Page 123 - Dust Explosions in the Process Industries

96 Dust Explosions in the Process Industries

It is felt that this correlation still holds good as a first approximation. For example, the
data in Figure 1.109 are reasonably well accounted for in Figure 1.110. However, as
reviewed in Section 9.3.7.5 in Chapter 9, the length-to-diameter ratio of the duct is a more
basic parameter than just the duct length.
Aellig and Gramlich (1984) studied the influence of various geometrical features of
the vent duct design, in particular the details of the coupling between vessel, vent, and
duct, and the geometry of the bends. They proposed an overall correlation that looks sim-
ilar to that of Walker in Figure 1.110, but the ratio of duct volume to vessel volume was
used as parameter instead of the duct length.
Pineau (1984a) conducted a comprehensive series of experiments with explosions of
wheat flour and wood dust in vented vessels of 0.1 m3 and 1 .O m3 volumes connected to
vent ducts of various diameters and lengths, with and without bends. Some experiments
were also conducted with larger vessels of volumes 2.5-100 m3 vented through ducts.
In general, the main trends observed in the small-scale experiments were confirmed for
the large-scale ones, and it was recommended that vent ducts be as short as possible and
have a minimum number of sharp bends.
More recently, Lunn, Crowhurst, and Hey (1988) conducted a comprehensive theo-
retical and experimental study of the effect of vent ducts on the maximum explosion pres-
sure in vented vessels. Experiments were performed in a 20 liter vessel (same experiments
as Crowhurst, 1988), and in a large-scale 18.5 m3 vessel. Figure 1.11 1 shows the 18.5 m3
vessel fitted with a straight duct, whereas Figure 1.112 shows the same vessel during a
coal dust explosion with a 90" bend at the end of the duct.

Figure 1.1 11 18.5 m3 vented explosion vessel connected to a straight vent duct (Courtesy of Health
and Safety Executive, United Kingdom). For a much clearer picture, see Color Plate 1.

In general, the trends of the experimental data for the five dusts-coal, aspirin, toner,
polyethylene, and aluminum-used by Lunn et al. were similar to that in Figure 1.109.
The maximum explosion pressure in the vessel increased systematically with duct length
and the length-to-diameter ratio of the duct. The theoretical analysis generally confirmed
this trend and yielded predictions in reasonable agreement with the experimental data,
although some discrepancies were found. The theory developed by Lunn et al. may serve
as a useful tool for estimating the influence of various types of vent ducts on Pred. The K,,