Quantifying Smoke Generator Output for Full-Scale Hot Smoke Testing to Enhance the Reliability of Smoke Control Assessment

Full-scale hot smoke tests are widely used to evaluate smoke control performance; however, the amount of smoke released during testing is often determined empirically, resulting in inconsistent and non-reproducible evaluations. This study develops a measurement-based method to quantify smoke generator output using certified addressable photoelectric smoke detectors as a reproducible concentration reference. A detector-defined threshold of 15% obscuration is adopted to represent a practical and standardized smoke condition for engineering evaluation. A controlled smoke accumulation enclosure is constructed to achieve well-mixed conditions, supported by buoyancy from an ethanol burner and gentle mechanical mixing. The uniformity of the smoke field is verified through optical attenuation measurements, multi-height thermocouple data, and consistent detector responses. Under a constant driving pressure of 138 kPa, stabilized detector readings are used to quantify smoke generator output at three preset settings. The reported smoke generator output should be interpreted as a detector-based equivalent smoke output established under controlled and well-mixed conditions. The corresponding quantified average smoke output rates are 8.29, 12.57, and 15.44 m³/s for the low-, medium-, and high-output conditions, respectively. The proposed method provides a reproducible basis for defining smoke output in full-scale testing and enables comparison with plume entrainment correlations for engineering interpretation. By establishing a link between controlled smoke generation and theoretical smoke production, the framework supports the selection of appropriate test conditions for performance-based fire safety design. Overall, the approach improves the consistency and reliability of full-scale hot smoke tests in complex building environments.