US3663314A - Bearing steel composition - Google Patents

Abstract

A bearing steel consisting of 0.55 to 0.78 percent of carbon, 0.50 to 2.00 percent of chromium, 0.10 to 1.15 percent of manganese and 1.00 to 2.00 percent of silicon by weight with the balance iron, said steel having been spheroidized during annealing, and afterward, heated to an austenitizing temperature ranging from about 810* to 870* C. for a period of about 30 minutes to dissolve 0.35 to 0.55 percent by weight of carbon into austenite retaining 3 to 6 percent by weight of undissolved spheroidized iron carbide, liquid quenched and tempered at about 150* C.

Description

United States Patent Monma et a1.

[ 51 May 16, 1972 [54] BEARING STEEL COMPOSITION [22] Filed: Oct. 14, 1970 [21] Appl. No.: 80,646

Related U.S. Application Data [63] Continuation-impart of Ser. No. 691,868, Dec. 19,

1967, abandoned.

[52] US. Cl. ..l48/36, 75/126, 75/126 Q, 148/144 [51] Int. Cl. ..C2ld l/00, C22c 39/14 [58] Field ofSearch ..75/126, 126 Q, 128; 148/31, 148/36, 134,143,144

[ 56] References Cited UNITED STATES PATENTS 2,325,088 7/1943 Wright et a1. ..75/126 Q 2,753,260 7/1956 Payson ....75/126 Q 2,844,500 7/1958 Peras.... ..75/l26Q 3,117,863 l/l964 Roberts et a1... ..75/126 3,155,550 11/1964 Mitchell et al.... 148/134 3,298,827 1/1967 Jatczak ..75/128 3,306,734 2/1967 Aksoy et a1. ..75/126 3,337,376 8/1967 Grange 148/143 3,595,711 7/1971 Faunce et a1. .148/36 OTHER PUBLICATIONS Tool Steels, Roberts et al., ASM, 1962, pp. 321- 339.

Primary Examiner-Charles N. Lovell Attorney-Herman, Davidson & Berman [57] ABSTRACT A bearing steel consisting of 0.55 to 0.78 percent of carbon, 0.50 to 2.00 percent of chromium, 0.10 to 1.15 percent of manganese and 1.00 to 2.00 percent of silicon by weight with the balance iron, said steel having been spheroidized during annealing, and afterward, heated to an austenitizing temperature ranging from about 810 to 870 C. for a period of about 30 minutes to dissolve 0.35 to 0.55 percent by weight of car bon into austenite retaining 3 to 6 percent by weight of undissolved spheroidized iron carbide, liquid quenched and tempered at about 150 C.

3 Claims, 13 Drawing Figures Pmmd my 16-, 1972 4 81:00 a-Sho 3 FIG.

FIG. 6.

0.45% CARBON X4000 O. 65 CARBON X4000 INVENTORS.

TOSH/RO YAMAM q BY I 6 M52 M W? ATTORNEYJ.

This application is a continuation-in-part of applicants previously filed application Ser. No. 691,868, filed Dec. 19, 1967, now abandoned and entitled Bearing Steel Composition and Method of Manufacture.

This invention relates to steels usable for making bearings, and more particularly to an improved composition and method of heat treating such steels in which the carbon content is lower than normal and a particular proportion of the carbon is dissolved in austenite by introducing an austenitizing step, the balance of undissolved carbon remaining as spheroidized iron carbide in martensite.

In the making of steel for bearings it is known that the carbon content of the steel afiects the fatigue life, the compressive breaking strength, and other important characteristics of the finished steel product. However, insofar as known to applicants, all conventional tests and research to discover the relation between the carbon content and the above-named characteristics have failed to establish a distinct correlation, because the total carbon content appears in the finished steel in varying proportions both as dissolved carbon in austenite and undissolved iron carbide in martensite. Accordingly, a specific total carbon content results in varying characteristics depending on the method of heat treatment and the variations in dissolved carbon and undissolved carbide.

It is a primary object of the present invention to provide a steel composition, or product, and method of manufacture in which the percent of carbon dissolved in austenite and the percent undissolved as carbide are predetermined to yield the optimum properties such as long fatigue life and high compressive breaking strength.

It is another important object of the invention to overcome the research problem mentioned above and to determine what steel composition and heat treatment are most suitable to provide long fatigue life and high compressive breaking strength in bearing steel.

Experimentally, according to the present invention, it has been determined that use of a less than normal quantity of carbon in the range from 0.55 to 0.78 percent by weight and the introduction of an austenitizing step after spheroidizing annealing to dissolve 0.35 to 0.55 percent of the carbon in austenite and leaving the balance of undissolved carbon as spheroidized carbide in martensite yields a bearing steel with optimum characteristics.

And furthermore, it has been found that longer fatigue life can be obtained by use of a greater percentage of silicon than usual, as for example, a percentage ranging from 1.00 to 2.00 percent.

As a result of determining the best steel composition and process for manufacture, additional advantages have been gained, such as: the soaking period is shortened because less large carbide appears in the finished bearing steel; the conventional normalizing process may be omitted because less net carbide, proeutectoid cementite, exists because of the lower carbon content; the time period for spheroidizing annealing can be shortened because of the reduction of net carbide; and the total cost of steel production is lowered because of the shortening of the steel fabrication processes.

Applicants have found that steel bearings attaining the foregoing objects can be produced from steels consisting of carbon, silicon, manganese and chromium within restricted ranges. The desired properties can be obtained within the following ranges:

With the balance iron and residual impurities.

A preferred range within the foregoing is as follows:

Carbon O to 1.80%

0.6 Silicon 1.5

Manganese Chromium 0.50 to 2.00% With the balance iron and residual impurities.

The most preferred range is as follows:

Carbon 0.65 to 0.70% Silicon 1.50 to 1.80% Manganese 0.10 to 1.15% Chromium 0.50 to 2.00%

With the balance iron and residual impurities.

The novel features that are considered characteristic of the invention are set forth with particularity in the appended claims. The invention, itself, however, both as to composition and method, together with additional objects and advantages thereof, will best be understood from the following description of specific embodiments when read in connection with the accompanying drawings, wherein:

FIG. 1 is a diagram showing the relations between carbon content and hardness, and between carbon content and compressive breaking strength in carbon steel samples, whose total carbon content is dissolved in martensite matrix.

FIG. 2 is a diagram showing the relation between carbon content and fatigue life of the carbon steel samples.

FIG. 3 is a diagram showing the relations between undissolved carbide and fatigue life, and the compressive breaking strength and hardness in steel samples containing various quantities of carbon and about 1.50 percent of chromium, about 0.45 percent of said carbon content being dissolved in martensite matrix.

FIG. 4 is a diagram showing the relation between silicon content and fatigue life in steel samples containing about 0.45 percent of carbon all of which is dissolved in martensite matrix.

FIG. 5 is a diagram showing the relation betweenv silicon content and fatigue life in steel samples containing about 0.70 percent of carbon and about 1.45 percent of chromium, about 0.45 percent of said carbon content being dissolved in martensite matrix.

FIGS. 6 to 13 are microphotographs of the steel samples used in plotting the diagram illustrated in FIG. 3.

The following is pertinent as background information for the invention. Recent demand for hearing steel has greatly increased. The main alloy elements of such bearing steel are carbon, chromium, silicon, manganese and iron. Depending on its use, the nature and size of bearing, there are differences in content of carbon, chromium, silicon and manganese in hearing steel. The Japan Industrial Standard accepts three kinds of bearing steel according to the following table:

TABLE I [Japan Industrial Standard for bearing steel] Percent Content of Designation carbon Chromium Silicon Manganese The industrial standards for other countries similarly designate bearing steels having a carbon content of about 1 percent by weight as in the United States, Great Britain, West Germany, France, and Sweden. Where lower carbon content steels are accepted for bearings they are usually for special purposes and are case hardened. Generally, a steel containing about 1 percent of carbon, 1.5 percent of chromium, 0.25 percent of silicon and 0.35 percent of manganese, as AISI No. 52100 for example, is most commonly used for bearing steel.

In conventional processes for manufacturing a bearing the steel is made by (l) forming a melt, (2) degassing, (3) casting an ingot, (4) soaking, (5) rolling, (6) normalizing, and (7) spheroidizing. The bearing is then fabricated by (8) turning, (9) quenching, (10) tempering, and l l) grinding or polishmg.

It is a well-known fact that the microstructure of bearing steel after quenching is observed to contain Fe C (cementite) as when a steel containing 1 percent of carbon is spheroidized, all of the carbon exists as ferrous compound of iron, Fe C (cementite) in the amount of 15 percent of the steel. Upon austenitizing of said steel as the quenching temperature, i.e., austenitizing temperature, is increased more cementite (Fe C) dissolves in austenite. According to austenitizing temperature and austenitizing time, the quantity of this Fe C (undissolved carbide) varies from 15 percent to zero (all carbon dissolved in matrix, and no Fe,,C remaining), and consequently the quantity of carbon dissolved in martensite matrix after quenching varies from zero to 1 percent. The longer the austenitizing time, the more cementite dissolves in the matrix, showing the same behavior as if the austenitizing temperature were raised. That is, both the quantity of carbon dissolved in martensite matrix and undissolved carbide (Fe c) vary correlatively with austenitizing conditions (temperature and time).

l-leretofore in attempts to improve fatigue life and high compressive breaking strength of bearing steel, studies were made only in relation to the austenitizing temperature, or time, and using only the conventional bearing steel whose main composition is almost limited to about 1 percent of carbon and about 1.5 percent of chromium. In these studies, for reasons indicated above, it was impossible to find how much carbon should be dissolved in the matrix and how much undissolved carbide would be required for obtaining the most desirable characteristics of bearing steel.

The applicants have examined the relation between the character of bearing steel and carbon content in martensite matrix, and between the former and undissolved carbide. That is, separate investigations were directed to the most desirable quantity of carbon dissolved in martensite matrix and then to the most desirable quantity of undissolved carbide, for improving the fatigue life and compressive breaking strength of the bearing.

To determine the most desirable quantity of carbon dissolved in martensite matrix, samples of several kinds of carbon steels containing about 0.20 to 0.80 percent of carbon were prepared according to Tables II and III.

Each sample was normalized and, omitting spheroidizing,

heated at the temperature higher than A transformation 5 The samples of Table ll were then tested for compressive breaking strength (size of test piece: outside diameter 20 mm, inside diameter mm, height 10 mm) and Rockwell hardness C, and the samples of Table III were tested for fatigue life.

As a result of the test of the samples of Table II, the effect of 10 carbon content in martensite on compressive breaking tts sfl ess re e rly ass ssin 9 by t data of Table II, wherein compressive breaking strength and ;hardness is plotted as a function of carbon content. As shown in FIG. 1, the hardness of carbon steel increases as the carbon 1 content increases, and maximum compressive breaking strength occurs at about 0.40 percent of carbon content diS-i solved in the matrix.

The fatigue life of the samples of Table III, obtained from a life test using a Bearing Steel Life Tester, was measured in millions of revolutions of a ball on a test piece surface having a roughness of about 0.3 microns required to cause flaking by abnormal vibration exceeding 10 microns.

From the data of Table III, the fatigue life is plotted as a function of carbon content in FIG. 2.

2 As shown in FIG. 2, both B life and B life had similar tendency, steel samples containing 0.35 to 0.55 percent of carbon dissolved in martensite matrix show long fatigue life, and especially, fatigue life becomes maximum at 0.40 to 0.50 percent carbon dissolved in martensite matric.

On the other hand, with a carbon content below 0.35 percent, the strength of martensite is low, and also martensite is brittle when the carbon content exceeds 0.55 percent.

It is concluded, therefore, that bearing steels containing 0.35 to 0.55 percent of carbon dissolved in martensite matrix are desireable to provide optimum fatigue life and compressive breaking strength, and that bearing steels containing 0.40

to 0.50 percent of carbon dissolved in martensite matrix are the most desirable to provide above optimum properties.

Next, the influence of undissolved carbide (which exists in steel after quenching) to the character of the bearing steel was examined. Eight kinds of steels whose carbon content varied from 0.45 to 1.17 percent were prepared according to Table .1 a... .1 l-

TABLE II Austenitizing Tempering Compressive Chemical composition breaking Temp. Time Temp. Time strength Hardness Sample C S1 Mn C1 C.) Quin.) C.) (min) (ton) (I'IrC l 0. 20 t). 20 0. 22 0. 14 .100 30 150 )0 4. 10 29. 8 .2 0. 38 0. 23 0. 26 0. 12 800 30 150 JO 4. 66 54. 0 d 0. 50 0 2d 0. 30 0. 13 850 30 150 80 3. 80 58. 8 4 0.61 0 2 0. 20 0. 12 830 30 150 '70 2. 57 (i1. 6 "1 0. H4 0 25 0. 24 0. 1). 800 30 151) I0 2. 40 04. 5

TABLE I11 Fatigue life Austenitizing Tempering' (X10 Chemical composition Temp. Time Temp. Time B10 B 0 Sample (3 S1 Mn C1 C.) (min.) 0.) (min) life life 6 0. 23 0. 25 0. 3i) 0. 14 900 30 150 90 0. 042 0. 27 7 O. 40 0. 26 0. 43 0. 14 860 30 150 90 0. 39 1. 70 8 0. 44 0. 20 0. 34 0. 14 860 30 150 00 0. 5. 60 t) 0. 0. 21 O. 35 0. 14 850 30 150 90 0. 22 0. 82 10 0. 0. 26 0. 44 0. 14 830 30 150 90 0. 060 0. 33 ll 0. 80 0. 26 0. 44 0. 14 S00 30 150 J0 0. 041 0. 25

TABLE IV Chemical composition, percent A B C, D, E perper- Fig- Sarnple C Si Mn Cr C. Min C Min cent cent are 12 0. 45 0. 33 0. 39 1. 44 870 30 150 90 0 0. 45 6 13 0. 55 0. 34 0. 41 1. 47 870 30 150 90 1. 5 0. 45 7 14 0. 0. 36 0. 38 1. 45 870 30 150 00 3 0. 45 8 15 0. 0. 35 0. 38 1. 50 870 30 150 3. 5 0. 45 9 16 0. 78 0. 35 0. 38 1. 50 850 30 90 4. 8 0. 45 10 17 0. 88 0. 35 0. 41 1. 48 840 30 150 90 6. 5 0. 45 11 I8 0. 5 0. 33 0. 3i) 1. 45 830 30 150 90 7. 8 0. 45 12 l!l 1.17 0. 35 0. 42 1. 45 810 30 150 00 10. 6 0. 45 13 FPO??? These samples were normalized, spheroidizing annealed (mean carbide particle size: about 0.6 microns) and austenitized to contain about 0.45 percent of carbon in each matrix. Austenitizing temperature and time were changed for each sample, (see Column A). Since austenitizing tempera- 1 ture and time are changeable correlatively, the figures in Column A are only examples, and not limiting. After austenizing, the samples were oil-quenched and tempered. Consequently, these samples contained the same amount (about 0.45 percent) of carbon in the matrix and different amounts of undissolved carbide, (cementite). The amounts of the dissolved carbon in the matrix were calculated by linear analysis counting the amount of undissolved carbide in comparison with the amount of carbon content in steel.

Samples in Table IV were tested for compressive breaking strength (size of test piece: outside diameter 25mm, inside diameter 10 mm, height 10 mm), hardness and fatigue life.

The test results are set forth in Table V and depicted graphically in-FIG. 3.

TABLE V Compressive Fatigue life (X10 breaking strength Rockwell C 10 life 50 life Sample (ton) hardness As apparent from FIG-3, the characteristics of the steel samples having carbon content in excess of 0.65 percent are as follows:

The fatigue life is the longest within the range of 0.65 to 0.78 percent of carbon, but lowers remarkably as the carbon in the steel goes over 0.78 percent, while the compressive breaking strength decreases gradually with increase of the amount of carbon and hardness reversely is higher, but not remarkably so.

The smaller the quantity of undissolved carbide, the finer the particles of carbide. When carbon content exceeds 1 percent and the amount of undissolved carbide exceeds 10 percent (see FIG. 13), particles of the undissolved carbide grow larger and their volume in matrix becomes great. Therefore, it is seen that fatigue life and compressive breaking strength of the steel with large volume of undissolved carbide are low, because the undissolved carbide itself will cause origin of fracture, or internal defect is induced by martensite of high carbon content produced on the surroundings of carbide.

Further, it is shown from FIG. 3, that the fatigue life and compressive breaking strength of the steel samples containing less than 0.55 percent carbon lower with the decrease of the amount of carbon in the steel, the quantity of undissolved carbide reduces and also hardness is lower remarkably.

It is desirable, therefore, for bearing steel to contain more carbon than the amount to be dissolved in the matrix. The

and Figure 2.

To have the carbon content just equal to that required to be dissolved in the matrix results in dissolving the total carbon content during the austenitizing treatment. If, however, any spheroidized carbon remains in the matrix, the carbon dissolved in the matrix will be less than the amount required. Absence of spheroidized carbide in heat treatment is not appropriate, as described above. therefore, it is preferably to have the carbon content in excess of about 0.55 percent in order to dissolve 0.35 percent of carbon in the matrix, retaining 3 to 6 percent of undissolved carbide, to give easy control of dissolution of carbon during austenitizing, and to stabilize the heat treatment.

That is, spheroidized carbide in the steel is cementite (Fe C) and carbon percentage in cementite is 6.67 percent, the amount of carbon in the carbide is calculated to be 3(percent) 6.67/100 0.2 percent, when 3 percent of undissolved carbide in the matrix of martensite is retained. Consequently, the total carbon content in the steel is calculated to be 0.55 percent, by adding 0.2 percent to 0.35 percent.

As is clear from the above, it is concluded that such bearing steel as contains 0.35 to 0.55 percent of carbon dissolved in matrix and also contains between 3 to 6 percent of undissolved carbide after tempering, is most desirable.

Therefore, on considering both carbon content in martensite and the amount of undissolved carbide, the desired carbon range in bearing steel is limited with the range 0.55 to 0.78 percent, and the better carbon range lies within 0.65 to 0.70 percent.

To add silicon greatly increases resistance to tempering of bearing steel, and it has an advantage of preventing defect of cracking in the course of grinding after quenching and tempenng.

Applicants experimentally have found that fatigue life of bearing steel is improved by adding silicon. In examining the effect of silicon content on fatigue life, steel samples were tested containing no undissolved carbide in martensite matrix and also steel samples with undissolved carbide in martensite matrix.

Firstly, the samples containing 0.45 percent of carbon, all of which is dissolved in martensite matrix, were used, because such steels have the longest fatigue life as mentioned above.

Each sample was austenitized at the temperature of 850 C. for 30 minutes to contain no undissolved carbide, then, quenched in oil and tempered at 150 C. for 90 minutes. Each of the samples was then tested for fatigue life.

The effect of silicon on fatigue life is clearly indicated by the data of Table VI, and the fatigue life is plotted as a function of silicon content in FIG. 4.

TABLE VI Chemical composition (percent) Fatigue life (X10 C Si Mn Cr B life B 0 life Sample 20 in Table VI and Figure 4 is equal to Sample 8 in Table III It is obvious from Table VI and FIG. 4 that the fatigue life of the tested samples is not improved when the silicon content varies within the range 0.25 to 0.80 percent and fatigue life inideal quantity of undissolved carbide was found to be 3 to 6 creases gr ly Over th rang of silicon content from 0.80

percent.

Moreover, as shown in FIG. 3, it is not desirable to have a carbon content in excess of 0.95 percent because such carbon content remarkably deteriorates fatigue life, while, as the carto 1.47 percent. Fatigue life of Sample 23 having 1.47 percent of silicon is two times that of sample 20 containing 0.20 percent of silicon. Since, as has been shown, the fatigue life of steel containing silicon in excess of 1 percent increases rebon content in the steel is lowered, the steel has better fatigue markably, therefore, a preferred range of silicon content is life and compressive breaking strength, and the steel with 0.78 percent of carbon exhibits the most advantageous improvement of these qualities. Further, the two steels with 0.65 percent and 0.70 percent of carbon have the longest fatigue life and the highest compressive breaking strength.

from about 1.00 to 2.00 percent.

Next, the fatigue life was examined by using steel samples with about 0.70 percent of carbon and 1.45 percent of chromium, in which the silicon content was varied from 0.35 to 1.99 percent.

Each of the samples was spheroidizing annealed (mean carbide particle size: about 0.6 microns), austenitized at the temperature of 870 F. for 30 minutes, quenched in oil and then tempered at 150 C. for 90 minutes. Consequently, each sample had about 0.45 percent of carbon content dissolved in martensite and retained about 3.8 percent of undissolved carbide.

The results of the test analysis are shown in Table VII and illustrated graphically in FIG. 5.

TABLE VII Chemical composition Fatigue life (pcrmn'it) (X Si Mn Cr B life B 0 life 0 35 0. 38 1. 5O 2. 00 ll. 00 0 88 0. 41 1. 44 2. 31 12. 33 "7 1 00 0. 38 1. 44 2. 75 14. 49 l. 2'.) O. 31) 1. 46 4. 40 22. 87 l). 6!) 1. 50 0. 3 1. 45 5. 51] 27. 45 (l. 70 1.67 0.40 1. 45 5. 82 21]. l 0. 70 l 80 0. 37 1. 43 5. 57 25. 58 32 0. 71 1 l1!) 0. 3!) 1. 46 4. 01 20.08

No'ric. Sample 25 shown in 'lahleVll and Figure 5150111111110 Sample in 'lnlilv 1V and l igul'v 3.

It is clearly depicted in FIG. 5 that the fatigue life of samples is virtually changeless within silicon content range of 0.35 to 0.88 percent, but increases gradually with increase of silicon proportion in the range of 0.88 to 1.50 percent of silicon, and is the longest within the range of 1.50 to 1.80 percent of silicon. However, fatigue life of Sample 32 with silicon content 1.80 percent decreases somewhat, but is still superior to that of Sample 27 containing 1.00 percent of silicon. Silicon in excess of 2.00 percent will lower the toughness of the bearing steel.

It is apparent from FIG. 4 and FIG. 5 that the fatigue life of steels with 3.8 percent of undissolved carbide is longer than that of steels containing no undissolved carbide over the range of 0.20 to 1.99 percent of silicon content, and the steel containing about 1.00 to 2.00' percent of silicon, with undis'solved carbide, shows preferable fatigue life, and especially the desired range of silicon is about 1.50 to 1.80 percent.

Chromium and manganese influence hardenability, resistance to tempering, wear resistance, ease of spheriodizing and the solution rate of carbide in austenite. In the martensite tate of about 150 C. of usual tempering temperature, however, they scarcely affect fatigue life and compressive breaking strength. They can be added adequately within the range of common use (0.50 to 2.00 percent of chromium, 0.10 to 1.15 percent of manganese) according to use, nature and size of the bearing.

Vanadium, molybdenum and nickel are contained in aome bearing steels, however, such elements scarcely affect fatigue life and compressive breaking strength of the steel in the state of martensite tempered 150 C. of usual tempering temperature. Therefore, the steel of the present invention does not contain vanadium, molybdenum nor nickel.

From the foregoing, it is clearly apparent that a bearing steel containing 0.55 to 0.78 percent of carbon, 1.00 to 2.00 percent of silicon, 0.50 to 2.00 percent of chromium and 0.10 to 1.15 percent of manganese, wherein the steel has 0.35 to 0.55 percent of carbon dissolved in martensite and retains 3 to 6 percent undissolved carbide, has the longest fatigue life, the highest compressive breaking strength and suitable hardness at which the present invention aims.

EXAMPLE An example of a process for manufacturing steel and a ball bearing according to this invention, is as outlined in the following steps:

l. The steel whose chemical composition by weight is shown in Table VIII below was melted in a l0-ton furnace:

TABLE VIII Percent Sample No 33 34 35 36 37 Chemical composition of the steel:

Carbon 0. 95 0. 78 0. 78 0. 65 0. 55 Chromium 1. 44 1. 46 1. 32 l. 00 1. 60 Manganese- 0. 41 0. 37 0. 25 0. 60 0. Silicon 0.25 0. 22 1. 52 1. 63 1. 70 Impurities:

Phosphorus 0.013 0. 011 0.011 0. 010 0. 011 Sulfur 0. 007 0. 009 0. 001) 0. 010 0. 009 Nickel. 0. 04 0.05 0.05 0. 06 0. 05 Copper 0. 07 0. 10 0. l1 0. 08 0. 09 The remainder: Iron 97. 82 17. 00 05. 95. 96 95. 09

2. Said melted steel was held in a vacuum atmosphere and degassed of oxygen, hydrogen and nitrogen to improve the fatigue life of the bearing.

3. The melt was cast in an ingot mold of 2.5 tons and allowed to solidify.

4. The ingot was soaked to decrease segregation of impurities under heat treatment at 1,265 c. for 5 hours, (conventional treatment is 15 hours).

5. The steel was rolled to a round bar of 65 mm.diameter to produce a bearing numbered 6206 as final product.

6. Normalizing which is essential to conventional steels, was omitted, and, spheroidizing in annealing was applied soon after rolling.

7. The outer and inner rings of the bearings were turned from the annealed round bar on a lathe.

8. The cut rings were quenched and tempered under the condition of temperature and time according to this invention, i.e., temperature: 810 to 870 C., time: 30 minutes, dissolving about 0.35 to 0.55 percent carbon in the matrix.

9. The rings were ground to eliminate the decarburized layer and to polish the track surface, and assembled to form the final product, a radial ball bearing No. 6206 having an outer diameter of 62 mm.,an inner diameter of 30 mm. and a thickness of 16 mm.

The result of fatigue life test of the bearing steels on samples using radialball bearing tester is shown in Table 1X.

Analysis of the data of Tables VIII and 1X is as follows: Fatigue life of Sample 34 (0.78 percent of carbon) is longer than that of Sample 33 (0.95 percent of carbon), conventional bearing steel (ex. .118 SUJ2), and fatigue life of Sample 35 (0.78 percent of carbon, 1.52 percent of silicon) and Sample 36 (0.65 percent of carbon, 1.63 percent of silicon) are the longest. And also, fatigue life of Sample 37 (0.55 percent of carbon, 1.70 percent of silicon) is shorter than that of samples 35 or 36, but is longer than that of Sample 33.

Applicants believe themselves to be the first to have discovered that less than 0.35 percent of carbon dissolved in the matrix reduces strength of the bearing steel, and that an excess of dissolved carbon above 0.55 percent also is detrimental is causing brittleness in the' martensite. The further discovery that both fatigue life and compressive strength are improved as the quantity of undissolved carbide decreases appears to be novel. The introduction of a heat treatment in which the bearing steel is spheroidized in annealing and austenitized at the temperature range of about 810 to 870 C. (at a quenching temperature), and consequently 0.35 to 0.55 percent of carbon is dissolved in the martensite matrix leaving a predetermined percent of undissolved carbide, is believed to constitute a new method of manufacture and a new composition of a bearing steel.

According to this invention, as described above, the soaking process for steel can be shortened, the normalizing process can be omitted, and the spheroidizing annealing process can be shortened. Further, bearing steel is produced with fairly fine spheroidized cementite structure. The conventional bearing steel specified by the Japan Industrial standard is hypoeutectoid steel having 0.95 to 1.10 percent of carbon content. If it is slowly cooled from rolling or forging temperature downward, there is firstly precipitated proeutectoid cementite at austenite grain boundary at temperature below Acm and then pearlite is formed at the point of A and the structure becomes cementite and pearlite structure. Particularly, in a large sized bearing, the steel is slowly cooled for preventing flakes and the net of its proeutectoid cementite develops much larger. Such thick proeutectoid cementite cannot be diminished by spheroidizing annealing in a later process, and, therefore, the steel is generally subjected to normalizing. In spite of such normalizing, it is quite difficult to prevent proeutectoid cementite newly precipitated by air cooling in such a large sized rolled and forged product. Also, in water cooling, the operation always accompanies danger of quenching crack and requires much higher and more careful technique. F ormation of such proeutectoid cementite is caused by high carbon content. Even in the precipitation of a less thick proeutectoid cementite, its dissolution rate in austenite is much slower than that of cementite in pearlite when it is heated over A point during the spheroidizing annealing process. If the steel is held a much longer time at temperature over A point during spheroidizing annealing in order to distribute spheroidized carbide finely and uniformly, the nucleus of the spheroidized cementite will decrease and the obtained particles of spheroidized cementite will become coarse and cause origin of fracture.

Huge carbide content of bearing steel seems to be formed by eutectic reaction upon solidification of steel ingot when impurities such as carbon and phosphorus increase in the melt among primary crystals during crystallization of proeutectic austenite, or is formed by easy precipitation of carbide particularly in phosphor rich part because of low solubility of carbide, when carbide beyond saturated solubility limit is produced by A transformation after solidification of steel ingot. In either case huge carbide is certainly formed in the segregation part of carbon, chromium and phosphor. In order to prevent defects occurring from the huge carbide, it is essential to consider the amount of carbon content which is the main cause of the formation of huge carbide as well as to dissolve the huge carbide or to diffuse the segregation of elements of phosphor.

The present invention lowers the carbon content in steel to about 0.55 to 0.78 percent, which does not develop proeutectoid cementite, and this cementite can be diminished in spheroidizing annealing process without normalizing processes. The solution of carbide into austenite is promoted and controlled by simple adjustment of austenitizing temperature and holding time. Fairly fine spheroidized carbide is obtained by spheroidizing annealing for a short period, and the formation of huge carbide is limited and, further, soaking time can be shortened.

Although certain specific embodiments of the invention ha been shown and described, it is obvious that many modifications thereof are possible. The invention, therefore, is not to be restricted except insofar as 15 necessitated by the prior art and by the spirit of the appended claims.

We claim:

1. A heat treated bearing steel consisting of 0.55 to 0.78 percent of carbon, 0.50 to 2.00 percent of chromium, 0.10 to 1.15 percent of manganese and 1.00 to 2.00 percent of silicon by weight with the balance iron and residual impurities, said steel having 0.35 to 0.55 percent by weight of carbon dissolved in matrix of martensite and retaining 3 to 6 percent by weight of undissolved spheroidized iron carbide said steel having been spheroidized during annealing, and afterward,heated to an austenitizing temperature ranging from 810 to 870 C. for a period of about 30 minutes, liquid quenched and tempered at about C.

2. A bearing steel according to claim 1, wherein the carbon is 0.65 to 0.78 percent, and silicon is 1.50 to 1.80 percent by weight.

3. A bearing steel according to claim 2, wherein the carbon is 0.65 to 0.70 percent by weight.

Claims (2)

1. 2. A bearing steel according to claim 1, wherein the carbon is 0.65 to 0.78 percent, and silicon is 1.50 to 1.80 percent by weight.
2. 3. A bearing steel according to claim 2, wherein the carbon is 0.65 to 0.70 percent by weight.

Images