The importance of uniformity in heating for forging and for annealing has been stated, and it has been shown how an error in this may be rectified by another and a more careful heating; when it comes to hardening, this uniformity must be insisted upon and emphasized, for as a rule an error here has no remedy.
There may be cases of bad work that do not cause actual fracture that can be remedied by re-heating and hardening, but these are rare, because even if incurable fracture does not occur the error is not discovered until the piece has been put to work and its failure develops the errors of the temperer.
If the error is one of merely too low heat, not producing thorough hardening, it will generally be discovered by the operator, who will then try again and possibly succeed; but if the error be of uneven heat, or too much heat, the probabilities are that it will not be discovered until the piece fails in work, when it will be too late to apply any remedy.
Referring to [Table I, Chap. V], treating of specific gravities, it is clear that all steel possesses different specific gravities, due to differences of temperature, and that these differences of specific gravity increase as the carbon content increases; it follows that if a piece of steel be heated unevenly, internal strains must be set up in the mass, and it is certain that if steel be quenched in this condition violent strains will be set up, even to the causing of fractures.
The theory of this action, as of all hardening, is involved in discussion which will be considered later; in this chapter the facts will be dealt with. When a piece of steel is heated, no matter how unevenly or to what temperature below actual granulation, and is allowed to cool slowly and without disturbance, it will not break or crack under the operation. If a piece be heated as unevenly as, say, medium orange in one part and medium lemon in another, and is then quenched, it will be almost certain to crack if it contains enough carbon to harden at all in the common acceptance of the term, that is to say, file hard or having carbon 40 or higher.
This fact is too well known to be open to discussion; therefore the quenching of hot steel, the operation of hardening, does set up violent strains in steel, no matter what the true theory of hardening may be.
Referring to [Chap. V], to the series of squares representing the apparent sizes of grain due to different temperatures, similar results follow from hardening, with the exceptions that the different structures are far more plainly marked, and the squares should be arranged a little differently; they are shown as continuously larger in [Chap. V], from the grain of the cold bar up to the highest temperature; this is true if a bar has been rolled or hammered properly into a fine condition of grain. Of course if a bar be finished at, say, medium orange it will have a grain due to that heat—No. 3 in the series of squares. Then if it be heated to dark orange and cooled from that heat it will take on a grain corresponding to square No. 2, and No. 1 square will be eliminated.
The series of squares to represent hardened grain will be as follows:
The heat colors being the same as before, viz.: