Numerical simulations of concealed residential sprinkler head activation time in a standard thermal response room test

Abstract

It is common for sprinkler heads in residential buildings to be of the concealed type. Key parameters for the thermal sensitivity of sprinkler heads are the response time index (m½s½) and conductivity factor (C factor, m½s−½). BS 9252:2011 and BS EN 12259-14:2020 outline test methods to determine whether sprinkler heads achieve adequate thermal sensitivity, stipulating that a room test be undertaken for concealed heads. In carrying out this test, neither the response time index nor C factor is defined. This paper uses computational modelling tools Fire Dynamics Simulator and B-RISK to simulate the room test. In running these simulations, maximum values for the ‘effective’ response time index and C factor needed to pass the room test can be estimated, assuming concealed heads can be represented as equivalent pendent heads. Initial deterministic simulations are used to calibrate results between Fire Dynamics Simulator and B-RISK. Monte Carlo modelling is then utilised within B-RISK, with variable parameters for the response time index and C factor (C), to estimate the relationship between the two parameters necessary to pass the room test. As a result, it is proposed that this relationship can be represented using a simple, empirical power law equation of response time index = 100 (5.4–C)2/3, where C < 5.4. Practical application: The results indicate that the minimum RTI and C factor values needed to pass the room test are greater than those needed to pass wind tunnel testing methods. In observing that equivalency is not demonstrated by the room test, and by defining the RTI/C factor relationship needed to pass the test, this paper provides fire safety engineers with amended values for concealed heads to be adopted in future assessments. In the absence of any detailed specification for sprinkler heads, it is recommended that an RTI of 290 m½s½ and a C factor of 0.5 m½s−½ may be applied.

Using Probabilistic Zone Model Simulations to Investigate the Deterministic Assumptions of UK Residential Corridor Smoke Control Design

“…B-RISK incorporates an underlying zone model which is used to calculate fire dynamics, smoke dispersion and temperature throughout enclosures, with each enclosure divided into two (upper and lower) gas layers. The fundamental equations are implemented as a system of differential equations which are solved to give outputs such as layer height, visibility, and layer temperature [19]. B-RISK also has the capability to run multiple iterations of simulations based on probabilistic inputs.…”

“…B-RISK also has the capability to run multiple iterations of simulations based on probabilistic inputs. The relative simplicity of a zone model means that B-RISK can run simulations in a short timeframe, with the capability to run thousands of simulations without any substantial reliance on computing power [19].…”

“…To examine this topic, the paper details probabilistic simulations of a single exemplar residential common corridor arrangement using the zone modelling software B-RISK [10], where the Monte Carlo method has been applied. Probabilistic distribution functions (PDFs) from previous research have been adopted to inform the input parameters of the fire, including the growth rate (t-squared) [11], maximum HRR [11], soot yield [12][13][14][15][16], etc., as well as for the sprinkler performance in the apartment of fire origin (e.g., the reliability [17,18] and thermal sensitivity [19]), and the behavioural characteristics of building occupants (the pre-evacuation time [20,21] and the apartment door open time [5]). From the Monte Carlo zone modelling simulations, an output distribution for the smoke clearance time in the corridor is determined and then compared to the estimated clearance time of an equivalent set of simulations undertaken using the deterministic assumptions of SCA guidance.…”

“…Uniform distributions have been used for the response time index (RTI) and conductivity factor (C factor). These distributions have been estimated from the work of Hopkin and Spearpoint [19], who used probabilistic modelling to estimate the relationship needed between RTI and C factor to pass the BS 9252:2011 [45]/BS EN 12259-14:2020 [26] thermal response room test for concealed residential sprinkler heads. This was then compared to previous experimental data for concealed sprinkler heads from Annable [46] and Yu [47].…”

Експериментальне Визначення Інерційності Спрацьовування Спринклерних Зрошувачів Автоматичних Систем Водяного Пожежогасіння

“…У статті [13] розглянуті питання застосування інструментів Fire Dynamics Simulator та B-RISK для комп'ютерного моделювання випробувань спринклерних зрошувачів прихованого типу в житлових приміщеннях. В результаті досліджень отримана емпірична ступенева залежність, що пов'язує індекс часу відгуку RTI та С-фактор.…”