In an effort to employ cast steel of a stronger structure than that found in the annealed steel castings, the possibilities of heat treatment which will increase the strength without materially decreasing the ductility may be resorted to.

The following abstracted report (which is a brief outline of what has already been done by the Pennsylvania Railroad Co. at its shops and Test Laboratories at Altoona, Pa.) is of material interest, as it indicates what may be done with steel castings when properly treated, thereby permitting in railway service greater strength of cast steel parts without any increase in weight or space.

The question of the heat treatment of alloy steel castings is not taken up in this paper.

The obscurity formerly surrounding the heat treatment of steel has been for the most part removed by the development of our knowledge of the critical points of steel, pyrometers, furnace construction, and the testing of the finished product.

The operations of the heat treatment proper will be taken up under the heads of:

1. heating for quenching
2. quenching
3. drawing

Heating for Quenching

Heating for quenching is best conducted slowly, especially in the case of castings of variable thickness. Cracks may occur either in heating or in cooling, due to different temperatures at different points of the casting. The castings should be thoroughly soaked at the maximum temperature (generally 1,500° to 1,600° F.), 1 hr. being sufficient for sections 1 ft. in thickness. The minimum temperature which will produce the desired hardening effect will, in all cases, be found to be the most satisfactory, as the grain coarsens when the critical range is exceeded to too great an extent. All temperatures should be governed by a checked pyrometer with the hot junction to the heated object, and with several couples in a large furnace to insure a uniform temperature.

Quenching Process

The casting should be transferred as quickly as possible from the furnace to the quenching bath, and in the case of large castings, such as locomotive frames, this is by no means a simple matter. The larger castings are best handled by means of cranes and rollers.

The quenching agent employed is generally water or oil, preferably the former, because of its cheapness and drastic cooling effect, more readily breaking up the coarse cast-steel grain. With intricate castings it is generally best to use oil. With water, it is possible to have a large tank and a large running stream, serving to maintain a uniform temperature. Castings should never be thrown an to rest on the bottom of the tank, but should be agitated to prevent the formation of a coating of vapor, retarding the quenching effect. It is also best, whenever possible, to quench the thicker portions first.

Drawing Operation

Whenever possible, the drawing should be done in a bath of some kind, such as lead, barium chloride, a barium chloride-salt mixture, or oil. In the case of large castings this is manifestly impossible, and great care should be exercised in obtaining a uniform temperature in the drawing furnace.

The use to which the casting is to be put determines the drawing temperature, railroad work, by reason of the shock and vibration of the road, requiring high ductility at the sacrifice of some strength.

The results of some tensile, chemical, and micrographic tests of commercially annealed and experimentally heat-treated cast steel bolsters accompany this paper. The following table shows the heat treatment and the results of tensile and chemical tests:

Examination of the accompanying photomicrographs, Figs. 1 to 8, inclusive, shows the inefficiency of the manufacturer’s annealing, which is by no means uncommon. The contrast between the annealed samples and the treated samples , is readily apparent: Figs. 1 and 3 show a coarse needlelike ferrite formation, traces of the casting structure, not obliterated by annealing. Figs. 2 and 4 show the structure of some

experimental high-carbon bolsters, which are not sufficiently ductile. Figs. 5, 7, and 8 show excellent heat treated structures, with a very fine ferrite network. Fig. 6, while given the same heat treatment as the others, shows very little free ferrite, checked by the low elongation of the tensile test. This may be due to the high manganese and silicon content. The heat treatment of these castings would have been more satisfactory with a drawing temperature of 1,100° F. instead of 900° F.

The plot, Fig. 9, shows the average results of a considerable number of tensile tests made from large castings. It will be observed that the heat treatment increases the elastic limit about 50 per cent, and the ultimate strength about 25 per cent., without any material change in elongation. The heat treatment to produce these results is also shown in the plot. In the event a fair amount of ductility is required, it is not desirable to have the carbon content above 0.35 per cent. Individual results will vary from 5 to 10 per cent, from the average results.