Lösulls brandisolerande förmåga vid gipsskivans bortfall i vindskonstruktioner

A commonly known problem with light attic joist constructions is the spread of fire to the attic in connection with a fire in multi-storey buildings. This has proven to have disastrous consequences, where this can result, among other things, in the roof construction losing its load-bearing capacity and the risk of collapsing thereby arising. It has also been shown that there is a great risk of fire from the attic spreading to other underlying spaces. In this report, light attic joists consisting of 45x95 primary studs, 28x70 laths of sparse panel, vapor barrier, loose wool of glass wool and a layer of gypsum boards have been studied under the influence of fire. The joists were subjected to temperature effects in the fire furnace at Luleå University of Technology, where the temperature effects were tried to be kept similar to the standard fire curve according to ISO 834. Six tests were carried out, where three different test setups were tested twice each. The test setups consisted of the following named “Normal”, “Ergolite” and “Hönsnät”. Normal constituted the zero test against which the other test setups were compared to and consisted of the previously named composition but with a normal gypsum board from Gyproc as its underlying cladding. Ergolite consisted of the same composition as Normal, but where the normal gypsum board had been replaced by Gyproc’s Ergolite board. This gypsum board is built with lightweight technology, weighs 25 % less than a normal gypsum board and has a fiberglass reinforced gypsum core. The final test, Hönsnät, had a normal gypsum board as its underlying cladding and consisted of a practice solution, namely attaching chicken netting underneath the loose wool. This was done as the loose wool should in theory fall down when the gypsum board falls away and thus not contribute any insulating ability to the attic floor above in the event of a fire. To carry out these tests, seven thermocouples were placed in the joists before they were exposed to fire. During the tests, the temperature and time when the gypsum board and loose wool fell off, insulation ability and when charring of the underlying primary stud started were measured. These were then compared with literature as well as calculations made with finite element method to check whether the insulating ability could be reflected in finite element analyzes both when the gypsum board falls away and remains. The results showed that the normal gypsum board fell away completely when the average temperature reached about 560 °C. This occurred after approximately 46 min. For Ergolite, there was no fall-off of the gypsum board during the test, but it had fallen off the day after when the fire furnace was opened for cleaning. During the 60 minutes that the tests on Ergolite were carried out, a maximum temperature behind the gypsum board of 650 – 700 °C was reached depending on the test attempt. When the gypsum boards were compared to calculations according to studies by Just and Eurocode, it could be seen that normal gypsum boards from Eurocode are calculated to fall-off after 1440 s (23.7 min) and according to calculations according to Just after 930 s (15.5 min). That the normal gypsum boards in these tests fell off later than according to Eurocode and literature may be due to different definitions of failure. In this study, the fall-off is defined as the time when thermocouples no. 1 and 2 increases rapidly, which was a result of large parts having fallen off, but according to Eurocode and literature, the fall-off is defined as when 1 % of the gypsum board area has fallen off. Ergolite boards that are glass fiber reinforced did not fall-off during the tests and can be more closely described as a fire gypsum boards than a normal gypsum board. The finite element calculations can be applied to simulate the loss of gypsum boards and also follows the temperature distribution in the joist relatively well. However, this only applies when the thermal properties of the materials have been modified. The conclusions that have been drawn are that Ergolite boards can withstand a higher temperature impact than normal gypsum boards. However, more test with higher temperatures are required to determine at what temperature these falls off. In addition, more tests are also required with different loads above the gypsum board to control the load’s influence on the temperature and time when the gypsum board falls down. Finally, it could be stated that the practice solution of using chicken netting under the loose wool has a better ability than without chicken netting to maintain the loose wool in the event of a fire.