A transfer beam supporting a four-storey building was notched to accommodate suspended ceiling fixings.
Because of planning restrictions on height, says a reporter, an engineer designed a shallow-depth steel transfer structure to support a four-storey building above. The main beams had flanges that were 50mm-60mm thick. When the building was structurally complete, there was obviously a certain amount of deflection in the main beams. A follow-on contractor, erecting aluminium T-section ceiling grids, found that the legs of the T-sections fouled the beam’s bottom flange. To overcome this, the contractor simply cut notches in the flange. Fortunately, this was spotted and repaired.
This is clearly a matter of the contractor’s competence, but it raises several questions:
- Was the contractor selected on the basis of their experience, or was it simply a matter of least cost?
- Was the designer of the structure aware of the ceiling grids, and did they allow for a deflection zone?
- What were the quality assurance processes on-site?
- What was the level of supervision of the contractor?
Although a simple error, this could have had serious consequences. Is education the key to mitigating basic errors? The operatives undertaking this work may have been trying to make things easier for themselves to get the job done, with a lack of appreciation for the consequences of their actions. Had there been any engineering presence on-site, or competent supervision, this could have been prevented. CROSS has seen similar examples that would also have potentially very serious consequences. Less competent contractors seem to think that designers factor in excessive safety measures and that a notch here or there is, therefore, not going to be a problem.
The basic safety violation is clearly that of making unauthorised changes. In this case, that was especially dangerous, since alterations were made to what was obviously a major structural item. Design includes the activity of co-ordinating a structural system so that all components fit together properly, negating the need for alterations. Necessarily, that process involves a proper appreciation of any movements and tolerances that might affect alignment.
A wider issue is the deflection in transfer beams. The use of engineering judgement is signalled by BS 5950, EC3 and numerous other reference documents regarding deflection control. Some would argue for a limit of between L/500 to L/1000 for total deflection in transfer structures, depending on the span, or the nature of the loads and structure supported; or an absolute limit of 15mm, rather than a proportion of the span. These figures aim to be comparable with the stiffness of foundations, which may typically settle a maximum of 12mm-15mm in low rise, three- to four-storey buildings. There is guidance available on the design of transfer structures, including the paper Structural Design of Transfer Structures. The following extract is worth reading in this context: “In general, building codes refer span/250 as an appropriate limit value for the vertical deflections of beams and slabs, for the quasi-permanent loads. Furthermore, span/500 is normally an adequate limit for deflection after construction, meaning the deflection which occurs after the addition of partition and finishes.”
Currently, there is not much guidance on deflection limits for transfer structures, although it is commonly accepted that their design should follow more severe criteria than normal beam or slab elements. This can be achieved either by imposing stricter limits, or by designing for a more severe load combination, such as frequent loads or characteristic loads. Despite this, the serviceability criteria for global transfer structures must be specified for each project and agreed with the client. The reporter does not mention how the remedial work was done. Repairs to a notch in a thick steel flange in tension require care and consideration.