The report below has been sent to CROSS by an engineer who wants to illustrate potential problems due to devolved responsibilities on site.
A project is supported by numerous steel columns which start at ground floor level. Because of various site constraints, several columns are kinked at upper floor levels; the resulting horizontal forces are taken back to a substantial concrete core through the floor beams or slabs. The second floor slab is post tensioned concrete and the contractor decided to erect the steel first and install the second floor slab later, when the steelwork and concrete subcontractors would not interfere with each other.
The steelwork subcontractor installed temporary bracing to two columns which kinked at second floor level. In the permanent structure these columns are unbraced from second floor to fifth floor. The steelwork subcontractor identified (correctly) that without the second floor in place there would have been a problem with major axis bending due to the kink, and a problem with minor axis buckling because the column was unrestrained over five floors. The main contractor removed the temporary bracing (because "it was in the way") shortly after the column was connected to the fifth floor steelwork. He was surprised to be told that the bracing was still required - he believed the column was properly restrained when it was connected at top and bottom. He claims that the engineer
should have identified a "hold point" (either on a sequence drawing or a health and safety note) - "column should be restrained at second floor if construction proceeds above second floor before permanent second floor structure is in place" - or similar.
The engineer's drawings clearly showed the kink points in plan and section, and identified the horizontal forces applied, in the permanent structure, by the kink to the second floor slab. The engineer believed that he should not have to tell a competent contractor that a kinked column needs temporary restraint at the kink; nor should he have to point out that a column designed to be unbraced over three floors needs extra restraint if it is erected unbraced over five floors. The problem appears to be that the main contractor has no in-house temporary works engineer. The contractor's staff rely on subcontractors or call in an engineer if they decide it is necessary - but because the decision is made by the site agent (and the site agent is not an engineer) they do not always get it right. There are "scope gaps" between the subcontractors and nobody to co-ordinate them. In this case the main contractor decided to remove some temporary bracing without understanding the consequences. The problem was picked up by the consulting engineer's REs and the steelwork subcontractor but there was a potential for a serious problem.
This report raises issues which go the heart of the construction process. The Report of the Advisory Committee on Falsework, June 1975, states: "…at each stage of the design and construction of Falsework a check or inspection should be made by a competent person. A senior designer must countersign the plans: an inspector must sign off the construction. We therefore believe that it is essential that one individual in the construction organisation be given the duty of ensuring that all procedures and checks have been carried out. ….we describe this person as the Temporary Works Co-ordinator." The message has been repeated many times since then.
BS 5950 has two particularly relevant clauses:
Clause 2.1.1.2---one engineer needs to be clearly identified with responsibility for overall stability and they need to “ensure compatibility” of parts by others contributing to overall stability even if designed by others.
Clause 2.4.2.3---where resistance to horizontal forces is provided by something other than a steel frame the steel design needs to “clearly indicate the need” and “state the forces acting on it”.
The need for temporary bracing therefore needs to be identified by the engineer if they are aware that the sequence of works requires it. Instability introduced by a change in the contractors’ method of construction is however out of the engineer’s hands if he is unaware of the change. It is not uncommon to find issues such as this when the structure is part complete and not fully restrained. Some contractors may not have the expertise to identify such problems, especially when the structure is unusual, and some designers may not be sufficiently conversant with erection processes so that potential problems may not be identified. There are lessons to be learned from this report.
1. It should always be assumed that steelwork is potentially unstable during erection and a method statement, which is agreed between all involved parties, should describe the erection sequence so that a competent and responsible engineer can check that the frame is stable in all stages up to completion.
2. Site visits by the designer during construction are to be encouraged so that any changes are identified.
Further information can be found in BS 5975 Code of Practice for temporary works procedures and the permissible stress design of falsework. Finally, the report raises an issue which has been of concern to SCOSS for some time: that on many sites there is an absence of competent structural engineers in the key decision making process.
Keep up to date with all the latest news and information from CROSS. If you have already registered on the old CROSS website there is no need to register again.
Reports can be submitted through our online form
or
by posting back a completed offline form