Page 311 - Dust Explosions in the Process Industries

280 Dust Explosions in the Process Industries

the vertical experimental glass duct of 0.2 m x 0.2 m cross section and 2 m height by
low-velocity elutriation from a fluidized bed of 600 g of starch resting on a porous mem-
brane at the bottom of the system. The average vertical air velocity was on the order of
0.1 m/s. A battery of parallel vertical steel plates 0.5 mm thick was inserted across the
whole cross section of the duct when quenching distances were measured. Average dust
concentrations were determined from the dust mass lost from the fluidized bed as a
function of time and the airflow through the system. A laser tomography system was used
to control the homogeneity of the dust cloud.
Laminar burning velocities were determined from the measured flame speeds and
photographically estimated flame surface areas, as in the case of Kaesche-Krischer
(1959), but the applicability of this method to flame propagation in tubes is not obvious.
(See specific comment in Section 9.2.4.2 in Chapter 9.) To obtain proper laminar flame
propagation, it is necessary to avoid the buildup of fundamental-mode standing acoustic
wave motion in the duct. Such waves are easily generated by the gas expansion follow-
ing the initial flame and can subsequently interfere with the flame propagation. Proust
and Veyssiere solved this problem by fitting a special damping diaphragm at the open
bottom end of the duct (see Guenoche, 1964).
A series of photographs of the propagating laminar maize starch flame is shown in
Figure 4.15. Figure 4.16 shows the upwards laminar flame front velocity (duct closed at
the upper end) as a function of the dust concentration. The velocity was measured by
means of ionization probes. The maximum value of 0.63 m/s occurred close to the stoi-
chiometric dust concentration 235 g/m3. A corresponding laminar burning velocity of 0.27
m/s was deduced by assuming that its value normal to the flame surface was uniform
across the entire flame hemisphere. However, this assumption is not necessarily justified.

Figure 4.1 5 Photographic records of an upward propagating laminar flame in a 120 g/m' cloud oi
maize starch in air (from Proust and Veyssiere, 1988).