Search for: All records

Firebrands are known to be able to ignite not only vegetation but also various structures found in wildland-urban interface (WUI) area. Especially, firebrands located close to each other on a combustible substrate increase the likelihood of ignition and the subsequent fire. To elucidate the ignition mechanism of firebrands, experiments are performed using a 3 by 3 square array of flaming firebrands deposited on a 6.35 mm thick birch plywood. The spacing of the firebrand is varied in each experiment, ranging from 10 to 30 mm. The deposited mass of firebrands lies between 13 and 15 g. Ambient wind is imposed parallel to the plywood surface to investigate its effect on the ignition and the subsequent flame spread over the fuel. Three different wind speeds 0, 0.5, and 0.75 m/s are tested. During the experiments, mass loss of the plywood and the deposited firebrands is recorded. Video cameras are used to monitor the burning process. An infrared camera is also used to monitor the temperature of the firebrands and the plywood. The experiment results indicate that the firebrands with the spacing greater than 20 mm are able to burn only the surface of the plywood until the firebrands burn out. When the spacing between firebrands is smaller than 20 mm, the plywood is ignited and continues to burn even after the firebrands are fully consumed. It is also observed that the flame is able to spread downstream at 10 mm spacing under ambient wind speed of 0.5 m/s. Results from this study demonstrate the significant influence of spacing between the firebrands on the ignition and the burning behavior of the substrate materials. 

Firebrand attack has been shown to be one of the key mechanisms of wildfire spread into Wildland-Urban Interface (WUI) communities. The ignition propensity of materials caused by firebrands depends on not only the attributes (e.g., shape, size, numbers) but also the distribution of firebrands after landing on the substrate materials. To help characterize this process, this study aims to first investigate the effects of gap spacing on the burning behaviors of a group of wooden samples. Experiments were conducted using 9 wooden cubes, 19mm-long on each side. These samples were arranged in a 3 by 3 square pattern on suspension wires. The gap spacing (s) between the cube samples varies from 0 to 30 mm. Burning process was recorded using video cameras. Sample mass loss and temperatures were monitored during the flaming and smoldering processes. The results show that when s ≤ 10 mm, flames from individual samples merged. When the gap spacing reduces, the mass loss rate first increases but starts decreasing at s = 10 mm where flame merging occurs. The flame height has a similar non-monotonic dependency on the gap spacing and the maximum flame height occurs at s = 5 mm. Compared to the case with s = 10 mm, cases with a smaller gap spacing (s = 2.5 and 5 mm) have a larger flame height but a smaller sample mass loss rate. This indicates that a reduced air entrainment leads to an increase in the flame height despite of a decreased flame heat feedback to the solid samples. The heating rates of each sample were also calculated to investigate the local burning behaviors. The analysis showed a weaker flame heat feedback to the sample at the center for cases with under-ventilated combustion. Last, gaseous flame height was corelated to the solid burning rate. The correlation was also compared with previous empirical equations concerning liquid pool fires of different heat release rates. 

Firebrand (ember) attack has been shown to be one of the key mechanisms of wildfire spread into wildland–urban interface communities. After the firebrands land on a substrate material, the ignition propensity of the material depends on not only the attributes (e.g. shape, size, and numbers) but also the distribution of the firebrands. To help characterize this process, this study aims to investigate the effects of gap spacing on the burning behaviors of a group of wooden samples. Experiments are conducted using nine wooden cubes, 19 mm on each side. These samples are arranged in a 3 × 3 square pattern on suspension wires and are ignited by hot coils from the bottom surface. The gap spacing (s) between the samples varies in each test (ranging from 0 to 30 mm). After ignition, the samples are left to burn to completion. The burning process is recorded using video cameras. Sample mass loss and temperatures are monitored during the flaming and smoldering processes. The results show that the flame height and the sample mass loss rate have non-monotonic dependencies on the gap spacing. When the gap spacing reduces, the flame height and the mass loss rate first increase due to enhanced heat input from the adjacent flames to each sample. When s ≤ 10 mm, flames from individual samples are observed to merge into a single large fire. As s further decreases, the air entrainment at the flame bottom decreases and the flame lift-off distance at the flame center increases, resulting in an increased flame height, decreased flame heat feedback to the solid samples, and a decreased mass loss rate. The decreased mass loss rate eventually leads to a decrease in the flame height as well. The gaseous flame height is correlated to the solid burning rate. The correlation generally follows previous empirical equations for continuous fire sources. For the smoldering combustion, compared to a single burning sample, the smoldering temperature and duration significantly increase due to the thermal interactions between adjacent burning samples. To help interpret the results of the burning experiments, thermogravimetric analysis is also performed in air and nitrogen, resulting in heating rates ranging from 10 to 100 K/min.