METHODS FOR MELTING TO TEST CAST IRON. 363
prevent physical results being- obtained from any desired mixture. By melting in the crucible, we can closely tell the physical properties in respect to what the chemical elements would define it in the original state, when not affected by the sulphur, etc., in fuel, but not what it would be when remelted. Why this is so involves elements most essential for the founder to understand and are treated further in Chapter XLV.
Melting a mixture in a crucible with the expectation of obtaining tests to denote what the physical qualities of a regular cupola mixture would be, is impractical. These can be told with fairness by taking tests from the regular cupola. Small cupolas ranging from fifteen to twenty indies inside diameter can often be well used to test single brands or grades or mixtures not having over four different kinds of iron. As there are cases where vsome would like to use a small cupola for crucible melting also, I have studied to the point of combining the two, and as a result present the following original device or small cupola, as seen in Fig. 69, next page. This cupola can be erected in any out-of-the-way place, or by the side of a regular " heat " cupola, so that the flue A can be attached to head off the sparks, etc., when used as a cupola, without risk of setting anything on fire, should there be any danger of this; if not, then, the cover B could be dispensed with and the flame, etc., permitted to pass out at the top. B is a cover made of cast iron, and having prickers on the under side for the purpose of holding a daubing of clay to prevent the heat of the furnace burning the cover. Trie handle D is for convenience in lifting the cover on and off when desiring to change or takeus that