Smoke production in the full-scale room fire test IS0 9705 (Commonly referred to as the Room Corner Test) and in the Cone Calorimeter I S 0 5660 has been analysed for three sets of building products comprising a total of 28 products. The smoke production may be critical for the fire classification of surface products since some products produce large amounts of smoke in the room fire test even if they do not reach flashover within 20 min. Several smoke parameters in the Cone Calorimeter and the room fire test have been analysed. Good correlations have been obtained when the products are divided into two groups: products with more than 10 min to flashover in the room fire test and those with less than 10 min. These two time categories correspond to the two heat output levels in the room fire test 100 kW for the first 10 min and then 300 kW up to 20 min. For products with more than 10 min to flashover the average rate of smoke production and the total smoke production seem to be useful parameters for predictions of smoke release in the room fire test. Both parameters have good correlations between data from the Cone Calorimeter and the room fire test. For products with less than 10 min to flashover no parameter seems to give useful predictions. For all products evaluated together, the correlations are not so good, but the same regression lines as for products with more than 10 min might be used as a first rough estimate. In this case the total smoke production in the Cone Calorimeter could be used to estimate the total smoke production in the room fire test for different building products, independenf of their estimated time to flashover.
It is suggested that the average rate of smoke production and the total smoke production from the Cone Calorimeter is reported in addition to the mass-based specific extinction area. This will be helpful in predicting smoke release in the room fire test and will also make the data on smoke release analogous to those on heat release.