An Evaluation of Methodologies for Determining Delay Times in the Cone Calorimeter Fire Test

Abstract

Abstract An important step in calibrating a cone calorimeter apparatus is the determination of gas delay times. Gas concentration measurements must be made simultaneously in time with the main combustion events and processes (i.e., ignition, changes in mass loss rate, smoke production, and observed phenomena) to produce accurate results from the test. The calibration methodology prescribed in ASTM E1354-14e1 does not specify direct measurement of the transit time of gases through the apparatus or the response time of the individual analyzers, instead estimating total delay time by relating the thermal lag in the stack thermocouple to the transient response of the gas analyzers. Other methods to account for gas delay times are used in practice, leading to varying opinions on the method that is most suitable for oxygen consumption calorimetry. Furthermore, studies have shown that analyzer delay times are not consistent test-to-test, but depend on both the characteristics of the gas analyzer and sampling system (gas transit time) and the rates of production or consumption of gaseous species during a particular test (analyzer response time). In the present work, a methodology was proposed for measuring analyzer delay times by injecting a gas mixture of known concentration into the cone calorimeter exhaust stream. Delay times were computed using various methods, including ASTM E1354 and gas injection, and were evaluated with a series of cone calorimeter tests on various materials. Gas delay times determined by the ASTM E1354 method were found to produce inconsistent results for the cone calorimeter used in this study; results were significantly improved when alternative criteria were applied to the method. The square wave method was found to produce very good results for specimens with heat release rates greater than 3 kW; however, delay times in carbon monoxide production were not well represented. The gas injection method was found to produce excellent results, closely tracking oxygen consumption, carbon monoxide production, and carbon dioxide production in time, and as a result, the derived heat release rate coincided with observed events.