A study of the performance of cellular composite floor beams at elevated temperatures could benefit long-span construction and steel industries.

Cellular steel beams have become widely used in multistorey buildings, as they are said to reduce total steelwork weight and help decrease the depth of floors by accommodating pipes, conduits and ducting.

However, as the popularity of cellular steel beams in composite floors has grown, so too has the need to pay increased attention to fire safety engineering design.

Professor Nadjai, of the University of Ulster, UK, says that recommendations for cellular beam designs ‘in fire limit states remain very primitive, and this is due to the lack of general research in the area’. He notes that, while some tests have ‘provided a significant step forward in terms of understanding the structural behaviour of steel-framed buildings under fire, the maximum span of the means in the tests was limited to nine metres, whereas spans of 15m are more common when cellular beams are used’.

To address this, the team designed a large-scale natural fire test on a composite floor slab supported by long-span cellular beams. The test was conducted in a 15x9x3m fire compartment sponsored by the European Coal and Steel Community.

Adding to the compartment fire test, 18 full-scale composite cellular steel beams using span lengths from 4,500mm to 7,500mm and subjected to one-point and twopoint loadings using symmetrical and assymetric composite beams were tested at the University of Ulster laboratory (FireSERT). The project was sponsored by the Engineering and Physical Sciences Research Council.

He says, ‘failure temperature observed in the fire tests indicated that failure by web post buckling of cellular beams cannot simply be estimated by applying temperature-dependent reduction factors on stiffness, as given in codes [fire engineering design codes EC3/4 Part 1.2 and BS5950 Part 8].’

The fire compartment test concludes that, ‘the reinforcement in the central region of the slab was under tensile force forming around the perimeter of the slab. Due to membrane action, the existence of secondary beams to support the slab is not necessary in the fire condition, and these beams can be left unprotected,’ he claims.

According to Nadjai, the team’s ‘real fire’ test has helped develop ‘uniform European design rules for protected and unprotected cellular beams’. He says, ‘the UK construction market is the major beneficiary. Designers, fabricators and contractors are in need of a validated performance-based design methodology for the fireresistant design of non-composite and composite steel floor beams.’

Their European design code for long cellular beams was approved by the European Commission, and will shortly be distributed to universities, consultant engineers and building controls.

Charles Fentiman, of Fentiman Consulting in Southwater, UK, says, ‘this is an important area of research with new building methods being established. However, in assessing the results of these tests, it is important to assess the state of the concrete after test. Portland cement concrete is not refractory and at high temperatures breaks down to form quicklime. Although the concrete may still be OK after the fire test, any incoming moisture could cause it to fail by rehydrating the quicklime’.