To ensure the safety of buildings in fire incidents, the fire resistance performance of structural components is evaluated based on the standard fire curve in a heating furnace, considering factors such as compartment size, openings, and types and qua...
To ensure the safety of buildings in fire incidents, the fire resistance performance of structural components is evaluated based on the standard fire curve in a heating furnace, considering factors such as compartment size, openings, and types and quantities of combustible materials. Although applying the standard fire curve is the most common method to assess fire resistance conservatively based on the temperature behavior of idealized building elements, there are significant differences between actual fire behavior and the standardized fire curve due to variations in compartment spaces and combustible materials. Therefore, the purpose of this study is to evaluate the fire behavior of steel columns in indoor spaces during real fire incidents and provide fundamental data.
In this study, experiments were conducted to investigate the fire behavior of steel columns exposed to real fire incidents in indoor spaces. The experiments were performed using a standard fire test chamber (2,400mm (W) × 3,600mm (L) × 2,400mm (H)) and standard opening size (800mm (W) × 2,000mm (H)) based on KS F ISO 9705, which represents the fire scenarios in compartment spaces. Additionally, the suitability and reliability of the Fire Dynamic Simulator (FDS) were verified for temperature distribution prediction using both the simple model and the pyrolysis model. The validated FDS model was then used to evaluate the fire behavior in indoor spaces, leading to the following conclusions.
(1) The quantitative assessment of fire characteristics and size was conducted for a compartment space, applying only half of the fire load of a single-family house. The maximum heat release rate (HRR) was measured as 1027.77kW at 5 minutes after ignition, and the total heat release (THR) was measured as 855MJ. The heat release rate increased rapidly with ignition and started decreasing at around 6minutes when the wood cribs began to burn. Based on the observed fire characteristics and the critical heat release rate, the fire behavior in this experiment was determined to be a local fire within the compartment without flashover.
(2) Examining the fire behavior at different locations within the compartment during the compartment fire, it was observed that the fire went through the growth phase until 6 minutes, the fully developed phase from 6 to 14 minutes, and the decay phase afterwards. The maximum temperature at the fire source location was measured as 1076.1°C at a height of 1.0m. The maximum temperatures at the front and rear locations within the compartment were measured as 661.6°C to 758.6°C during the fully developed phase. Particularly, at the front location where the opening was present, the temperature was distributed approximately 6-12% higher than at the rear locations due to heat flow.
(3) Investigating the fire behavior at different heights within the compartment, it was found that the temperatures were higher at heights of 1.0m, 1.5m, and 2.0m in ascending order for the fire source location. For the front and rear locations within the compartment, the temperatures were higher at heights of 2.0m, 1.5m, and 1.0m in ascending order due to heat accumulation from the ceiling. Furthermore, the condition for flashover, which is a temperature above 600°C, was measured at a height of 2.0m between 4.5 to 7 minutes. However, since the wood cribs started burning at 6 minutes, there was not enough combustible material to create a high-temperature layer, resulting in no flashover occurrence in this experiment.
(4) Examining the fire behavior according to the fireproofing methods for H-section steel and angle steel columns within the compartment, for the uncoated specimens, the surface temperature of the steel remained at ambient temperature for 3 minutes, and the maximum temperatures reached 383.8°C for H-section steel and 394.3°C for angle steel columns during the fully developed phase (14 minutes). For the specimens coated with 10mm fireproofing, the surface temperature remained at ambient temperature until 6 minutes during the fire growth phase, and the maximum temperatures reached 95.8°C for H-section steel and 91.5°C for angle steel columns during the decay phase. Based on this, it can be concluded that for steel columns in real fire incidents, regardless of the cross-sectional shape, the temperature can be reduced by up to 75-77% with a 10mm fireproofing coating.
(5) Examining the fire behavior according to mortar filling and finish mortar attachment for angle steel columns within the compartment, when mortar was filled inside the angle steel columns, the temperature rise trend was similar to the cases without filling, but it was approximately 20-23% lower. When finish mortar was attached, the temperature remained at ambient temperature until 7-8 minutes and then started rising. The maximum temperature reduction of approximately 73% was observed when finish mortar was attached. Therefore, based on the fire behavior according to the fireproofing methods, it was shown that attaching finish mortar along with 10mm fireproofing coating can reduce the temperature rise by up to 82%.
(6) Comparing the thermal heat release rate and temperature distribution predicted by the FDS model with different input methods to the results of the experimental study, the simple model showed an error of more than 50%, while the pyrolysis model predicted the heat release rate with an error of 6% and the spatial temperature distribution with an error of 5-12%. The large error of the simple model can be attributed to the fact that it calculates the heat release rate based on the surface temperature of the combustible material reaching the ignition temperature according to the input heat release rate per unit area. If the surface temperature of the combustible material does not reach the ignition temperature due to the fire source, combustion does not occur, resulting in significant discrepancies. Therefore, the reliability of the pyrolysis model in predicting the heat release rate and temperature distribution was considered valid within an acceptable accuracy range.
(7) Using the validated FDS model, the fire behavior in the living room of a single-family house was evaluated based on the fire load. The fire went through the growth phase until 4 minutes after ignition, the fully developed phase from 4 to 11 minutes, and the decay phase afterwards. The maximum temperature at the fire source location was measured as 1052.46°C at a height of 1m. The maximum temperatures at the front and rear locations within the compartment were measured as 713.90°C to 777.28°C at around 8-9 minutes. In particular, at the front location with the opening, the temperature distribution was higher by 4-10% compared to the rear locations due to heat flow and flame ejection through the opening.
(8) Comparing the fire behavior in real fire incidents with the standard fire, it was observed that in the compartment without flashover, the temperature measured was up to 26% higher than the temperature range of the standard fire between 3 and 7 minutes. In the compartment with flashover, the temperature measured was up to 24% higher than the temperature range of the standard fire between 1.57 and 11.4 minutes. Therefore, although there were some temperature ranges beyond the standard fire temperature, the temperatures were distributed within the standard fire duration (945°C), indicating that steel columns would be safe with appropriate fireproofing. Based on this study, continuous research on the fire behavior of building components exposed to real fire incidents is necessary to promote performance-based fire-resistant design in Korea.