Page 452 - Dust Explosions in the Process Industries
P. 452

Ignition of Dust Clouds and Dust Deposits  4 7 9

                 Gubin and Dik (1986) developed a.  mathematical model assuming that the oxidation
               occurred as a heterogeneousreaction between oxygen from the gas phase and the parti-
               cle surface. They further assumed that the spark discharge initially generated a certain
               quantity of heat located within a narrow channel in the spark gap. The heat drain from
               the channel to the surroundingswas assumed to occur essentially by conduction,radia-
               tion and convection being neglected. The basic heat balance equation was of the same
               form as that derived by previous workers. As in the case of other workers, the ignition
               criterion was the establishment of self-sustainedlaminar flame propagation in the dust
               cloud. It would appear that Gubin and Dik may not have been aware of the other inves-
               tigations mentioned previously.


               5.4
                   NlTlON OF DUST CLOUDS BY HEAT
                FROM MECHANICAL RUBBING, GRINDING,
               OR IMPACT BETWEEN SOLID BODIES


                5.4.1
                BACKGROUND

               Whether or not metal sparks or hot spots from accidental impacts, rubbing operations,
               and the like between solid bodies can initiate dust explosionshas remained a controversial
               issue for a long time. Many attempts have been made at resolving the puzzle by analyzing
               past accidents with the objective to identify the ignition sources.A summary with ref-
               erence to the grain, feed, and flour industry is given in Table 5.6.


                Table 5.6  Percentage of dust explosions in the grain, feed,  and food indust-y assumed to be initi-
                ated by “friction sparks“ or unknown sources











                Source: Pedersen and Eckhoff, 1987.

                 As can be seen, “friction sparks” are claimed to play a significant part. If one further
               takes into account that it is often tacitly implied that a substantial part of the “unknowns”
               may have been initiated by some untraceable source, such as metal sparks and electro-
                static discharges,the friction spark becomes the most suspect of all the potential ignition
                sources.
                 The situations in which metal sparks and hot spots can be generated in an industrial
               process plant fall into two main categories. The first is grinding and cutting operations,
   447   448   449   450   451   452   453   454   455   456   457