Page 437 - Dust Explosions in the Process Industries
P. 437
404 Dust Explosions in the Process Industries
5.2.4.4
Self-Ignition in Dust Deposits in Bag Filters in Steel Works
This problem was studied by Marchand (1988).Two specific cases were discussed to illus-
trate how hot spots and smoldering combustion can develop in fabric filter plants in steel
works. The cause of accumulation of deposits of very fine dust fractions in the clean sec-
tion of some filters and the various possibilities of ignition were analyzed. The dusts in
question contained a large fraction of combustible material, including carbon, various
organic compounds, and metallic iron. The typical ignition sources were burning metal
droplets expelled from the molten metal and conveyed to the filter.
5.3
IGNITION OF DUST CLOUDS BY ELECTRIC SPARK
DISCHARGES BETWEEN TWO METAL ELECTRODES
5.3.1
HISTORICAL BACKGROUND
Holtzwart and von Meyer (1891) were probably the first scientists to demonstrate that
dust clouds could be ignited by electric sparks. They studied the explosibility of brown
coal dusts in a small glass explosion vessel of 50 cm3capacity, fitted with a pair of plat-
inum electrodes, between which an inductive spark could pass.
A few years later Stockmeier (1899), who investigated various factors affecting the
rate of oxidation of aluminum powder, was able to demonstrate that aluminum dust,
shaken up in a glass bottle, ignited in the presence of an electric spark.
Since the publication of these pioneering papers, numerous contributions to the pub-
lished literature on the spark ignition of dust clouds have been produced. Indeed, they have
confirmed that ignition of dust clouds by electric discharges is a real possibility and the
cause of many severe dust explosionsover the years, in mines as well as in industrialplants.
5.3.2
THE OHMIC RESISTANCE OF A SPARK CHANNEL
BETWEEN TWO METAL ELECTRODES
Ohm's law can be applied to a spark channel just as to any other current-carrying con-
ductor. However, the resistance per unit length of channel is not a constant but depends
on the extent to which the gas in the gap between the electrodes is ionized. This in turn
depends on the energy dissipation in the gap per unit time, which determines the tem-
perature in the ionized zone. If equilibrium has been established, one would, for a given
gas at a given temperature and pressure, expect a consistent relationship between the gap
resistance per unit length and the current flowing through the gap. This has been inves-
tigated for air at atmospheric pressure and normal temperature by several workers, as
summarized in Figure 5.11.