Technical: Responsibilities in steel frame design

NSCJuly2016digi

Technical The Structural Engineer of April 20161 posed a number of questions about the responsibilities of the structure designer and the connection designer – presuming the connections are to be designed by the steelwork contractor. David Brown of the SCI offers a detailed response. 26 NSC July/Aug 16 Responsibilities in steel frame design In the April 2016 edition of The Structural Engineer, the ‘Verulam’ section presented a series of 6 scenarios presenting ‘grey areas’ where the correspondent suggested that responsibility was unclear. This article summarises the key elements of the question and provides a response. 1. Connections with high tying forces. The scenario presented is that high tying forces demand ‘strong’ connections, which are likely to be stiffer than ideal – no longer nominally pinned - and transfer significant moments into the columns. The question related to the responsibility for verifying that the columns are still satisfactory. The short answer is that the original structural designer must have an appreciation of the likely connection. The designer of the structure must anticipate that if the forces are so large that a nominally pinned connection is not physically possible, the design rules for “columns in simple construction” are no longer appropriate and the columns should be designed to accommodate the larger moments. The Green Books on Simple Connections2,3, give tabulated resistances in shear and in tying for nominally pinned connections, so developing this necessary appreciation of the likely connection is not onerous. In fact, a more realistic scenario is when a designer specifies axial tensions in the beams that are not tying forces – for some reason they are ‘real’ forces. Immediately, this is at variance with the concept of “simple” or nominally pinned connections, which are “shear only”. Although nominally pinned connections can be verified for shear and, as an entirely separate check, a tying force, the Green Books do not contain any design rules for the combination of shear and axial forces. In the original question, it was suggested that BS 5950 was “a little hazy” about requiring the connection flexibilities to be checked to ensure that they comply with the frame design concepts. Not so – clause 2.1.2.1 requires that “in each case the details of the joints should be such as to fulfil the assumptions made in the relevant design method” although it might be argued that BS 5950 does not specify how stiffness is to be calculated. It might also be said that BS 5950 puts the onus on the connection designer to meet the structure designer’s assumptions, but this cannot be reasonable or sensible if those assumptions are unrealistic. The Eurocodes place the responsibility squarely with the original designer. To paraphrase BS EN 1993-1-1 clause 5.1.2, the effects of the behaviour of the joints… must be taken into account when they are significant. In clause 5.5.1(2), “the calculation model and basic assumptions should reflect…. the anticipated type of behaviour of the cross sections, members, joints and bearings”. This leads on to BS EN 1993-1-8, where rules are presented to calculate joint stiffness and compare this with limits on nominally pinned, semi-rigid and rigid behaviour. Rather than follow the calculation procedure, the Eurocode points out that a joint may be classified on the basis of “experience of previous satisfactory performance in similar cases”, which seems a more attractive option if that experience exists. In the UK, designers have the advantage that the National Annex notes that connections designed in accordance with the principles in the Green Book on Simple Connections3 (Figure 1) are nominally pinned, without justification by calculation of stiffness. 2. Flange to web welds in a plate girder. This question has reached SCI on a number of occasions. The responsibility lies with the designer of the member, not the connection designer. 3. Joint resistances in hollow section trusses. The situation described was when checked by the connection designer, the joints required expensive stiffening (although it was really strengthening that was required). When the truss designer has selected members, the joint resistance has also been set. Joints should be checked as part of the design process, as judicious choice of members and geometry can lead to nodes which do not need strengthening. As the question in Verulam noted, there is published guidance on this specific subject in Steel Industry Guidance Note SN484. All these guidance notes are available on Steelbiz. Although checking joint resistance can appear daunting (see Figure 2 over the page showing part of BS EN 1993-1-8), software is available. Free software can be obtained from Tata Steel Tubes, in Corby – the contact number is listed on SN48. 4. Holding Down Bolts and foundation design. The question focused on the design responsibility when holding down bolts are in tension. As the original contributor noted, this is covered in Steel Industry Guidance Note SN515. Once the loads in the anchors have been calculated by the steelwork contractor, it is for the consulting engineer to design and specify the anchorage arrangement and the base reinforcement. Figure 1: One of the Eurocode ‘Green Books’  Continued on p28


NSCJuly2016digi
To see the actual publication please follow the link above