distorted the anchor housings.
Mace project director Tim Renwick this week confirmed that it will take four weeks to redesign, test and install replacement anchor housings, but insisted that the wheel would still be up and turning by 31 December 1999.
'We can bring forward parallel activities and increase the amount of prefabrication off site,' he said. 'Providing we don't have another setback we should still make it.'
The temporary cables are necessary to reduce loading on the Wheel's permanent spokes during the lift and prevent its rim from distorting. In the horizontal position around 80% of the load is in these cables but, as the wheel approaches the vertical, this falls off as load transfers to the permanent spokes.
Problems arose during the final testing when 110% loading was being induced in the cables prior to lifting. Slight misalignment of the temporary cables at the point of entry into the anchor blocks caused the securing wedge collars to fall out and allowed the cables to pull through.
Mace design manager Neil Thompson said the temporary cable anchor block head would now be fixed so the only rotation allowed would be in the pin joint attaching the anchor plate to the inner rim chord.
'The distance from the cable anchor to the pin joint means there will be sufficient moment to overcome any friction in the joint,' he explained.
The wheel is being designed and constructed by Dutch contractor Hollandia with the design checked by consultant Allott & Lomax. Renwick defended the design and testing procedure, which he said was second to none. He said: 'The design checking and testing is a managed procedure. The fact it failed during the test proved that our procedure was right.'