US2736673A - Axle and method of making the same - Google Patents

Description

Feb. 28, 1956 R. K. POTTER AXLE AND METHOD OF MAKING THE SAME Filed Aug. 25, 1952 United States Patent AXLE AND METHOD oF MAKING THE SAME Rodney K.' Potter, Hobart, Ind., assignor to United States Steel Corporation, a corporation of New Jersey Application August 25, 1952, Serial No. 306,247

2 Claims. (Cl. 148-12) This invention relates to improvements of the art of quenching axles to impart residual compressive stresses therein.

Wide experience of railroads has amply demonstrated a superior performance of axles used in rolling stock after these axles have been given a treatment which induces compressive stresses in the layers of metal constituting the outside shell of the axles.

A conventional manner of imparting such stresses in thev outer layer of axles is set forth in tbe Making, Shaping and Treating of steel by C. B. Francis, 5th edition, pages 819-821. This practice consists in oil or water quenching steel axles from above their upper critical temperature followed, if desired, by a drawing treatment at a temperature below the upper critical point. Theoretical reasoning underlying the selection of such treatments was based on the assumption that the transformation of austenite to martensite on quenching causes compression, since martensite has a larger space lattice parameter than austenite.

Quenching or quenching and drawing provided mproved results, but at the cost of certain negative features. Railroad axles and similar bodies are conventionally forged and roughly machined before heat treatment, and then turned down on a lathe to the linal accurate dimensions. A reasonably high carbon content used for providing axles with the necessary strength develops an excessive hardness when in quenched condition which renders machining operations somewhat dicult. When a drawing operation follows the quenching treatment in order to bring the hardness to the desired machnability range, the drop in compressional stresses present in the outside layers of axles is frequently suciently large to affect markedly the benelicial results produced by th original treatment.

I have discovered that the mechanism of stress formation previously relied upon for treating axles, which depends on allotropic transformations of steel, is not necessary and that superior fatigue-resistance properties and improved machinability can be imparted to axles by quenching them from below their lower critical temperature. The subcritical quench produces the proper stress pattern without materially altering the level of hardness in as forged and double normalized axles.

Accordingly, it is an object of the present invention to provide an improved method of heat treating axles.

It is a further object to provide a method of heat treating axles to produce properly balanced stresses therein.

It is another object to provide an axle having enhanced resistance to fatigue.

The foregoing and further objects will be apparent from the following specification when read in conjunction with the single figure of the drawings which is a graph of the stresses obtained by my improved treatment and by prior art methods.

I have found that the best resistance to fatigue failure can be provided by producing at the surface of the axles residual compressive stresses such that the sum of residual ffice surface stresses and those developed on the compression side of the axle by loading in service do not exceed the compressive yield point of the steel used for axles. Furthermore, the surface compressional stresses should be sutliciently large to prevent the algebraic sum of the surface compressional stresses and of the tensional stresses induced on the tension side of the axle by loading from exceeding 10,000 pounds per square inch in tension. Such axles may be produced from steel containing from .30 to 50% carbon, 50 to 80% manganese, .02% maximum phosphorus, .04% maximum sulphur and .30% maximum silicon with the balance iron.

During a comprehensive search for means suitable for producing the desired stress distribution in axles it was discovered that conventional heat treatment is not satisfactory. Table I presents some of the results recorded in applying conventional treatment, namely quenching from above the upper critical point either in water or in oil, and tempering to sections of seven-inch diameter cylinders machined to size from axles containing 0.46% carbon,v0.68 manganese, 0.016% phosphorus, 0.028% sulphur, 0.23% silicon. The Aci point of the steel was 1353 F. and the Ari point was 1235 F. These cylinders were heated in a neutral atmosphere to 1600 F. until thoroughly soaked, quenched in oil or in water, and drawn at different temperatures for varying times.

Heat treated cylinders were studied by appropriate methods for stress distribution, only the calculated longitudinal stress being presently reported in detail.

Calculated Longitudinal Stress, lbs./sq. in.

Brlnell Time 0i Hardness Type of Cylinder g Draw` Water 011 Hollow, open end A corresponding set of figures obtained on specimens subjected -to quench alone without any subsequent drawing is given in Table Il, in which the treatments employed were strictly comparable to those used in collecting data given in Table I.

TABLE II Test cylinders quenched only Calculated Longi- Qunch- Hardness tulcinl Stress,

g s. sq. in. Type of Cylinder TgmFp.,

- Water Oil Water Oil Hollow, open end 1,600 51,000 Hollow, plugged en 1, 600 305 243 104, 000 80, 000 Do 1,300 193 158 60,000 60,000 Solidl, 300 195 72,000 Do..- 1,000 180 47,000 35,000 Do 900 37, 000 D0 700 176 2, 000

A comparison of values given in the above tables can be preferably made inthe light of practical experience which indicates optimum fatigueproperties being asso ciated with a surface compression between substantially 30,000 and 40,000 pounds, per square inch for a steel having a yield point of about 70,000 andthe requirement of a hardness below about 200 Brinell for satisfactory machining. Viewed from this standpoint, conventional quenching followed by drawing does not produce the desired combination or" surface compression and hardness, while a single subcritical quench can develop themV when applied within a specific range of factors. A cylinder quenched in water from 1000 F. gives a surface compression of about 47,000, which is too high, and Water quenching from 700 F. causes a surface compression of about 22,000 pounds per square inch which'is too low to be acceptable. Water quench-ing from 900 F. produces a surface compression of about VIV37,000 together with a satisfactory hardness, thus indicating a preferred water quenching practice at temperature above 700 F. and below l000 F.

in case of oil quenching,it has been found that. quite satisfactory surface compression of 32,000- pounds per square inch and good hardness can be effected by quenching from 1000" F. in oil, and accept-able results can be expected when the oil quenching range extends substantially from 800 to ll-F.

It has been found, furthermore, that a pronounced eduction in hardness and improvement inrnachinability of axles treated by the method of the present invention does not affect adversely the stress distribution on the surface of the axles. in the single figure of theV attached drawing, longitudinal compression and tension stresses in axles heat treated in the conventional manner and according to the present invention are plotted as a function of distance from the center of shafts. It can be seen therefrom that quenching in oil from 1600 Efollowed by drawing at 800 F. produces compression of 32,000 pounds per square .inch on the outside surface. Axles quenched in oil from l000 F. show a stress pattern which compares very favorably with thev pattern produced by oil quenching from 1600" Fyand drawing at 800 F. developing a surface compression of 35,000 pounds lper square inch. Likewise, water quenching from 900 F., produces stresses of approximately `the same level, the actual compression value in this case being 37,000 pounds per square inch. A reference to Table I will `show that the Brinell hardnessof axles-quenched at 1600D F. in oil and drawn at 800 F. is 250, which is too high for good m-achinability, while Table II shows that the oil quenched and Water quenched cylinders have a Brinell hardness of 1530 and 190, respectively, both remaining in good machinability range.

The present method offers therefore means for producing heat treated axles which have the fatigue resistance characteristics substantially equal to the best obtainable by conventional methods, but which, at the same time, possess improved machinability unavailable by the prior art methods.

While I have shown and described several specific embodiments of my invention, it will be understood that these embodiments are merely Afor the purpose 4of illustration and description and that various other forms may be devised within the scope of my invention, as dened in the appended claims.

I claim:

1. The method of producing railway car axles having enhanced resistance to fatigue, comprising forming a roughly machined forged blank of steel containing .30to .50% carbon, .50 to .80% manganese, .30% maximum silicon with the balance iron and residual impurities, heating sai-d Irough machined blank to a subcritical temperature between 700 and l F., then quenching said heated blank topr-ovi-de aBrinell hardness value less than 200 and residual surface stresses which combined with the stresses developed on the compression side of the axle during loading are less than the compression yield point of the steel, and then machining said quenched blank to its lfinished size.

2. A railway axle produced in accordance with the process described in claim l.

References Citedin the tile of this patent UNITED STATES PATENTS `Coin Mar. 12, 1889 Longford Aug. 13, 1929 OTHER REFERENCES

Claims (1)

1. THE METHOD OF PRODUCING RAILWAY CAR AXLES HAVING ENHANCED RESISTANCE TO FATIGUE, COMPRISING FORMING A ROUGHLY MACHINED FORGED BLANK OF STEEL CONTAINING .30 TO .50% CARBON, .50 TO .80% MANGANESE, .30% MAXIMUM SILICON WITH THE BALANCE IRON AND RESIDUAL IMPURITIES, HEATING SAID ROUGH MACHINED BLANK TO A SUBCRITICAL TEMPERATURE BETWEEN 700 AND 1100* F., THEN QUENCHING SAID HEATED BLANK TO PROVIDE A BRINELL HARDNESS VALUE LESS THAN 200 AND RESIDUAL SURFACE STRESSES WHICH CONBINED WITH THE STRESSES DEVELOPED ON THE COMPRESSION SIDE OF THE AXLE DURING LOADING ARE LESS THAN THE COMPRESSION YIELD POINT OF THE STEEL, AND THEN MACHINING SAID QUENCHED BLANK TO ITS FINISHED SIZE.

Images