Page 661 - Dust Explosions in the Process Industries

628 Dust Explosions in the Process Industries

area. By introducing a vent opening efficiency factor 0 < EF<1, Faber defined the rela-
tion between the vent area for ideal covers,AE,and the area actually required,A,, as A, =
AE/EF.Tamanini (1996~)presented a model of the effect of the inertia of vent covers, on
the efficiency of the venting process, based on a simplified lumped-parameter approach.
A further dimension of complexity is added to the venting problem if the initial pres-
sure (or temperature) deviates from atmospheric.Results from venting ofdust explosions
in air of elevated initialpressure were reported by Siwek, Glor, and Torreggiani (1992).
In dust explosion venting, maintaining the integrity of the enclosureis not the only con-
cern. Venting implies that both pressure waves andflames are emitted into the surround-
ings; and this may present a hazard, depending on the size of the emitted flame and the
magnitude of the blast wave. Crowhurst, Colwell, and Hoare (1994) and Crowhurst et al.
(1995) reported from a series of large-scaleventing experiments (40 m3and 20 m3)where
external blast waves and flame lengths were determinedand compared with existing empir-
ical correlations. Schumann and Rastogi (1995) presented further data of lengths and
velocities of the flames expelled from the vent of a 1 m3vessel in the case of dust explo-
sions in the vessel. Peng et al. (1994) developed a labyrinth-type flame arrester for miti-
gating flame and pressure and pressure effects from dust explosion venting. Experiments
with a 2.7 m3vented enclosuresuggested that this arrester concept works satisfactorily for
Stl dusts. The maximum explosion pressures in the vented enclosure, with the arrester
mounted in the vent, were only a few percent higher than the pressures without the arrester,
whereas both flame and pressure effects were substantially reduced. As a part of an exten-
sive program of large-scale vented dust explosion experiments, Crowhurst et al. (1996)
investigated the characteristicsof explosionpressures measured around complex structures
near the vented explosions.They were able to suggest some empirical equations for esti-
mating expected pressures in various locations relative to the vented enclosure.
Forcier and Zalosh (2000) reviewed various correlationsof experimentaldata that have
been developedfor estimating externalpressures to be expected from vented dust and gas
explosions.In addition, they explored the possibility of applying approximate spherically
symmetric and elipsoidal blast wave models.The results indicatethat the simplifiedmodels
can produce predictions in good agreement with the more complex data correlations.
Harmanny (2001) reexamined some existing guidance on blast emission into the sur-
roundings by vented dust explosions, by taking into account a well-established rela-
tionship predicting blasts from vented underground ammunition storage explosions.
Existing guidance tends to overestimate the near-field blast effects and underestimate
the far-field effects. The predictions according to the revised approach are in con-
siderably better agreement with experimental data than predictions by the existing
guidance.
Holbrow, Hawksworth, and Tyldesley (2000) described a research project studyingthe
effects of thermaE radiation from vented dust explosion flames. The objective was to
establish critical borderlines around a dust cloud fireball, beyond which the thermal
radiation would no longer be a risk to people. Dust explosions were created in a large
vented vessel, using six dusts and a range of different experimentalconditions,and the
thermal radiation pulse from the fireball expelled from the vessel was measured at var-
ious distances from the fireball.
Li et al. (1994) presented a new version of the @pipe, which was first developed in
Norway (see Section 1.4.6.6 in Chapter 1),in which the conventionalbursting diaphragm
had been replaced by a reusable vent cover. This solution is of interest in situations