A plunge-tunnel similar to what has been used to measure response time index
(RTI) of automatic sprinklers was utilized in the bench-scale tests. As a true thermal
response sensitivity index of a detector should be independent of surrounding
conditions that can be caused by different sets of fires, eight different plunge test
conditions were used to measure the detector response times. The eight conditions
consisted of two set temperatures, 197_C and 291_C, with four velocities, ranging
from 0.5 to 2.5 m/s, at each set temperature. An ideal sensitivity index of each
detector obtained through these measurements should be an invariant. Multiple tests
across the eight test conditions were conducted with one sample per each restorable
detector type.