Small scale experiments and theoretical aspects of flame extinguishment with water mist

Research output: Book/ReportReport


The present study focuses on extinction of flames with water mist where surface cooling effects are neglected i.e. water mist as a total flooding system where the direct spray cannot reach and cool all burning items in the protected volume. The study includes a survey of the production and properties of water mist such as different types of nozzles and means to describe droplet size distribution. Properties of jets and sprays are discussed as well as water droplet movement, fall and
evaporation. Three different series of experiments were conducted with different hydraulic atomising nozzles. In the first series, droplet size distribution and water spray distribution measurements for the nozzles were conducted. The measurements showed that the droplet size distribution ranged from a Sauter Mean Diameter of approximately 35 - 85 µm, all dependent on the water pressure and the configuration of nozzles. The measurements show that considerably larger droplets are formed when individual nozzles are placed together. This can be explained by a coalescence effect when droplets from the sprays of the nozzles collide. The second series of experiments were done using a tubular propane gas bumer where water and propane were mixed prior to reaching the burner outlet. Based on the tests, the specific amount of extinguishing medium required (Required Extinguishing Medium Portion, REMP = m, / m, i.e. the ratio of the agent quantity to fuel quantity consumed) is given as a quantitative measure of the efficiency of the agent. The lower the REMP value, the more efficient the agent. The water pressure ranged between 40 and 80 bar which provided for droplet sizes with a Sauter Mean Diameter in the order of 35 pm. The results show that the amount of water needed for extinguishment by weight is between 1.2 - 2,2 times the amount of propane gas. The decrease in droplet sizes decreased the amount of water needed. Another observation was that the heat release rate of the fire is not affected until extinction
occurs. A REMP-value of 1,2 - 2,2 corresponds to a water content of 100 – 200 g/m3 protected volume which is in agreement with theoretical values. Finally, a series of tests were conducted in a 113 scale room using a propane gas fire. Parameters such as location of the fire, the location of the nozzle, water flow rate and the size of the room opening were varied. In these tests the water content needed was in close agreement with the values obtained from the REh4P experiments and the theoretical values. These tests also highlighted the problem of delivering the droplets to the fire. To achieve "total flooding" in an actual situation, nozzles covering the complete protected compartment, with additional nozzles under obstructions would be needed. To make droplets follow the air flows inside a room and behave more like a gaseous total flooding agent, requires droplets of a size in the order of 1 - 20 µm.


  • Petra Andersson
  • Magnus Arvidson
  • Göran Holmstedt
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Other Civil Engineering
  • Building Technologies


  • Water mist, Extinction
Original languageEnglish
Publisher[Publisher information missing]
Publication statusPublished - 1996
Publication categoryResearch

Publication series


Bibliographic note

Project No. 621-921, 610-922

Total downloads

No data available