Provided the cylinders are hot the wood is thoroughly charred in 21--3* hours. The gas and tar from the wood passes through a c d and a into the fire, which is found greatly to economize fuel, and to be the readiest means of ascertaining when the charring is properly and thoroughly done. This is shown by the flame which issues from the pipe leading into the fire becoming of a violet tint, indicating the formation of carbonic oxide. Aa soon as this is observed, the doors of the cylinders are opened, the alips are hoisted out of them by chain tackle and blocks, and lowered into large iron extinguishers having close-fitting lids, in which they remaiu for half a day, the charcoal is then diacharged into coolers, large cylindrical cases of sheet-iron fitted with lids, and aent to the charcoal store. Wood yields about 25 per cent. of charcoal. Charcoal is occasionally made in large pita, as may be seen described in various works on chemistry and metallurgy. Pit charcoal is preferred for some explosive compositions, but why it is difficult to say.
It is of the highest importance that the charring of the wood should always be conducted as nearly as possible at the same temperature, for the chemical composition of the charcoal, and the temperature at which it will ignite, is undoubtedly effected by the temperature at which it has been charred. Charcoal prepared at a low temperature is softer, more inflammable, and contains more gaseous elements than charcoal prepared at a higher heat, and the gunpowders made from these charcoals would be similarly affected. At many gunpowder manufactories, pyrometers to ascertain the temperature of the cylinders are in use.
The fitness of charcoal for gunpowder depends on its chemical composition, which is indicated by its physical qualities. If properly made it should be jet-black in colour, its fracture should show a clear velvet-like surface, and it should be light and sonorous when dropped on a hard aurface. Underburnt charcoal, that is, charcoal prepared at a very low temperature, is at once known by its reddish-brown colour ; ovorburnt charcoal, by its hardness and density. The former is greatly more inflammable than the latter, charcoal prepared at a temperature of 260° (500° F.) being readily ignited at a temperature of 338° (640° F.), while charcoal prepared at 982° (1800° F.) requires a temperature nearly double the laat to inflame it. Underburnt charcoal has found favour for some small-arm powders. It certainly appears to render the powder more inflammable, and consequently quicker, but it has the disadvantage of being more hygroscopic than denser charcoal, and of rendering the powder therefore more liable to suffer damage from damp.
The analysis of some specimens of gunpowder charcoal showed :— Proust, who paid special attention to the study of gunpowder, tried a series of experiments with mixtures of saltpetre and equal quantities of the charcoals of various woods, which he burnt together, measuring the quantity of gas evolved. His experiments have been repeated by some English chemists with very interesting results, of which the following is an abstract. 12 gr. of charcoal mixed with 60 gr. of saltpetre gave the following proportions of gas in cub, in. :—
The saltpetre as imported is known as grough saltpetre; it contains various impurities, the principal of these being sulphate of soda, chloride of sodium, sand, and moisture. To eliminate this, an apparatus similar to Fig. 594 is employed. Here A is a boiler, B a filtering apparatus, C a cooler, D drainer, E washing vat, F F liquor tanks, G evaporating pans, H filter.
About two tons of grough saltpetre is placed in a large open copper pan A, and about 270 gal., of water are added to it. Over the bottom of each pan is placed a false bottom of iron, perforated with holes 1 in. in diameter, to allow the sand and insoluble impurities to fall. In about two hours the saltpetre is dissolved and the solution boiling. The false bottoms are pulled out just before the solution begins to boil, and the scum removed from the surface. The solution is allowed to boil until no more scum rises ; the copper is then filled up with cold water, again boiled briskly for a few minutes, after which it is allowed to cool.
When the temperature of the solution has fallen to about 93° (200°F.), with a specific gravity of about 1.53, a hand-pump is lowered into the copper, and the solution is pumped into a wooden trough leading to another larger one termed the supply trough, furnished with six holes in the bottom, beneath which the filtering bags are suspended, Wooden plugs are provided for these holes, so that if the bags become clogged the flow of solution may be stopped.
The bags, made of dowlas, are suspended on iron hooks underneath the holes in the supply trough. They are always rinsed with hot water before the filtering commences, and require occasionally a little poured over them to prevent the formation of crystals. Occasionally a solution is found to contain so much organic impurity that it will not run through the filters. In this case a little glue, about 1 lb. to 2 tons of saltpetre, is added to the solution in the copper, which has the effect of throwing up a great part of the impurity as a scum, which can be removed before the liquor is pumped out. As soon as it is all removed the pumps are pulled up, and the coppers, if necessary, cleaned out. A wooden trough placed directly underneath the filtering bags receives the solution as it runs from them, and conducts it directly into the cooler or crystallizing cistern, a large shallow flat trough of sheet copper about 12 ft. long, 7 ft. wide, and 1 ft. deep. By the time the solution runs into it the temperature will have fallen to between 82° and 88° (180° F. and . 190° F.). As the temperature continues to fall the excess of saltpetre crystallizes out ; the coolers must be kept in constant agitation, to oause it to deposit the salt in the form of flour or minute crystals. This is effected by constant stirring by means of a large wooden hoe, with which also the flour is drawn to the side of the cooler, to be shovelled out with a copper shovel. As it is renioved it is first thrown on to an inclined board or drainer, to allow the excess of liquor to run back into the cooler. It remains on the drainer for some minutes, after which it is transferred to the washing vat. The crystallizing process may be very materially hastened by artificial cooling.