A set of experiments was conducted to expose different types of energized electrical cords for lamps, office equipment, and appliances to a developing room fire exposure. All of the cords were positioned on the floor and arranged in a manner to receive a similar thermal exposure. Six types of cords commonly used as power supply cords, extension cords, and as part of residential electrical wiring systems were chosen for the experiments. The non-metallic sheathed cables (NMB) typically found in residential electrical branch wiring were included to provide a link to previous research. The basic test design was to expose the six different types of cords, on the floor of a compartment to a growing fire to determine the conditions under which the cord would trip the circuit breaker and/or undergo an arc fault. All of the cords would be energized and installed on a non-combustible surface. Six cord types (18-2 SPT1, 16-3 SJTW, 12-2 NM-B, 12-3 NM-B, 18-3 SVT, 18-2 NISPT-2) and three types of circuit protection (Molded case circuit breaker (MCCB), combination Arc-fault circuit interrupter (AFCI), Ground-fault circuit interrupter (GFCI)) were exposed to six room-scale fires. The circuit protection was remote from the thermal exposure. The six room fires consisted of three replicate fires with two sofas as the main fuel source, two replicate fires with one sofa as the main fuel source and one fire with two sofas and MDF paneling on three walls in the room. Each fuel package was sufficient to support flashover conditions in the room and as a result, the impact on the cords and circuit protection was not significantly different. The average peak heat release rate of the sofa fueled compartment fires with gypsum board ceiling and walls was 6.8 MW. The addition of vinyl covered MDF wall paneling on three of the compartment walls increased the peak heat release rate to 12 MW, although most of the increased energy release occurred outside of the compartment opening. In each experiment during post flashover exposure, the insulation on the cords ignited and burned through, exposing bare conductor. During this period the circuits faulted. The circuit protection devices are not designed to provide thermal protection, and, thus, were installed remote from the fire. The devices operated as designed in all experiments. All of the circuit faults resulted in either a magnetic trip of the conventional circuit breaker or a ground-fault trip in the GFCI or AFCI capable circuit protection devices. Though not required by UL 1699, Standard for Safety for Arc-Fault Circuit-Interrupters as the solution for detection methodology, the AFCIs used had differential current detection. Examination of signal data showed that the only cord types that tripped with a fault to ground were the insulated conductors in non-metallic sheathed cables (12-2 NM-B and 12-3 NM-B). This was expected due to the bare grounding conductor present. Assessments of both the thermal exposure and physical damage to the cords did not reveal any correlation between the thermal exposure, cord damage, and trip type.
The effect of intumescent mat on post-fire performance of carbon fibre reinforced composites