Modern Gasworks Practice

REFRACTORIES FOR GASWORKS PURPOSES 205 Two important empirical facts have been established, namely :— («) The refractoriness of a fireclay under load diminishes as the load increases in accordance with a definite law. (6) The sensitiveness of a fireclay to load is greater the greater the alumina content of the clay, and the sensitiveness is less the greater the silica content of the clay. Tn carrying out his tests Mellor employed a furnace as illustrated in Fig. 109, loading of the test piece being arranged for by means of a lever system. The refractoriness under load is determined in the following manner : The test piece is cut from the brick and ground to a rectangular block 3J incb.es long and 2 inches cross-section. This presents a surface of 4 square inches to the load. The final measurements of the piece are.taken with micrometer slide gauge. Much depends npon the careful preparation of this block. If the ends are not parallel, the weight will not act in a true vertical direc-tion, but tend to thrust sideways. A solid refractory support is bedded. at the bottom of tb e inner tu b e o f the furnace. Athin layer of powdered carborun- Fig. 109.—Apeakatus foe Testing Refkactories under Load. dum is spread over the top of this basal support to prevent the sticking of the test piece. Small Seger con.es, arranged in series of four, are fixed on the basal support round the test piece. A cube 2| inches square, made of similar material to the bottom support, is placed on top of the test piece. This is followed by a cylinder, say, magnesite—and finally a carbon rod mounted on. an. iron ring. The rate of heating is controlled by means of variable transformers and resistances. The current used varies from 8 to 10 k.w. As the temperature of the furnace rises, the thermal expansion of the brick causes the lever arm at first to rise, and later on the lever beging to fall. The rate at which. the deformation proceeds depends on the character of the test piece. When the squatting has reached the limit decided upon, the current is switched off, the weights, lever, and supports removed, and the cones observed. The test piece is left in the furnace until its temperature has dropped. As a result of a number of tests, Mellor concludes that in general silica bricks stand up under load very much. better than firebricks made from fireclays ; but the finer grained silica bricks do not stand under load so well as the coarser grained silica bricks. Indeed, very fine grained silica bricks show as large a difference between the normal refractoriness and the refractoriness under load as bricks made from fireclays.