It is for this reason (among others) that where very exact results are to be obtained and a large number of pieces are to be tempered, fluxes, heated to the required temperature, are very often employed.
Color tempering is conducted in different ways. In a muffle, in heated sand, with hot pieces of flat iron, and in boxes heated to the requisite temperature in an oven, the temperature being indicated by a pyrometer or heat-gauge. The articles to be tempered remain in the oven a length of time determined by experiment or experience, these being influenced by the size and substance, or thickness, of the pieces.
A muffle is a tube or cylinder receiving its heat from the outside and open at the end or ends to receive the steel. Where tempering is carried on continuously the muffle is kept in the fire, although it is claimed by many that better results are obtained by removing it from the fire when heated. It is obvious that if the muffle is heated evenly the steel will temper most evenly by being held in the centre of the muffle, or the piece may be revolved and moved endways in the muffle in order that the steel may heat evenly. The tempering should always proceed slowly, otherwise the heat may not have time to penetrate the steel to the centre, the outside tempering more quickly, thus the tool will be weak because of the undue hardness of the interior metal. Furthermore, protruding edges, or slight sections of the steel, may reduce to the required temper before the main body of the steel, which induces either serious weakness of the insufficiently tempered part, or softness in the thin sections, providing that the steel is kept long enough in the muffle to temper the main body to the proper degree.
In heating steel to harden it, especial care is necessary, particularly when the tool is one finished to size, if its form is slight or irregular, or if it is a very long one, because unless the conditions both of heating and cooling be such that the temperature is raised and lowered uniformly throughout the mass, a change of form known as warping will ensue. If one part gets hotter than another it expands more, and the form of the steel undergoes the change necessary to accommodate this local expansion, and this alteration of shape becomes permanent. In work finished and fitted this is of very great consideration, and, in the case of tools, it often assumes sufficient importance to entirely destroy their value. If, then, an article has a thin side, it requires to be so manipulated in the fire that such side shall not become heated in advance of the rest of the body of the metal, or it will become locally distorted or warped, because, though there may exist but little difference in the temperature of the various parts, the more solid parts are too strong to give way to permit the expansion: hence the latter is accommodated at the expense of the form of the weakest part of the article.
Pieces, such as long taps, are very apt to warp both in the fire and in the water. In heating, they should rest upon an even bed of coked coal, and be revolved almost continuously while moved endways in the fire; or when the length is excessive, they may be rested in a heated tube, so that they may not bend of their own weight. So, likewise, spirals may be heated upon cylindrical pieces of iron or tubes to prevent their own weight from bending or disarranging the coils.
Experiments have demonstrated that the greater part of the hardness of steel depends upon the quickness with which its temperature is reduced from about 500° to a few degrees below 500°, and metal heated to 500° must be surrounded by a temperature which renders the existence of water under atmospheric pressure impossible; hence, so long as this temperature exists the steel cannot be in contact with the water, or, in other words, the heat from the steel vaporizes the immediately surrounding water. The vapour thus formed penetrates the surrounding water and is condensed, and from this action there is surrounding the steel a film of vapour separating the water from the steel, which continues so long as the heat from the steel is sufficiently great to maintain the film against the pressure of the water and the power of the water which rushes toward the steel to fill the spaces left vacant by the condensation of the vapour as it meets a cooler temperature and condenses. The thickness of the vapour film depends mainly upon the temperature of the steel; but here another consideration claims attention. As the heated steel enters the water the underneath side is constantly meeting water at its normal temperature, while the upper side is surrounded by water that the steel has passed by, and, to a certain extent, raised the temperature of. Hence, the vapour on the underneath side is the thinnest, because it is attacked with colder water and with greater force, because of the motion of the steel in dipping. For these reasons it is desirable, especially with thick pieces of steel, to inject the water in a full stream upon the article, as is done in the Brown & Sharpe hardening tanks.
In cases where a great many pieces are to be hardened and tempered to an even degree, the steel is heated for the hardening in a flue with the advantage that contact between the heated steel and the impurities (as sulphur or silicon) of ordinary fuels is avoided, and also that all the pieces may be heated, and therefore hardened, to a uniform degree. The capacity of this system is great, because a number of pieces can be heated without fear of any of them becoming overheated if not attended to immediately. Thus the Waltham Watch Co. heat their mainsprings for the hardening in a flux composed of melted salt and cyanide of potash, the latter serving to clean the surface of the steel; but as the latter wastes it requires to be added occasionally.
The Watch Company, however, find this mixture will not do for the hair springs, as it alters (to a very small degree, however) the nature of the steel; hence these springs are heated for hardening in melted glass.
The Pratt and Whitney Co. heat their taps, &c., for hardening in a composition of equal quantities of salt and cyanide of potash, adding the latter as it wastes, and temper them by the cold test.
The Morse Twist Drill Co. use a similar compound for heating to harden, and the following apparatus for dipping. In a large tank having a free water circulation, stand two pots of a capacity of about five gallons each, one of these contains cyanide of potash and salt, and another sperm oil. The heated work is dipped for an instant into the pot containing the potash and salt, which clean the surface of the steel, and then cooled in the main water tank; but if the work is, from its shape, liable to crack, it is at the final cooling dipped in the pot of sperm oil instead of in the water.