Window Glass Breaking by Small Compartment Fires - Literature review Example

WINDOW GLASS BREAKING BY SMALL COMPARTMENT FIRES By Student’s name Course code and name Professor’s name University name City, State Date of submission Introduction Window glass breaking in compartment fire is a significant problem of study within the fire engineering profession based on the fact that prior to breakage the window usually poses as a wall. Various researchers have carried out a wide coverage on this topic with a zillion of articles resurfacing within the internet libraries on this matter. Further, symposiums and conferences have been held with a focus on the techniques to estimate glass breakage from the time a fire starts keeping in mind the accumulation of pyrolyzates in the hot layer and a sudden drift within the backdraft and flashover stage. Most of these studies however are not specific on the compartment type hence the need to investigate window glass breakage in small compartment fires. This literature review therefore covers some of the most important experiments that are relevant with regard to compartment fire glass breaking as a launching platform for the main issue of concentration. Literature Review According to a research carried out by Babrauskas (2010), glass tends to crack at a temperature of around 150°C – 200°C. These cracks do not however possess any importance to this study unless they aggravate further to an extent that they may cause breakage or eventual fallout of glass pieces. These cracks according to Pagni and Joshi (1991) are caused by thermally induced tensile stress. These researchers further give a breakdown of how the surrounding edge and the size of the room is important to the determination of the tensile forces that are responsible for this problem. This problem is brought about due to the nature of glass as a poor conductor of heat prompting infrared immersion and convectional processes to thermally expand glass. This leads to tension and eventual fracturing of glass due to the irregular distribution of heat throughout the surface. Bifurcation occurs on glass depending on its characteristics for example tempered glass takes longer to break in case of exposure to fire. The manner in which glass is mounted also affects the nature and time taken for breakage for example a double glass mounted window takes longer to break. The importance of glass as a weak point on the compartment wall is described by Keski-Rahkonen (1988) who researches on the role played by openings. Due to the fire resistance nature of glass, breakage creates an opening at the advanced stage of fire. In coming up with a concrete backing Rahkonen seeks to unearth answers to the topic of coverage “Breaking of Window Glass Close to Fire.” The approach applied though aims at the periodical loss of tightness through statistical analysis as a method of coming up with a rate of heat release against time curve. Although this might have been carried out through deployment of theoretical methods, the confidence levels that are posed by experimental techniques are usually higher. The experiments staged sought to investigate the spread of fires caused by large openings due to glass breakage. In a similar study as Keski-Rahkonen (1988), Skelly et al., (1991), set up several experiments to establish conditions under which glass breakage in compartment occur with respect to edge protection and exposure. Using a simple remark, Skelly et al. instil sense in this topic in that they illustrate ice cube breakage to explain the thermal expansion and eventual breakage of glass. They use the simple equation (1) below to express the sentiments shared with Keski-Rahkonen (1988) who is the main innovator in this area. The proportionality in the difference between average and glass temperatures (local) are however determined prior to determination of temperature differences in order to come up with a suitable compartment size for experimental consideration. (1) Where = Normal stress at failure = Young’s Modulus = Linear thermal expansion coefficient = Temperature of heated glass = Temperature of insulated edge The breakage mechanisms that are identified by Keski-Rahkonen (1988) are worth observation in case of a full blown research on this topic. Apart from the thermal stress, one of the most significant mechanisms identified is the intense heat flus that occur abruptly on the glass due to differences in thermal gradient. Thermal shock as this is referred causes a deep gradient which is responsible of this transient phenomenon as soon as radiation kicks off. Thermal gradients are also blamed for planar deformation of tightly held glass depending on the glazing and tempering. Such pane conditions usually create a condition necessary for bending to occur depending on the glazing work. Non-uniform heating on the other hand causes tension on weak locations and a heat sink that eventual breaks the glass. Skelly et al. (1991) stage an experiment to study this phenomenon under conditions as those found in normal building fire. Using a compartment with dimensions framed using an angle of and iron slats, fire insulation board, Kaowool ceramic fibre attached to the walls the experiment is set into a small scale building condition. Ventilation path was provided through a rectangular window provided at the centre of the insulation board. The opening area provided was at above the compartment floor which was precisely below the roof. On the other hand, the plenum was framed using the same size of angles and of steel plating at a 58cm height. The vanes were built using a mild steel plate in order to provide ventilation to the compartment. The incoming air was let in through the lower plate in toward the centre of the compartment through the lower position. The figure below shows the experimental setup that was used in this experiment. Figure 1: Experimental setup for a glass breaking in compartments (Skelly et al., 1991). The experimental glass applied in this experiment was normal soda-ash with a thick cut using a hand scribe. The edges of this glass were not in any way ground but a protected edge was maintained all round the window frame. Hexane as the experimental fuel was set on fire at the centre of the compartment while held using aluminium trays. Skelly et al., (1991) undertook a total of seventeen experiments on edge protected and unprotected edge in order to draw their conclusions accordingly. The following graph was obtained as a standard for these experiments: Figure 2: A graph of temperature against time for a glass of (Skelly et al., 1991). The consistency achieved from this experiment creates an average temperature difference of 90°C as a valuable support to theory above. The temperature difference theory is backed up as a means of glass breakage in two different situations i.e. protected and unprotected. Theoretical predictions place this value at 70°C which is considered as being 30% below the experimental value. These differences according to the experimenters, is created by differences in thermocouple designs. The mechanisms of breakage mentioned above are therefore valid for investigation within this study as they shall reaffirm these findings. Although breakage patterns are also mentioned, they do not add any value to this study and therefore shall be ignored. In all the cases investigated, the bifurcations are the source of breakage and in the case of edge-protected windows no glass is spared. This prompts a research to be carried out to investigate the incorporation of vents instead of windows as a replacement windows through a computer aided simulation (Skelly et al., 1991). A similar experiment was carried out by Hassani et al., (1995) in same conditions as Skelly et al., (1991) not only expose the breakage mechanism but also the effects of surface exposure. Temperatures of up to 380°C are applied as an exposure methodology – only achieved after the flashover region. The exposed section of the glass used in this particular experiment on reaching 150°C – 175°C breaks while the unexposed side is at 75°C to °C. This experiment is further refined by Shields et al., (2001) who use of 6mm of glass as a guide thereby breaking at a gross temperature of around 110°C. However the shortcomings are noticed as the researchers do not identify the exact room temperature during which the glass fall out although the initiation temperatures are set at 450°C. A similar experiment has been staged although in a different perspective to study the effects of glazing procedures on glass breakage in compartments. A single glazed thick glass was imposed on this test through a painstaking process meant to come up with a graphical presentation of the glass breaking path. Although the results established from this experiment reflect the same graph as shown in figure 2 above, the mean gas temperature at which the glass pane breaks are set at around 240°C – 360°C with a deviation of close to 50°C suggested. Further experimenting on double glazed windows exhibits deterrent characteristics since the wavelength required for transmission from inner surface to outer pane is quite longer. At the beginning of fire, the outer pane does not at all absorb the heat and may not even end up breaking in case the fire is terminated within a short period of time. Although this experiment does not give the fallout temperature for the second pane, it is evident from theoretical calculations that for thickness, the required amount of heat is (Shields et al., 2005). An experiment has also been staged in the past to investigate the effects of double glazed windows with a specific objective to establish the time taken for a thick glass to fall out under the same conditions as the above experiment. The fuel used in this experiment is however altered from the normal methanol to of wood which manages temperatures of up to 600°C and within 8 to 10 minutes, the glass start falling off thereby offering a new life to the fire due to ventilation. In a full scale experiment, thick glass starts falling at 5 minutes after the launch of fire during which the temperature is observed to be 600°C (Loss Prevention Council, 1999). In addition, Cohen and Wilson (1994) who carried out a similar experiment confirmed that the corresponding fluxes for the temperatures above range from. Theoretical simulations have also been employed by the likes of Dembele et al. (2012) and Babrauskas (2010) with a positive outcome noted in their conclusions. Dembele et al. uses a simulated model to establish that thermal breakage of window glass in typical fires is actually due to the stresses and strains that are critically detailed in their generated pictograms. The panes boundaries conditions (constrained and unconstrained) are investigated with a conclusion that unconstrained glass faces multiple fractures as compared to its counterpart which takes longer to experience such kinds of cracks. The edge conditions are thus noted as important considerations to make while designing glass window panes. Another point that is worth for this study is that the edge condition (as-cut, polished and ground) does matter in the breakage experiment since analysis of these models exhibit as-cut edge to be stronger than the rest. The authors of this research do not however venture into the different types of window glass products in their study citing that the normal soda-ash glass is of more importance (Dembele et al., 2012). Figure 3: Glass conditions studied by Dembele et al. (2012). Away from the glass breakage matter, the effects caused on ventilation are worth investigation too. This approach has notably been used by most authors whose works are discussed in the literature on glass breakage above. Creating an abrupt opening within the building is likely to cause an increase in heat release rate which in turn becomes catastrophic and a hindrance to the fire fighting process. Creation of openings above the flame causes an increased outflow of smoke within the compartment leading to an increase in combustion levels. The thickness of the hot layer is also increased as a result of enlarged smoke release area although this factor is underestimated in most studies that major on the ventilation issue (Merci & Vandevelde, 2007). Tang et al. (2012) is keen to exploit the research loop holes left behind by previous researchers for their specific topic of study, “Experimental study on flame height and temperature profile of buoyant window spill plume from an under-ventilated compartment fire” in which they establish a very important observation. Window geometry is actualized as a point of contention in specific circumstances such as the study of window glass breaking in small compartment fires. Creating an opening therefore poses as a temperature release valve thus leading to reduction of heat at fire bases in case of breakage. However the determination of geometry is not well described in this study thereby eliciting an urge to study the effects of window geometry with regard to glass breakage identified in some of the studies above. This further determines the amount of plume spilling from the compartment thereby suggesting further damage or even flame transfer if mitigation measures are not kept in place as a matter of urgency (Tang et al., 2012). In a report by Kerber (2010), he notes that fire ventilation as a source of spontaneous increase in openings causes an unnecessary pressure due to an increased rate of burning. Backdraft conditions have been noted by fire service operators when methane fuel is used in small scale experiments in which the window breakage experiment is replicated. Computational fluid dynamics in under-ventilated simulations have also been applied in establishing the effects of ventilation on heat release thereby giving a conclusion that glass breakage has an adverse effect. This however varies depending on the size and the configuration of the openings. Although window size is not significant to the fire behaviour, the significant conclusion is that the increased fire load is also of importance. Positive pressure due to abrupt increase in ventilation area as glass breakage offers a positive ventilation which in turn offers an increased heat release rate. As much as the breakage may seem small scale, the effects on fire spread are immense and worth studying – an energy increase of is discovered by Kerber (2012). This is a magnification level of 6:1 which is responsible for a temperature rise of 800°C to 1050°C within a matter of seconds. Therefore, there is a strong connection between the positive pressures and extra ventilation. Ventilation is also an established causal agent of reduced temperature at low elevations according to the literature in this study. Finally it is worth to explore the extent to which ventilation is influenced by the path travelled by air to stoke fire further – and as a result of the main issue of contention which is glass breakage. Conclusion Several studies have been staged to study the effect of glass window breakage by small compartments with various approaches in the minds of the authors. Glass breakage has been largely attributed to poor heat conduction properties and poor distribution of tensile stress. The issue worth further research however is the effect on ventilation and fire properties in small compartment fires with respect to the size of opening created. Positive pressure ventilation has also been discovered when further ventilation is introduced in the middle of fire in a compartment. This background study thus forms enough ground for the study of this topic in perspective of properties imparted to fire and ventilation. The experimental procedure mentioned in this literature shall however be modified to fit the course that this study shall choose. List of References Babrauskas, V., 2010. Glass Breakage in Fires. Fire Science and Technology Inc., pp.1-7. Cohen , J.D. & Wilson, P., 1994. Current Results from Structure Ignition Assessment Model (SIAM) Research. In The Biswell Symposium: Fire Issues and Solutions in Urban Interface and Wildland Ecosystems. Kananaskis, Alberta, 1994. Dembele, S., Rosario, R.A. & Wen, J.X., 2012. Thermal breakage of window glass in room fires conditions e Analysis of some important parameters. Building and Environment, 54, pp.61-70. Hassani, S.K..S.T.J..a.S.G.W., 1995. An Experimental Investigation into the Behaviour of Glazing in Enclosure Fire. Journal of Applied Fire Science, 4, pp.303-23. Kerber, S., 2010. Impact of Ventilation on Fire Behavior in Legacy and Contemporary Residential Construction. Northbrook, Illinois: Underwriters Laboratory inc. Underwriters Laboratory. Keski-Rahkonen, O., 1988. Breaking of Window Glass Close to Fire. Fire and Materials, 12, pp.61-69. Loss Prevention Council, 1999. Fire Spread in Multi-Storey Buildings with Glazed Curtain Wall Facades (LPR 11: 1999). Borehamwood, England: Loss Prevention Council. Merci, B. & Vandevelde, P., 2007. Experimental study of natural roof ventilation in full-scale enclosure fire tests in a small compartment. Fire Safety Journal, 42, pp.523-35. Pagni, P.J. & Joshi, A.A., 1991. Glass Breaking in Fires. In Fire Science Safety - Proceedings of the Third International Symposium. California, 1991. International Association for Fire Safety Science. Shields, T.J..S.G.W.H..a.F.M.F.., 2001. Performance of Single Glazing Elements Exposed to Enclosure Corner Fires of Increasing Severity. Fire and Materials, 25, pp.123-52. Shields, J., Silcock, G.G. & Flood, F., 2005. Behaviour of Double Glazing in Corner Fires. Fire Technology , 41(1), pp.37-65. Skelly, M.J., Robby, R.J. & Beylar, C., 1991. An experimental Investigation of Glass Breakage in Compartment Fires. Journal of fire protection engineers, 3(1), pp.25-34. Tang, F. et al., 2012. Experimental study on flame height and temperature profile of buoyant window spill plume from an under-ventilated compartment fire. International Journal of Heat and Mass Transfer, 55, pp.93-101. Read More