Experiment Which Allows Students to Develop an Understanding of Compartment Fires - Assignment Example

NAME: XXXXX TUTOR: XXXX DATE: XXXX INSTITUTION: XXXX ©2016 Aims (5 marks) The experiment allows students to develop an understanding of compartment fires. The aim of this experiment is to assess how a compartment fire develops from ignition to flashover and decay. The primary objective of the research gets achieved through burning PMMA inside the fire box in the laboratory. Apparatus (2 marks) The firebox’s roof floor and walls are made out of 0.025 m thick ‘Monolux 500’ lining with one wall constructed from fire-resistant glass. An adjustable door opening, 225mm high and opens between 0 and 0.15m wide. The fuel rests on an axle in the bottom of the floor with a balance to measure the fuel’s weight. [BJM81] Figure 1 –Side view of fire box with thermocouples showing nominal measurements Thirteen thermocouples, mounted on three columns (positioned at 0.15m, 0.315m and 0.48m from the opening), are used for the measurement of temperature during the experiments. These thermocouples are ‘type K’ (nickel – chromium / nickel - aluminium) with stainless steel sheet. Twelve measure the inside temperatures of the compartment and one protrudes out. A ‘Squirrel’ data logger is used to record the temperatures of the compartment. As fuel, square slabs of PMMA (Polymethyl Methacrylate) are burnt in the small-scale fire compartment. One PMMA samples are burnt per test. The fuel thickness is approximately 0.012m (12mm) or 0.25m (25mm) measuring approximately 100mm X 100mm, or 150mm x 150mm or 200mm x 200mm were tested. Figure 2 –Fire box front view, full and half- sized opening showing nominal measurements Methodology (5 marks) The following is the generic procedure used for conducting the experiments. The procedure was identical for all samples. The internal dimensions of the fire box were measured and the location of where the sample will be throughout the test identified and measured The size of the ventilation was recorded All thermocouples on the inside of the firebox were cleaned with a paper towel (so that the readings would not affected by soot or dirty). The thermocouples were connected to the Squirrel data logger. (Thermocouples and the channels to which they match on the data logger were recorded). The balance was switched on. An appropriate sized sample tray was placed on the support plate in the fire box and “tare” pressed on the balance. The sample was then added to the tray and some fine PMMA powder added over the fuel sample. The initial mass of fuel PMMA was recorded. All thermocouples were straightened so they were protruding horizontally into the firebox. The Squirrel data logger was then started. The sample was then ignited using a lit taper while ensuring the appropriate PPE was worn. The vent opening was then opened to the required size (Half or full). A small piece of paper was placed at the front of the box and signs, and time of flashover during the experiment were observed. The mass of the sample and the height of the smoke layer were recorded every minute. Any other observations were also recorded. At a specified time (15-20 minutes after ignition) for all experiments record if conditions are suitable for back draught to occur. The data logger was stopped when the test is complete. What was measured and accuracy (2 marks) The location of the fuel sample, the time, position and nature of ignition, the decomposition of the sample and the peak temperatures for each experiment were measured. The accuracy of the results obtained was dependent on the measuring instruments used. Timeline of the fire (2 marks Time (seconds) Observation Incipient stage 0 - 20 The fuel (PMMA) is ignited and starts burning. Start weight is 540g 80 No smoke being produced yet Growth stage 288 Sample surface is taking hold about one third is alight. The flames are about 100 mm high 366 Temperature at the middle of the box is 1670C Fully developed stage 506 Flames are bigger 520 Flames have covered ¾ of the sample 546 Thermocouples 6 is at 2600C 567 There is quite a bit of smoke exiting the door 682 All surfaces are now ignited, flames are touching the ceiling. Mass is 465g 767 There is quite a bit of turbulence in the flame and smoke. Flames are touching the door 980 - 1002 Flash over stage takes place 1024 - 1054 Consistent flaming out of door. Smoke has high velocity. The weight of the sample is 244g Decay stage 1362 - 1436 Sample mass is 20g. Sample is nearly extinguished 1455 Temperature is 3550C 1495 Sample has gone out Personal observations (2 marks) The powdered PMMA was added for easy ignition of the fuel. This is due to the high surface area to volume ratio offered by the powder for the fire to act upon. At about 980 seconds, a piece of paper was placed in front of the box, and the paper smoldered and ignited at about 1002 seconds. The smoldering was taken to be a warning sign that flashover would occur. This was recorded as the flashover event. The time taken for the piece of paper to smolder and ignite was approximately 20 seconds. This shows that the fire was relatively slow. At about 1040 seconds, smoke spewed out of the combustion chamber at a high velocity. It is a result of the turbulent nature of the combustion reaction when the fire is at a fully developed stage. Data obtained, Explanation & analysis of data (10 marks) Temperature - Time Distribution The graph shows the various temperatures recorded by the thermocouples placed at different locations from ignition of the fuel, during combustion till its decay. It displays the temperature distribution in the fire box. Thermocouples located lower than the fuel source and have similar temperature distribution. Thermocouple 1 however, has the lowest value due to its position farthest from the fuel source. Thermocouple 9 has the second lowest temperature because it is located at the bottom of the ventilation door, where fresh air flowed into the fire box, bringing the ambient temperature down. Graph 1: Temperature distribution with time for Sample 100x100 Flame impingement could have caused the temperature spikes shown by thermocouple 6, due to the close location of the fuel bed, while the temperature exhibited by thermocouple 4 could have resulted by smoke dispersion. The smoky thermal layer of fire generated at the top of the firebox could have caused the highest temperatures in the experiment were exhibited. Flame impingement due to very close flame could have caused the highest temperature spikes of 3220C were recorded at thermocouples 7 and 8. The flow of smoke over thermocouple 12, located close to the vent explains the spike in temperature recorded. The graph for Sample 200x200 shows that the highest temperature was achieved much quicker (730 sec to get to highest temperature as compared to 1430) and was significantly higher than for Sample 100x100 (702.50C as compared to 2510C). The faster rate of combustion of the Sample 200x200 can be attributed to the availability of extra mass for combustion during the experiment. Graph 2: Temperature vs Time for Sample 200x200 Rate of change of temperature The rate of change of temperature increases rapidly in the first 60 – 90 seconds due to the introduction of the lit paper and initial ignition of the fuel sample. The rate then increases at a much more stable, slower rate as the fuel combusts, up to a peak area (1320 – 1440 seconds) when all the fuel has been ignited. As the fuel sample is diminished, the fire eventually burns itself out. This, coupled with the build- up of smoke in the combustion chamber chokes the fire, leading to a fall in the rate of temperature increase until ambient temperature is achieved. The fluctuations in the graph (crests and troughs) can be attributed to the build-up of smoke in the combustion chamber. The temperature of the smoke is periodically recorded by the thermocouples, which is higher than the ambient temperature of the combustion chamber. Graph 3: Mean Temperature against Time for Sample 100x100 Rate of Mass loss Graph 4: Mass Loss Rate Curve The graph illustrates the rate at which the mass of the fuel sample decomposes with time during combustion. After ignition, the initial rate of loss of mass of the sample is slow. This is due to the solid nature of the PMMA, which reduces the surface area on which the fire can act. As burning continues, more of the fuel is exposed to the fire, increasing the rate of combustion, resulting in a, more or less exponential decrease in the mass of the fuel sample. This coincides with the growth and fully developed phases of the fire timeline. After this stages, the fuel available is being rapidly depleted, causing it to be a limiting factor for the combustion reaction[Kin91]. The rate of mass loss, therefore, slows down in the later stage (decay stage) and comes to zero when all the fuel has been depleted The rate of combustion of the 200x200 sample is faster due to more fuel being available for burning, resulting in a faster mass loss rate. The 100x100 sample exhibits a more even mass loss rate throughout the combustion process until all the fuel is depleted. Evaluation of the experiment (10 marks) The stages of fire identified are ignition, developing, fully developed and decaying stages. However, in compartment fires, another phenomenon takes place referred to as flash over the stage, lying between the developing and fully developed stages[Pop06]. The hot gas layer of the compartment can be predicted by a simple formula known as MQH correlations. The correlation states that for simple compartment fires, the upper layer of the gas needs to be 6000C for flash over to take place. [Bru57] Using an estimated burning area of, the predicted mass loss rate based on area is given by; This is higher than the mass loss rate at flash over obtained from experimental data. The peak mass loss rate achieved from experimental data is The thermal diffusivity of ‘monolux’ is given by The depth of thermal penetration is; The width of the wall (0.025m) is smaller than the depth of thermal penetration, thus the wall is thermally thin. The heat transfer constant is given by; Since flash over occurred, we can apply the McCaffrey [BJM81] flashover condition; References BJM81: , (B.J.McCaffrey, 1981), Kin91: , (Tu, 1991), Pop06: , (Pope & Bailey, 2006), Bru57: , (Bruce, 1957), Read More