Page 376 - Dust Explosions in the Process Industries

Propagation of Flames in Dust Clouds 345

of fairly large-scale experiments with obstacle- and jet-induced turbulence. It has been
suggested, for example, by Nagy and Verakis (1983), that there may be similarities
between the influenceof turbulence on gas and dust explosions.One of the first systematic
comparative studies of turbulence influence on dust and gas explosions was conducted
by Bond, Knystautas, and Lee (1986). They concluded that the relative burning rate vari-
ations caused by turbulence were equal in a 300 g/m3maize starch-in-aircloud and in pre-
mixed 7.5 vol% methane-in-air. However, they also emphasizedthe need for further work.
Pu (1988) and Pu et al. (1988) made further comparison of turbulent flame propaga-
tion in premixed methane in air and in clouds of maize starch in air, in identical geome-
tries and at identical initial turbulence intensities. The experiments under turbulent
conditions were conducted in closed vertical cylindrical vessels of 190mm diameter and
length either 0.91 m or 1.86 m. All experiments were conducted with initial turbulence
generatedby the blast of air used for dispersing the dust. The influence of ignition delay
on the flamepropagation and pressure developmentwas studied.In the gas experiments,
the initial turbulencewas generatedby a blast of compressed methaneh, from the same
reservoir as used for the compressed airfor dust dispersion in the dust cloud experiments.
In some experiments,a battery of concentricring obstacleswere mounted in the tube for
studying the influence of the additional turbulence generated by the expansion-induced
flow of the unburned gas or dust cloud past the obstacles.
Acomparableset of Yi Kang h’s results are shown in Figures 4.52 (gas) and 4.53 (dust).
On average, the combustion of the gas is twice as fast as that in the dust cloud. The lam-
inar burning velocity of 550 g/m3maize starch in air, as determinedby Proust and Veyssiere
(1988), is about 0.20 ds. Extrapolation of Zabetakis’s (1965) data for methane in air to
5.5 vol% methane gives lower values, in the range of 0.15 m/s or less. It is thereforeclear

- -
E E -
I
I-
z c
0 z
n 0 L1:
LL
g 15 LL 1.5 -
4 Y
x
6
LL LL
-I
c
0
0 c
I- 8 -
=; 10 c 1.0
m c
0
m
_1 __I
w
VI
E W
VI m
> Y
0.5 -
E
Y
u
z
4 c
0 0 05 0.10 0.15 0.20 0.0 0.1 02 0.3 04
TIME FROM IGNITION 151 TIME FROM IGNITION Is1
Figure 4.52 Pressure rise and flame front loca- Figure 4.53 Pressure rise and flame front loca-
tion during cornbustion of 5.5 vol% methane/air tion during combustion of 550 g/m3maize starch
in a 1.86 m long closed vertical tube of diameter in air in a 1.86 m long closed tube of diameter
790 min, as a function of time, under the influence 190mm, as a function of time, under the influence
of obstacle-induced turbulence. Three different of obstacle-induced turbulence. Three different
ignition delay times z; are shown, and ignition is ignition delay times z; are shown, and ignition is
at the tube bottom (From Pu, 1988). at the tube bottom (From RI,1988).