The alterations introduced are chiefly directed towards obtaining a light as steady as possible from a comparatively small generator, and, secondly, towards the entire elimination of smell, which obviously is far more serious in a lecture hall than, for instance, on a motor car. The generators in most common use may be divided into two classes, i.e. those on the gasometer principle in which the carbide is gradually lowered into the water, and those in which the water is allowed slowly to gain access to the carbide. A good example of the former is perhaps that made by Messrs. Moss of Birmingham, though there are several others equally good, and clear and explicit directions for working should be supplied by the makers. The Moss Generator (Fig. 5) consists of a tall iron vessel A fitted with a gas tap at bottom, this communicating
with a vertical iron tube within the vessel. Into this container fits the inner bell or container B, divided internally into two concentric portions entirely separated from each other, but connected by the pipe P P and the tap T.
A guide inside the bell encircles the iron tube in the outer tank and prevents rotation. Into the inner portion fits again the carbide-container (shown separately on the left), which is locked when in place by giving it a half turn, when a hook inside the bell engages with the lower edge of the carbide container and prevents it from falling.
The carbide container is fitted with a series of shelves, and the contents of a 2 lb. tin of carbide should be roughly divided among them; there is no need to make any accurate division. The carbide used should be that known as ½ inch mesh, and should be pure. That described as 'chemically' treated is apt to give trouble by over-generation in these gasometers and should be scrupulously avoided.
The carbide having been placed in the receptacles, these should be closed by means of the loose flap and the whole pushed into the bell and secured.
Water should be poured into the outer vessel up to a mark on the iron tube, and the bell placed in position. The lower tap being then turned on and the upper one closed, air from the outer portion of the bell can gradually escape by means of the iron tube and lower tap, and the bell gradually sinks by its own weight until it is on the bottom, but still
no water can reach the carbide, the air imprisoned in the inner portion of the bell effectually excluding it.
The lower tap should now be connected by means of india-rubber or flexible metallic tubing to the burner in the lantern (of which more anon), and the upper tap on the generator turned on, the tap or taps on the burner being likewise opened. The air from the inner portion of the bell can now escape by the pipe P P into the outer part, and thence through the iron tube, and out through tubing and jet, and as it does so water will rise in the interior and attack the carbide.
In a few moments the burner can be lit; but the gas, being generated far in excess of requirements, and filling both the inner and outer portions of the bell faster than it can escape, lifts the latter until the carbide is entirely out of the water, when in a few minutes generation ceases.