The Fire Performance of Post-Tensioned Timber Beams
Thesis DisciplineFire Engineering
Degree GrantorUniversity of Canterbury
Degree NameMaster of Engineering in Fire Engineering
Post-tensioned timber frames have recently been undergoing heavy research and development at the University of Canterbury. The recently developed post-tensioned timber system utilises engineering wood products such as Laminated Veneer Lumber (LVL) and glue laminated timber (Glulam), which are formed into box sections and post-tensioned with high strength steel tendons made from stranded steel wire or solid steel bars. The post-tensioning serves to counteract some of the bending actions imposed on the timber beam from loading through a variety of mechanisms. Previous research has focused on the seismic performance and gravity frame performance of post-tensioned timber, both of which yielded promising results. There is however a commonly perceived increase in fire risk with timber building, particularly multi-storey timber buildings, and the fire performance of post-tensioned timber had not previously been investigated. Therefore, the focus of this research was to investigate the fire performance of post-tensioned timber beams. This was completed through a series of full-scale furnace tests, and the development of a fire resistance design method. Three 4.36m span post-tensioned timber beams were exposed to the ISO 834 standard fire. Each of the test beams were glued box beams made from 63mm LVL and were of varying external dimensions. Each beam was intended to demonstrate a specific failure mechanism at approximately 60 minutes of fire exposure. The failure mechanisms demonstrated were a shear failure in the lower corner of due to corner rounding, and a combined bending and compression failure at the end of the beam. These failure mechanisms are unique to post-tensioned timber in fire. The results of the experimental testing were used to validate and refine the proposed fire resistance calculation. Also tested during the full-scale testing were five different forms of anchorage fire protection. These were tested as a secondary objective, but useful thermal data was collected. Through the full-scale testing and the calculation method development it was found that it is important to consider shear during fire design. The post-tensioning increases the bending capacity of a beam but doesn’t affect its shear capacity, therefore when more loading is applied to utilise the increased bending capacity the shear action is increased which leads to shear governing the design in many cases. It is also important to consider shear not only in the webs at the centroid where the shear flow is greatest but also in the lower corners, which can become much thinner than the webs. Without calculation it is not possible to determine where the shear stress will be greatest and therefore both the web and the lower corners need to be checked. It was also found that as the timber section chars on three sides the post-tensioning eccentricity increases which can lead to the moment at the end of the beam becoming critical. Other failure mechanisms which need to be checked include, combined bending and compression at mid span, and tension in the bottom most fibre at mid span. It was found that the proposed calculation method, when used with a char rate of 0.72mm/min and an additional allowance of 7mm for temperature-affected timber beneath the char layer, provided good predictions of the failure times for the full-scale experiments.