Page 372 - Dust Explosions in the Process Industries
P. 372

Propagation of Flames in Dust Clouds  34 1


               Table 4.13  K,,  values measured for clouds of maize starch dust in air in different closed vessels and
                arranged according to vessel volume: Ks, = (dP/dt),,,  V’l3























               *Arithmetic mean values, 11% moisture in starch.
               Source: Extended and modified version of  table from Pu,  1988.


                 It should be mentioned that Bradley et al. (1988) were able to express K,,  in terms of
               a “massburning rate” and the initial and final pressure. The K,,  concept was then defined
               by equation (4.85).


               4.4.4
               TURBULENT FLAME PROPAGATION IN PARTLY OR FULLY
               UNCONFINED GEOMETRIES


               The important work of Tamanini (1989), and Tamanini and Chaffee (1989) is discussed
               more extensively in Chapter 6 on venting of dust explosions. In the present context, it
               should only be briefly mentioned that explosion experimentswere conducted  in a 64 m3
               vented vessel ah a series of different,known turbulence intensitiesat the moment of igni-
               tion. The turbulenceintensitieswere measured by means of a bidirectional impact probe.
               For a given dust, dust concentration, and vent characteristics,the maximum pressure in
               the vented explosion increased systematically with increasing initial turbulence inten-
               sity in the experimental range 2-12  ds.
                 Hayes et al. (1983) investigatedthe influence of the speed of four shroudedaxial fans,
               mounted above the channel floor, on the dust flame speed in a horizontal channel of 1.5 m
               length and 0.15 m x 0.15 m square cross section, open at both ends. A cloud of dried
               wheat flour of mean particle size 100 pm was produced in the channel and ignited by a
               propane/air flarme while the fans were running. Some results are shown in Figure 4.48.
                 It was anticipated that the flame speed would increase markedly with fan speed, and
               this was also observed up to a fan speed of about 1500 rpm. However, as the fan speed
               was increased further, the flame speed exhibited a marked decrease, to about 3000 rpm,
               beyond which ignition of the dust cloud by the propane flame was no longer possible.
   367   368   369   370   371   372   373   374   375   376   377