著者

神 忠久

出版者

Japan Association for Fire Science and Engineering

雑誌

日本火災学会論文集 (ISSN:05460794)

巻号頁・発行日

vol.19, no.2, pp.1-8, 1970 (Released:2013-06-18)

被引用文献数

1

---

A smoke chamber is designed to perform the measurement of visibility through fire smoke (see Fig.3). The chamber is provided with an illumination apparatus and two kinds of sign. One of the signs is the placard and the other is the lighted sign which is backlighted with a projector. The brightness of the lighted sign can be controlled freely, and that of the placard can be set in four stages. Nearly white smoke is generated by heating filter-paper in an electric furnace.Experiments were performed to get the relation among the brightness of sign, the visual distance, and the extinction coefficient of smoke in the instant of obscuration threshold. In this case the distances the between observer and the object were 5.5 m, 10.5 m, and 15.5 m.The extinction coefficient (σ) in the instant of obscuration threshold may be theoretically given byσ≒1/V ln BE0/δckL ………(1)where V ; Visual distanceBE0 ; Brightness of signδc ; Threshold of brightness-contrast (δc≒0.01 under the condition of general illumination)L ; Intensity of external light (Illumination)k=σs/σ (Ratio depending on the nature of smoke, k≒1 for nearly white smoke)σs ; Mean light-scattering coefficientFrom Eq. (1), the extinction coefficient in the instant of obscuration threshold is logarithmically proportional to the brightness of sign for a given smoke (nearly white smoke), a given intensity of external light, and a given visual distance. If the dimensionless brightness (BE0/L) is constant, the relation between σ and V is given byσ · V ≒ const.The results of experiment are shown in Fig.4~Fig.12 The agreement between the results and the theoretical values calculated from Eq. (1) concerning the brightness of sign is good as shown in Fig.11. The visibility of the placard is about (2~4)/σ and that of the lighted sign is about (5~10)/σ.In the case of escape through real fire smoke, the visibility should be lower than that of experiment, because the effects of physiology and psychology must be considered.