Page 442 - Dust Explosions in the Process Industries

Ignition of Dust Clouds and Dust Deposits 409

.
I

I
i
/ I I I I I I
0 50 100 150 200 250 300 350
NOMINAL DUST CONCENTRATION rq/rn31

Figure 5.1 5 Probability of electric spark ignition of clouds oflycopodium in air as a function of dust
concentration, for three different spark energies. Spark discharge duration is 5- 10 ps. External circuit
resistance is 0.01 a. Circuit inductance is 2 pH. Spark gap length is 2.7 mm (From Eckhofi, !970).

ignition, and the mechanical separationof the dust from the hot spark kernel by the blast
wave, which counteractsignition.The results in Figure 5.14 even show a drop in the fre-
quency of ignition as the spark energy increasesfrom 1J to 3 J. Eckhoff and Enstad (1976)
demonstrated that the blast wave from capacitive discharges of durations on the order of
1 ys and energies of 100-200 mJ, could push a 4 x 5 mm paper penclulum, supportedby
thin threads, an appreciable distance from the spark. The results are given in Table 5.5.

Table 5.5 Displacement distances of a 4 x 5 mm paper pendulum due to blast waves from capac-
itive spark discharges (initial distance between paper and spark gap Is 1 mm)

Source: Eckhoff and Enstad, 1976.

Table 5.5 clearly demonstrates that as the spark energy increased beyond 100 mJ, the
displacement of the paper by “short” sparks was appreciable. On the other hand, as the
spark energy decreased below 10 mJ, the displacement was practically negligible even
for the “short” sparks, which means that the minimum ignition energy may not neces-
sarily increasewith decreasingdischarge durationin the range of low spark energies below
10 mJ. This was confirmed by the results of Parker, discussed later. However, first, the
theoretical analysisby Enstad (1981) of the interferenceof the blast from a “short” spark
discharge with the surrounding dust particles is outlined. Enstad made the following
assumptions: