Page 718 - Dust Explosions in the Process Industries

Appendix 685

A.1.2.9
Flammability

The dusts are classified according to their ability to propagate a combustion wave when
deposited in a layer. Ignition is accomplishedusing either a gas flame or a glowing plat-
inum wire at 1000°C.The test sample is a 2 cm wide and 4 cm long dust ridge resting
on a ceramic plate. Ignition is performed at one end. The classifications are
0 Class 1. No self-sustainedcombustion.
0 Class 2. Local combustion of short duration.
0 Class 3. Local sustained combustion but no propagation.
0 Class 4. Propagating smoldering combustion.
0 Class 5. Propagating an open flame.
0 Class 6. Explosive combustion.
The numbers in brackets refer to a modified test procedure according to which 20
weight% diatomaceous earth is mixed with the powder or dust to be tested. By this
means, some materials that otherwisewould not propagate a flamebecause they melt may
show sustained flame propagation.

A.2
APPLUCABILITY OF EARLIER USBM TEST DATA

A.2.1
BACKGROUND

The U.S. Bureau of Mines in Pittsburgh,PA, developed a comprehensiveset of laboratory
test methods for characterizingignitabilityand explosibilityof dusts, and published a large
number of test data, which have been widely used throughout the world. The test appara-
tuses and procedures were describedby Dorsett et al. (1960).Test data for 220 agricultural
dusts were reported by Jacobson et al. (1961); for 314 dusts in the plastics industry, by
Jacobson,Nagy, and Cooper (1962);for 314metal powders,by Jacobson,Cooper, and Nagy
(1964); for 241 carbonaceous dusts, by Nagy, Dorsett, and Cooper (1965); for I75 chem-
icals, drugs, dyes, and pesticides, by Dorsett and Nagy (1968); and for 181miscellaneous
dusts, by Nagy, Cooper, and Dorsett (1968); that is, for 1445 dusts altogether.
In more recent years, alternative test methods have been developed, and there is a need
to indicate the extent to which the substantialamount of the earlier USBM data are com-
patible with more recent data, as for example those in Tables A.l, A.2, and A.3.

A.2 .2
MINIMUM IGNITION TEMPERATURE OF THE DUST CLOUD

The apparatus used was the original Godbert-Greenwaldfurnace, which is essentially
the. same apparatus as the Godbert-Greenwaldfurnace used for determining the data in
TableA.1.The eadier USBM data should thereforebe compatiblewith those in TableA.1.