REACTION TO FIRE CLASSIFICATION OF BUILDING PRODUCTS: ASSESSMENT OF THE SMOKE PRODUCTION: Hazard Assessment, ISO Fire Test Methods and Development of Empirical Smoke Prediction Models
MetadataVis full innførsel
The thesis presents a comprehensive study on assessment of smoke production properties of building products. Primarily the hazard related to the loss of visibility is considered. The starting point has been to investigate whether smoke prediction models can be developed to predict smoke test data from the full scale ISO Room Comer Test by the use of test data from the bench scale ISO Cone Calorimeter. The thesis outlines a theoretical base for fire parameters, then a set of fire parameters is developed. All fire parameters are then used in empirical approaches to develop smoke prediction models. The approaches are based on test data from 34 building products,and only models with a scientifically sound basis are accepted. By use of a found model,full scale smoked at a from pre-flashover fires are sufficently well predicted. The predictions are principally based on bench scale CO data. It is argued that bench scale smoke data might not have general suitability for such predictions.The study shows that the success of prediction depends on the elimination of the temperature dependency of the smoke production Then the generality of the full scale test data is considered This means verifying whether the smoke data are predominantly governed by the building products and less the fire scenario.The verification is done by comparison of the smoke production (generation) from the ISO Room Comer Test with corresponding data from three other enclosure tests. The verification validates smoke data from pre-flashover fires.This verification also implicitly validates the ISO Room Comer Test to be consistent with the fire scenarios the test is meant to cover.This leads to the conclusion that the ISO Room Comer Test can be used to assess smoke production from building products in growing fires. The relationship between bench scale and full scale data seems to disappear as a fire grows. This has been pointed out to be caused by complex full scale combustion conditions which need further investigation There are some indications that post-flashover smoke and CO data might possibly be predicted by reduced scale enclosure tests. Finally,the hazard related to the loss of visibility is considered The assessment relates to the occurrence of flashover. In pre-flashover fires the smoke hazard is identified to occur both outside and inside the plume.The lethal hazard outside the plume is identified to be a combination of high irradiance level and loss of visability, while lethal hazard inside the plume (fire eftluents) is identified to be a combination of loss of visibility and the initant effects of the fire eftluents.TheISO Room Comer Test is validated as applicable to evaluate the pre-flashover smoke hazard It is argued that a benchmark of smoke hazard occurs when the combustible changes from a moderately smoky product like wood to a highly smoky plastic based product. A proposal for how this could be implemented in a functionally based classification system is also given. The hazard assessment of post-flashover fires shows that the loss of visibility can be a significant hazard even when the fire atmosphere has a relatively low toxicity. It is argued that the hazard related to loss of visibility in post-flashover fires should provisionally be left out of classification.