US1517354A - Process for improving metals and alloys in resistance to repeated stress - Google Patents

Description

Dec, 2, 1924.

H. W. GILLETT PROCESS FOR IMPROVING METALS AND ALLOYS IN RESISTANCE TO REPEATED STRESS Filed April l0. 1923 Inl/anion @SMQ SQ. am

Patented e.i2,19`24. i

UNITED STATES HORACE W. GILLET@ F ITHACA, NEW-YORK.

' PROCESS FOR IMPRO'VING METALS ALLOYS IN'TANCE T0 BEPEATED STRESS. f

Application nieaapru 1o,

1923. Serial Nb. 081,191.

' (rrrnnumira rnE Ac'r or mncn s, 1era, ze sur. L., eas.)

To all whom t may concern.:

Be it known that I, I-'IoRAc W. GILLETT, al citizen of the United States, residing in Ithaca, in` the county of Tompkins and 5 State of New York, have invented a new and useful Process for Improving- Metals and Alloys in Resistance to Repeated Stress,

of which the following is a specification. n

This application'is made under the act ofi Ma1)ch 3, 1883, C. 143, (U. S. Stat., 22, p.

One object of my'invention is to alter the properties of a metal or alloy so that vit will withstand repeated stress or cyclic varia-V tions of stress fora longer period than such' cyclic stress can be resisted by unaltered metals and alloys.

Another object is to increase the range of cyclic stress which a metal or alloy can withstand without -failure.

A further object is to increase the resist- .ance of metals and alloys to repeated stress to an extent greater than, and superimposed upon, the known increase in such resistance that can be obtained by such an operation as forging, rolling, drawing, spinning, etc., or by heat-treatment, or both. If dilerent specimens of a given metal or alloy be subjected to cyclic 'variations of 3o stress, as in repeated bending, in rotary bending of a loaded bar, or in reversed axial stress, i. e., in repeated tension and compression, it is found that as the stress range to which the material is subjected is decreased a5 the number of alternations of stress ,which the specimen will withstand before cracking or breaking increases. If the stress or stress range be plotted against' the number of cycles, a curve of the general type of Fig. 1 4o (in which Vstress is plotted on an ordinary scale, and number ofalternationson a logarithmic scale) is obtained in which, at some particular stress or stress range as at A, the relation between stress or stress ran e and life changes: At stresses above A t e life increases slowly with decrease in stress, while at those below A, lthe life increases rapidly with decrease inlstress.

The stress corresponding to the point A,

"the curvature or slope of the 'curve at stresses above and below that corresponding to A, and the number of cycles corresponding to the -point A, may be different in diiferent' metals and alloys, in dile'rent states of hardness, internal structure,I purity,

cleanliness, etc.

The stress correspondin to the point Av may, in some metals or a oys be above the static elastic limit or the static ield point of the'material when tested without cyclic stressin and in other metals or alloys, it

may be low.

The stress corresponding to point A may be considered as a cyclic elastic limit or an endurance limit, as'it is commonly termed. It is also called the safe range of stress, and it is sometimes considered that at lower stress ranges than the safe range failure will never ensue. But as no tests can be made over an infinite period of time it is impossible to establish experimentally whether or not there is a range of stress at which lfailure will never occur. Forengineering purposes, however, it is of value to know at what stress range the relation between stress and life changes, and the endurance limit 1s the stress at which this relation changes. In the actual case of a quenched and tempered alloy steel with a Brinell hardness of about 400 in a reversed bending test, the point A is-at about 100,000 lbs.l per sq. in. stress in tension and compression, or a stress range of 200,000 lbs. per sq. in., and at about 750,000 alternations, undenthe conditions of the test. n

If a bar of this material is stressed to 116,000 lbs. per sq. in. it will break/in about 100,000 alternations, and one stressed to 110,000 lbs. per sq. in., in about 140,000 alternations.

If, however, instead of initially stressing the bar above'the endurance limit of about 100,000 lbs. r sq. in. it is initiall stressed at a stress low the endurance 't but near to it, inthe vactual case, at 97,000 to 99,000 lbs. per sq. in., for a considerable number of alternations, in the actual case 1,400,000. to 1,500,000, an improvement in the bar takes lace by which the bar is strengthened. ff the stress is then raised, the strehened bar does not breakin as few alternations as an uri-strengthened bar does. If, after a period of stressing at the new stress the stress is again increased, further strengthening takes place. The increase of stress can take placein steps or-it may gradually and pro Le: inc :u L I los ln the actual case, shown in Fig. 1, a bar (B) first stressed for 1,500,000 alternations at 97,000 lbs. per sq. in. was then stressed at 101,000 lbs. per sq. in. (see point B) for 1,100,000 alternations, and then when tested at 110,000 lbs. per sq. in. (see point Bm), it broke at 650,000 alternations, although an un-strengthened bar (B) breaks at 140,000 alternations at that stress. The life at 110,000 lbs. per sq. in. has thus been increased over 4,1/2 times by the strengthening process.

Another bar (C) first stressed at 99,000 lbs. per sq. in. for 1,400,000 alternations, then at 104,000 lbs. per sq. in. (see point (1) for 1,250,000 alternations, then at 111,- 000 lbs. per sq. in. (see point Cm) for 1,350,000 alternations and then tested at 116,000 lbs. per sq. in. (see point C), broke after 1,800,000 alternations, although an uri-strengthened bar C breaks after 95,- 000 alternations at that stress. llfhe life at 116,000 lbs. per sq. in. has thus been increased about 190 times by the strengthening process. In Fig. 1 the points showing the stresses and alternations of the bars referred to above are plotted with circles and arrows where the bars did not break and with crosses where they did break.

Although a hard heat-treated alloy steel has been chosen as an example, the strengthening process applies to iron and soft steels and to non-ferrous metals and alloys as well.

l am aware that it has been suggested that the properties of metals and alloys may be improved through raising the yield point by overstressin i. e. stressing above the yield point or e astio limit either in tension or compression, after which overstress the f material may behave elastically up to the stress so impressed, and that this may either he done at elevated temperatures or the temperature may be thereafter raised for tern pering or annealing purposes. But such overstress, producing permanent set in the material, is attended by a decrease in the elastic limit in compression if the applied over-stress is a tension stress, or a decrease in the elastic limit in tension of the applied over-stress is a compressive stress. That is, by over-stressing above the elastic limit the material may be materially A improved against either tensile or compressive stress, but not against both.

My process for "improvement by understressing below the endurance limit is sharply dierentiated from the above-mentioned processes, because the understressing process improved the material against both tensile and compressive stresses, as is shown by the results plotted in Fig. 1, in which the test specimens were all subjected to equal alternatin tensile and compressive stresses in reverse bending.

It is further dierentiated in that instead of a single or a very few applications of stress, my process involves the repetition of the stress below the endurance limit many times.

Still further dii'erentiation is found in the fact that in the overstressing process no improvement is shown in metals or alloys unless the .stress is so great as to cause a detectabley permanent set in the material, while the process of repeated understressing will generally, and in such materials as most commercial steels always, produce no detectable permanent set whatever. My process involves no change in the external dimensions of the material operated upon in cases when the endurance limit is below the elastic limit, which is usual.

The understressing process is so called because it must start by stressing below the endurance limit. After the material has been strengthened and its endurance limit raised above its initial value, the process can be hastened by raising the stress, but still keeping it below the now increased new endurance limit. The stress may be increased in steps, or it may be steadily increased, or increased at varying rate. An automatic increase of stress may according ly be arranged for in the apparatus by which the stress is applied so that as the endurance limit rises the applied stress rises, following the endurance limit but keeping below it. When the stress is steadily increased, it might, for example, be increased at the rate of a tenth of a pound per square inch per cycle of stress, while, when it is increased at a varying rate it might be increased first at say one tenth of a pound per square inch and later at two tenths of a pound per square inch per cycle.

The strengthening process may be carried out at elevated temperatures since the interval change or readjustment which results in strengthening appears to be accelerated by increase in temperature. Of course in the case of a metal or alloy whose initial properties have been obtained by cold rolling ora similar strain-hardening process or by quenchin and tempering, too high a temperature wou d soften the material by annealing or tempering, so that there will be an upper.- limit of temperature for the strengthening process which will vary with the particular material in question. process could also be carried out at lowered temperatures, although this will usually be a disadvantage rather than an advantage.

The strengthening process may be applied The,

to a finished part of a machine, for example,

by running the machine first at light loads and later increasing the load. This is common practice, and while the usual reason for it is to wear in the bearings, it may readily 'bring about some of the strengthening process, even though the operator of the machine does not know that strengthening takes place.

It is, however, a new procedure in the preparation for use of metals and alloys to intentionally subject them before final assembly and initial use in the finished machine to a strengthening process of repeated stressing with the object of increasing their ability to withstand repeated stress.

The strengthening process can be applied to a finished part by loading it in a similar mannerv to that in which it will be loaded in use. It can be applied to a partly finished part or to unfinished material. For example, an axle shaft could be rotated in an apparatus similar tothe well known Farmer-type of rotary bending endurance testing machine. A finished connecting rod for a gas engine could be put in repeated tension and compression, or, the rough forging could be similarly stressed and later finished to size. Or, a block of metal could be so stressed, strengthened throughout, and any piece desired can be later machined therefrom. Other applications of my invention to other cases are self-evident.

I do not limit myself to any particular apparatus or method of producing repeated stress, to any specific metals or alloys, to any specific temperature of operation, nor to any particular combination of applied repeated stresses. The applied repeated stress may vary from zero or from a finite tensile or compression stress, for example, to another nite tensile or compression stress. The important limitation is that the initial stress or stress range which is to be repeated be below the endurance limit (respectively stress or stress range) of the particular material for that sort of stress. The initial stress may be close to or well below the endurance limit, but in practice, it will naturally be not very far below it. One advantage of using an initial stress near the endurance limit and of repeating that stress for a considerable number of alternations is that a defective piece, containing surface imperfections or internal flaws of any nature which tend to increase the local stress at any point as distinguished' from the stress calculated from the load applied andthe dimensions of the piece, will fail under the initial repeated stressing so that the process will tend to seek out and eliminate such defective pieces as well as to strengthen pieces that are not notably defective.

I do not limit myself to any particular steps of increase in stress, to any particular rate of increase ofstress with a number of alternations, nor any particular speed of alternations of stress, since these will vary with the material, with the result desired and with the apparatus used to bring about the strengthening process.

Nor do I limit myself by any particular theoretical explanation of the exact phenomena occurring in the strengthening process, since the process is an experimentally observed fact for which different explanations may be adduced.

I am aware that other investigators have observed a similar strengthening of materials in the course of. endurance tests and I do not claim the process as a method of testing.

I do broadly claim the` application of the strengthening process as a step in the industrial preparation of materials to withstand re eated stress.

What claim as new and wish by Letters Patent is:

1. The process of strengthening metals or alloys in their endurance against repeated stress by subjecting them to repeated stress to protect of magnitude not greater than the stress v corresponding to the endurance limit of the material.

2. The process of claim 1 followed by further subjection to repeated stress at one or more stresses above the original endurance limit of the material which it had before undergoing the process of claim 1.

3. The process of claim 2 carried out at stresses increased at a constant rate.

4. The process of claim 2 carried out at stresses increased at a varying rate.

5. The process of claim 2 carried out at stresses increased partly at a constant rate and partly at a varying rate.

6. The acceleration of the strengthening process of claim 1 by carrying out the process at a temperature above room' temperature.

7. The acceleration of the strengthening process of claim 2 by carrying vout the process at a temperature above room temperature.

8. The acceleration of the strengthening process of claim 3 by carrying out the lprocess at a temperature above room temperature.

9. The acceleration of the stren thening process of claim 4 by carrying out t e process at a temperature above room temperature.

10. The acceleration of the stren hening process of claim 5by carrying out t e proc` Aess at a temperature above room temperature.

Images