Currently there are many buildings with systems intended to control smoke movement in building fire situations. Network computer models have been developed to analyze these systems for research and design purposes. This paper presents a general description' of such models, a discussion of validation techniques used to check the computer algorithms, and the results of full scale tests conducted in a pressurized stairwell at Champs Sur Marne, France to validate the basic assumptions of these computer models. Agreement between computer simulations and the test data was good for all tests analyzed. The appropriateness of using exclusively a flow exponent of 1/2 for smoke control design is reevaluated and is found to have only a slight effect on the results of a computer simulation.

Units: Dual