JP2011215139A - Method and device for estimating fatigue limit surface pressure of rolling bearing material for railroad vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for estimating the fatigue limit surface pressure of a rolling bearing material for a railroad vehicle capable of precisely estimating the fatigue limit surface pressure of the rolling contact metal material such as rolling bearing steel or the like from the result of a short-term fatigue test.SOLUTION: The relation between the shearing stress amplitude and load frequency of a metal material is calculated by a perfect alternating ultrasonic twist fatigue test (S1) and the shearing fatigue strength τin an ultralong life region is determined from the relation between the shearing stress amplitude and load frequency of the metal material according to an S-N diagram or the like (S2). The maximum contact surface pressure Pat the time of action of the load, wherein the maximum alternate shearing stress amplitude τacting on the inside of the surface layer of the metal material determined from contact dimension data and the load becomes equal to the shearing fatigue strength τ, is set as the estimate value of fatigue limit surface pressure P(S3).

Description

TECHNICAL FIELD The present invention relates to a method and an apparatus for estimating fatigue limit surface pressure of rolling bearing materials for railway vehicles, and obtains shear fatigue characteristics up to an ultra-long life range by experiments, and is an internal origin type of a machine element that makes rolling contact such as a rolling bearing. The present invention _relates to a method and an apparatus for estimating a maximum contact surface pressure P _{max at} which _separation does not occur as a fatigue limit surface pressure P _{max 1im} , and a method for selecting a bearing material using the estimation method.

It is thought that the internal origin type delamination of rolling bearings is caused by cracks that are generated and propagated by repeated alternating shear stress (almost both swings) with the maximum amplitude inside the surface layer. In the case of the tensile compression fatigue test (axial load fatigue test, rotary bending fatigue test), it is customary to use the fatigue strength at 10 ⁷ times as the fatigue limit. On the other hand, in the case of rolling bearings with good lubrication conditions, even if a considerably high load is applied, the internal starting type separation does not occur at a load frequency of about 10 ⁷ times. There is a torsional fatigue test as a test for fatigue fracture by shear stress, but the load frequency of the hydraulic servo torsional fatigue test is at most 10 Hz. For example, it takes 3 years or more to reach 10 ⁹ times. Therefore, it is practically impossible to obtain the shear fatigue characteristics up to the ultra-long life range.

  Instead, non-metallic inclusions contained in an arbitrary volume are considered from the idea that non-metallic inclusions that are inevitably contained in steel and are structurally discontinuous, and that cause non-metallic inclusions as the source of stress concentration, will be the origin of internal-origin separation. A method of estimating the maximum size of an object by extreme value statistical analysis has been devised, and a method has been adopted in which the maximum size of nonmetallic inclusions is used as an index of steel quality (for example, Patent Documents 1 to 4).

JP 2004-251898 A JP 2005-105363 A JP 2006-128865 A JP 2006-349698 A

Yukio Fujii, Kikuo Maeda, Akio Otsuka, NTN Technical Review, 69 (2001) 53-60. Y. Murakami, C. Sakae and S. Hamada, Engineering Against Fatigue, Univ. Of Sheffield, UK, (1997), 473p. Tee. T. A. Harris, Rolling Bearing Analysis (Third Edition), Wiley-Interscience, New York, (1991), 147p. Material Society of Japan, Revised Material Strength Science, Material Society of Japan, Kyoto, (2006), 94p. Material Society of Japan, Revised Material Strength Science, Material Society of Japan, Kyoto, (2006), 211p.

The mode of fatigue crack propagation at the surface of the rolling contact surface prior to internal origin type delamination is considered to be mode II. As a method of estimating the fatigue limit surface pressure from the maximum size of the non-metallic inclusions, there is a concept described in the consideration of Non-Patent Document 1. As shown in FIG. 13 of Non-Patent Document 1, when the Hertz contact pressure moves, a circle having a diameter 2a at a depth b / 2 (b is the minor axis radius of the contact ellipse) at which the alternating shear stress amplitude is approximately maximized. It is considered that a plate crack exists. Think of this crack as the diameter of the largest inclusion. In Non-Patent Document 1, an original mode II fatigue crack growth experiment is performed, and the lower limit value of the stress intensity factor at which fatigue crack growth does not occur is determined as ΔK _IIth = 3 MPa√m. In FIG. 14 of Non-Patent Document 1, in the case of ΔK _IIth = 3 MPa√m, it is assumed that the coefficient of friction between crack surfaces is 0.5, and the critical pressure on whether or not the maximum contact surface pressure and fatigue crack progress. The relationship of the crack diameter 2a is shown. For example, when 2a = 50 μm, the fatigue limit surface pressure is estimated to be P _{max lim} = 2.5 GPa. However, in this method, the coefficient of friction between the crack surfaces is unknown and must be assumed to be a certain value. Non-Patent Document 2 also conducts an original mode II fatigue crack growth experiment and obtains the lower limit value of the stress intensity factor at which fatigue crack growth does not occur as ΔK _IIth = 13 MPa√m. It is very different from ΔK _IIth .

  Rolling bearings for railway vehicles travel with a large number of passengers and are frequently used for many years, so that high reliability is required. Therefore, if it is possible to estimate the fatigue limit surface pressure by conducting a torsional fatigue test for each supplier or lot of materials used, it is effective for improving reliability. However, with the conventional technology, as described above, the torsional fatigue test requires a long time, and it is impossible to estimate the fatigue limit surface pressure of the material used. For this reason, there was no idea of adopting fatigue limit surface pressure as one of the test items for bearing materials.

The object of the present invention is to provide a rolling bearing material for rolling stock that can accurately estimate the fatigue limit surface pressure of a metal material with high fatigue strength that is in rolling contact with the rolling bearing steel or the like from the result of a short-term fatigue test. It is an object of the present invention to provide a fatigue limit surface pressure estimation method, estimation apparatus, and estimation system.
Another object of the present invention is to provide a method for selecting a rolling bearing material for a railway vehicle that can improve the reliability of the rolling bearing for a railway vehicle by adopting test items that were not previously conceived.

The outline of the method of the present invention is as follows. An ultrasonic torsional fatigue test is used to determine the shear fatigue characteristics up to the ultra-long life region, and the internal origin type separation of the metal material that will be used as the bearing ring or rolling element of rolling bearings for railway vehicles. This is a method of estimating the maximum contact surface pressure P _max that does not occur as the fatigue limit surface pressure P _{max 1im} .
The method for estimating the fatigue limit surface pressure of a rolling bearing material for a railway vehicle according to the present invention is a method for estimating the fatigue limit surface pressure P _{max 1im} of a metal material used as a race or rolling element of a rolling bearing for a railway vehicle. ,
A test process (S1) for determining the relationship between the shear stress amplitude of metal material and the number of loads by an ultrasonic torsional fatigue test;
A shear fatigue strength determination process (S2) for determining the shear fatigue strength τ _1im in the ultra-long life region from the relationship between the obtained shear stress amplitude and the number of loads, according to a predetermined criterion,
Maximum alternating shear stress acting on the surface of the metal material determined from the shape and size of the surfaces of the object made of the metal material and the object that is in rolling contact with the object, the size and the load that gives the contact surface pressure amplitude tau ₀ is the shear fatigue strength tau determined by calculation formula defined maximum contact surface pressure P _max at which the load is applied equal to the _HM, the maximum contact surface pressure P _max fatigue limit surface pressure P _{max And} a fatigue limit surface pressure calculation process (S3) with an estimated value of 1 _im .
The ultrasonic torsional fatigue test is preferably a full-twisted torsional fatigue test that gives torsional vibration in which the torsion in the normal rotation direction and the reverse rotation direction is symmetric with respect to the test piece. The metal material may be a rolling bearing steel used as a bearing ring or rolling element of a rolling bearing.

The above-mentioned “shear fatigue strength in the ultralong life region” is synonymous with “shear fatigue limit”, but in this specification, it will be described as “shear fatigue strength in the ultralong life region”.
The “defined standard” used in the process of determining the shear fatigue strength is, for example, a curve obtained by fitting the relationship between the shear stress amplitude and the number of loadings of the test result to the established theoretical curve indicating the shear fatigue strength. The shear fatigue strength is determined from the curve. Specifically, the SN diagram (fatigue strength diagram with 50% fracture probability) obtained by applying to the fatigue limit type polyline model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials Science Can be used. The SN diagram may be obtained by applying not only to the fatigue limit type broken line model but also to a continuously decreasing curve model. However, in that case, τ _1im needs to be defined as “a value on an SN diagram at 10 ¹⁰ times”, for example.
Non-Patent Document 3 describes a “defined calculation formula” used in the fatigue limit surface pressure calculation process. FIG. 5.13 of Non-Patent Document 3 is the circumferential distribution of the alternating shear stress at the depth at which the alternating shear stress acting below the contact surface in the line contact state is maximum, and is four times the maximum alternating shear stress τ ₀ . Is equal to the maximum contact surface pressure _Pmax . Therefore, in the case of a metal material that can be regarded as a line contact state,
(Fatigue limit surface pressure P _{max 1im} ) = 4 × (shear fatigue strength τ _1im )
It becomes.

According to the method of the present invention, since the fatigue test is performed by an ultrasonic torsional fatigue test, an extremely high-speed load is possible, and the relationship between the shear stress amplitude of the metal material and the number of loads can be obtained in a short time. From the relationship thus obtained, the shear fatigue strength τ _1im in the ultra-long life region is determined, and the maximum alternating shear stress amplitude τ ₀ acting on the inside of the surface layer from the contact dimension specifications of the metal material becomes equal to the shear fatigue strength τ _1im. to estimate the maximum contact surface pressure _{P max} when the load acts as a fatigue limit surface pressure _{P max HM,} it is possible to precisely estimate the fatigue limit surface pressure _{P max HM} from the results of the torsional fatigue test. For this reason, when estimating the fatigue limit surface pressure P _{max 1im} of the rolling bearing steel, which is a material having a strong shear fatigue strength τ _1im , the effect that only a short test is required is more effectively exhibited. The Therefore, the shear fatigue characteristics of the metal material used as the rolling ring or rolling element of the rolling bearing for a railway vehicle can be obtained, and the fatigue limit surface pressure P _{max 1im} can be evaluated from the obtained shear fatigue characteristics. By using a rolling bearing having a metal material evaluated for this fatigue limit surface pressure P _{max 1im} , it is possible to realize a bearing life in an extremely long life region.

It should be noted that the stress governing the fatigue fracture of the material is either normal stress or shear stress. Several years have passed since the ultrasonic axial load fatigue tester (full swing) was put on the market to evaluate fatigue characteristics due to normal stress at high speed. On the other hand, there has been little research on ultrasonic torsional fatigue tests to evaluate fatigue characteristics due to shear stress at high speed, and materials evaluated so far have a maximum shear stress amplitude (full swing) of 250 MPa or less. It is a mild steel or aluminum alloy that undergoes fatigue failure. On the other hand, the fatigue limit surface pressure of rolling bearings defined in ISO-281: 2007, which is the standard of dynamic load rating and rated life of rolling bearings, is 1500 MPa. The maximum alternating shear stress amplitude acting on τ ₀ = 375 MPa. Therefore, an ultrasonic torsion tester that can be evaluated with a maximum shear stress amplitude of 375 MPa or more is required, but there is no example of an ultrasonic torsion tester that can be evaluated with such a large maximum shear stress amplitude. Therefore, the present invention develops an ultrasonic torsion tester and determines the maximum contact surface pressure P _max when a load is applied in which the maximum alternating shear stress amplitude τ ₀ acting inside the surface layer is equal to the shear fatigue strength τ _1im. This is based on a comprehensive idea with the knowledge that the fatigue limit surface pressure P _max can be estimated as 1 _im .

In the method of the present invention, in the process of determining the shear fatigue strength, a value of 85% with respect to the shear fatigue strength determined according to the predetermined criteria may be used as the value of the shear fatigue strength τ _1im used in the fatigue limit surface pressure calculation process.
This is because in the ultrasonic torsional fatigue test, when the volume subjected to a large load (dangerous volume) is substantially equal to the conventional fatigue test, the shear fatigue strength tends to be evaluated higher.

In the method of the present invention, in the shear fatigue strength determination process, a value of 80% with respect to the shear fatigue strength determined according to the predetermined criteria may be used as the value of the shear fatigue strength τ _1im used in the fatigue limit surface pressure calculation process.
In the torsional fatigue test, the shear stress is maximum on the specimen surface and zero on the shaft core. That is, a fatigue test with a stress gradient. Of the tensile and compression fatigue tests, the axial load fatigue test is known to have a uniform normal stress in the cross section of the smooth part and show a constant fatigue limit regardless of the diameter of the smooth part. In the fatigue test, it is known that the fatigue limit decreases as the diameter of the smooth portion increases, and exhibits a dimensional effect that gradually approaches the fatigue limit in the axial load fatigue test. According to Non-Patent Document 4, the fatigue limit in the axial load fatigue test is about 80% of the fatigue limit in the rotating bending fatigue test with a smooth part diameter of 4 mm. As long as it has a stress gradient, dimensional effects are unavoidable even in torsional fatigue tests. Therefore, assuming that the standard of the tensile compression fatigue test can be applied to the torsional fatigue test as it is, the shearing strength using 80% of the shear fatigue strength determined according to the determined standard is used in the fatigue limit surface pressure calculation process. It is appropriate to use it as the value of the fatigue strength τ _1im .

In the method of the present invention, in the test process, the ultrasonic torsional fatigue test is performed a plurality of times to obtain a plurality of relationships between the shear stress amplitude of the metal material and the number of loads, and in the process of determining the shear fatigue strength, the plurality of times A _PSN diagram having an arbitrary failure probability is obtained from the relationship between the shear stress amplitude and the number of loads obtained in the test process, and from this _{PSN diagram, the} shear fatigue strength τ _{1im in the ultralong} life region is _obtained. You may make it decide.
The size effect that appears in fatigue tests with the above stress gradient is caused by a mechanical factor called stress gradient and a statistical factor that increases or decreases the volume subjected to a large load (dangerous volume). From the viewpoint of statistical factors, a PSN diagram may be obtained by evaluating multiple lines at multiple stress levels.

In this case, in the process of determining the shear fatigue strength, a value of 85% with respect to the shear fatigue strength in the ultra-long life region determined from the PSN diagram is used in the process of calculating the fatigue limit surface pressure. It is good also as a value of fatigue strength (tau) _1im .
In order to estimate it most safely, in the same manner as described above, 85% of the value of the shear fatigue strength in the ultra-long life region determined from the PSN diagram, and a value obtained by further adding 80% to the fatigue limit surface pressure. It is preferable to _set the value of the shear fatigue strength τ _1im used in the calculation process.

In the present invention, in order to safely estimate the absolute value of the shear fatigue strength, the ultrasonic torsional fatigue test is performed a plurality of times in the test process to obtain a plurality of relationships between the shear stress amplitude of the metal material and the number of loads, In the process of determining the shear fatigue strength, a PSN diagram of an arbitrary fracture probability is obtained from the relationship between the shear stress amplitude and the number of loads obtained in the plurality of test processes, and from this PSN diagram, The fracture probability correction, which is a correction for determining the shear fatigue strength τ _1im in the ultra-long life region, and a value of 85% with respect to the shear fatigue strength determined according to the determined criteria in the process of determining the shear fatigue strength, and overestimation correction is a correction to the value of the shear fatigue strength tau _HM used in limited surface pressure calculation process, put in the ultra-long life region decided by the shear fatigue strength determination process 80% of the value for the shear fatigue strength, of the three correction with correction at a size effect correction to a value of shear fatigue strength tau _HM used in the fatigue limit contact pressure calculation process, any two or more correction The _shear fatigue strength τ _1im obtained in combination may be regarded as an absolute value. Thus, by combining two or more corrections, the shear fatigue strength can be safely estimated, and the fatigue limit surface pressure can be estimated more safely.

  In the method of the present invention, the ultrasonic torsional fatigue test includes, for example, a torsional vibration converter that generates a torsional vibration that rotates forward and backward around a rotation center axis when AC power is applied, and a test piece concentrically at the tip. A mounting portion to which the vibration converter is attached and fixed to the torsional vibration converter at the base end, and an amplitude expansion horn that expands the amplitude of the torsional vibration of the vibration converter applied to the base end, The shape and dimensions resonate with the vibration of the amplitude expanding horn driven by the torsional vibration converter, and the vibration converter is driven at a frequency in the ultrasonic region to resonate the test piece with the vibration of the amplitude expanding horn to cause shear fatigue destruction. By doing.

In the method of the present invention, in the test process, the test piece may be forcibly air-cooled in order to suppress heat generation of the metal material test piece in the ultrasonic torsional fatigue test. Moreover, in order to suppress the heat generation of the test piece, the load and the pause may be alternately repeated. In the test process, a continuous load may be applied in a low load region where heat generation of the test piece of the metal material does not become a problem in the ultrasonic torsional fatigue test.
In the present invention, an ultrasonic torsional fatigue test capable of applying a load at high speed is used. For example, an ultrasonic torsional fatigue test with an extremely high excitation frequency of 20000 Hz is performed. As a result, if the vibration is continuously applied, the load number reaches 10 ⁹ times in just half a day. However, when the sample is continuously vibrated with a somewhat high shear stress amplitude, the test piece generates heat, and it is impossible to obtain a precise relationship between the shear stress amplitude and the number of loads. Therefore, it is preferable to forcibly air-cool the test piece. If the heat generation of the test piece is not sufficiently suppressed by forced air cooling alone, it is preferable to alternately repeat excitation and pause. Although the actual load frequency is reduced by resting, if an ultrasonic torsional fatigue tester with an excitation frequency of 20000 Hz is used, the suspension time is about 2000 Hz even if the suspension time is about 10 times the excitation time. If there is a week, it reaches 10 ⁹ loadings.
In the method of the present invention, the metal material may be any one of SNCM420, SUJ2, SUJ3, and SCr420, and the test piece may be a heat-treated product.

  In the method for selecting a rolling bearing material for a railway vehicle according to the present invention, the fatigue limit surface pressure estimated by the fatigue limit surface pressure estimation method according to any one of the above configurations of the present invention is equal to or greater than a predetermined fatigue limit surface pressure. A metal material is used as a material for a bearing ring or rolling element of a rolling bearing for a railway vehicle.

  According to the fatigue limit surface pressure estimation method of the present invention, the fatigue limit surface pressure of a metal material for a rolling bearing can be accurately estimated from the result of a short-time fatigue test. Therefore, fatigue limit surface pressure can be adopted as one of the test items of the material used for the bearing ring or rolling element of the rolling bearing. By using only a material having a fatigue limit surface pressure obtained by an actual fatigue test that is equal to or higher than a predetermined fatigue limit surface pressure as a bearing material, it greatly helps to improve the reliability of rolling bearings for railway vehicles. Employing fatigue limit surface pressure as one of the test items for the materials used has been difficult for the past because it took many years to test and was far from the actual situation. Yes, it can be used to improve bearing reliability. In addition, what is necessary is just to set "the defined fatigue limit surface pressure" used as a criterion suitably according to the objective. In addition, the fatigue limit surface pressure is estimated, for example, for each lot of material, for each purchased amount, for each supplier, and the like.

The apparatus for estimating the fatigue limit surface pressure of a rolling bearing material for a railway vehicle according to the present invention is an apparatus for estimating the fatigue limit surface pressure P _{max 1im} of a metal material used as a bearing ring or rolling element of a rolling bearing for a railway vehicle. ,
An input means 22 for storing the relationship between the shear stress amplitude of the metal material and the number of loads obtained by a complete double-swing ultrasonic torsional fatigue test in a predetermined storage area;
A shear fatigue strength determining means 23 for determining the shear fatigue strength τ _1im in the ultra-long life region from the relationship between the stored shear stress amplitude and the number of loads, according to a predetermined criterion;
Maximum acting on the inside of the surface of the metal material determined by the shape and size of the surfaces of the object M1 made of the metal material and the object M2 which is in rolling contact with the object M1 and the contact surface pressure. alternating shear stress amplitude tau ₀ is the maximum contact surface pressure determined by P _max were determined equation, the fatigue limit surface the maximum contact surface pressure P _max when the load equal to the shear fatigue strength tau _HM acts Fatigue limit surface pressure calculating means 24 that is an estimated value of the pressure P _{max 1im} is provided.
The metal material may be rolling bearing steel used as a rolling ring or rolling element of a rolling bearing for a railway vehicle. The input means 22 uses, for example, an input device that performs manual input such as a keyboard, a reading device for a recording medium, a communication network, and the like, for example, a file that summarizes the relationship between the shear stress amplitude of the metal material and the number of loads. It is a means for memorize | storing so that a predetermined | prescribed storage area or its memory location can be specified for later calculation.

According to the apparatus of the present invention, the ultrasonic torsional fatigue test capable of extremely high speed load can be used as described for the method of the present invention, and the shear stress amplitude and load frequency of the rolling bearing steel can be measured in a short period of time. The relationship can be obtained, and the fatigue limit surface pressure P _{max 1im} can be accurately estimated.
When it can be regarded as line contact, the determined calculation formula in the fatigue limit surface pressure calculation means 24 is, for example,
(Fatigue limit surface pressure P _{max 1im} ) = 4 × (shear fatigue strength τ _1im )
And

In the present invention device, the input means 22 has a function of storing the relationship between the shear stress amplitude of the metal material and the number of loads obtained in each ultrasonic torsional fatigue test in a plurality of times in a predetermined storage area, The shear fatigue strength determining means 23 obtains a PSN diagram having an arbitrary fracture probability from the relationship between the shear stress amplitude and the number of loads in the plurality of tests, and from the PSN diagram, The shear fatigue strength τ _{1im in the ultralong} life region may be determined.

The system for estimating the fatigue limit surface pressure of rolling bearing materials for railway vehicles according to the present invention is the complete torsional ultrasonic torsional fatigue test of a test piece 1 of a metal material that becomes a race or rolling element of a rolling bearing for railway vehicles. An ultrasonic torsional fatigue testing machine main body 3, a testing machine control device 4 for controlling the ultrasonic torsional fatigue testing machine main body 3 in accordance with input test conditions, and a fatigue limit surface having any one of the above-described configurations of the present invention This is a system including a pressure estimation device 5.
In this system as well, the ultrasonic torsional fatigue test capable of extremely high-speed loading can be used as described for the method of the present invention, and the shear stress amplitude and the number of loads of rolling bearing steel can be measured in a short period of time. The relationship can be obtained, and the fatigue limit surface pressure P _{max 1im} can be accurately estimated.

An estimation method, an estimation apparatus, and an estimation system for a rolling bearing material for rolling stock according to the present invention obtain a relationship between a shear stress amplitude of a metal material and the number of loads by an ultrasonic torsional fatigue test. decide shear fatigue strength tau _HM in life region, maximum contact surface when the maximum alternating shear stress amplitude tau ₀ is equal to the shear fatigue strength tau _HM load acting on the surface layer inside the contact dimension specifications of the metallic material acts Since the pressure P _max is estimated as the fatigue limit surface pressure P _{max 1 im} , an ultrasonic torsional fatigue test capable of _applying a very high speed can be used, and a metal material having high fatigue strength such as a steel for rolling bearings. However, the relationship between the shear stress amplitude and the number of loads can be obtained in a short period of time, and the fatigue limit surface pressure P _{max 1im} can be accurately estimated.

  In the method for selecting a rolling bearing material for a railway vehicle according to the present invention, the fatigue limit surface pressure estimated by the fatigue limit surface pressure estimation method according to any one of the above configurations of the present invention is equal to or greater than a predetermined fatigue limit surface pressure. Since the metal material is used as the material for the bearing ring or rolling element of the rolling bearing for the railway vehicle, the reliability of the rolling bearing for the railway vehicle can be improved by adopting test items that were not previously thought of.

(A) is a flow chart showing a fatigue limit surface pressure estimation method according to an embodiment of the present invention, (B) is a schematic diagram of the fatigue limit surface pressure estimation system, (C) is a fatigue strength diagram and shear fatigue Explanatory drawing which shows intensity _| strength (tau) _1im , (D) is sectional drawing of the object manufactured with the metal material used as a test object, and the object which touches this. It is a block diagram of the estimation system of the fatigue limit surface pressure. It is a conceptual diagram of a testing machine control device and a fatigue limit surface pressure estimation device in the fatigue limit surface pressure estimation system. It is a block diagram which shows the conceptual structure of the estimation apparatus of fatigue limit surface pressure. It is a front view of the main body of an ultrasonic torsional fatigue testing machine. It is a schematic diagram of a test piece. It is a front view of a test piece. 6 is a graph showing an axial distribution of torsion angle θ and surface shear stress τ (when end surface torsion angle θ _end is 0.01 rad). It is a microscope picture which shows the test piece shoulder part cylindrical surface lower end at the time of stationary. It is a microscope picture which shows the test piece shoulder part cylindrical surface lower end at the time of vibration. It is explanatory drawing which shows the relationship between the range 2a of FIG. 10, and end surface twist angle | corner (theta) _end . It is a graph which shows the relationship between amplifier output P and end surface twist angle | corner (theta) _end . It is the microscope picture of the example of the test piece which carried out torsional fatigue destruction, and explanatory drawing of the whole test piece. It is a front view of the test piece evaluated with a hydraulic servo type torsional fatigue testing machine. It is a graph which shows the relationship between the shear stress amplitude obtained by the ultrasonic torsional fatigue test and the number of loads, and the SN diagram (solid line). FIG. 16 is a graph showing a P-S-N diagram (broken line) having a fracture probability of 10% and an original S-N diagram (solid line) obtained from the relationship of FIG. Description of the distribution of alternating shear stress τ _yz and normal stress σ _{z in} the depth direction ( _y : circumferential direction, _z : depth direction) when P _max = 1500 MPa acts in the line contact state FIG. It is the microscope picture of the circumferential cross section which shows the micro crack parallel to the surface seen around the depth where the absolute value of an alternating shear stress becomes the maximum. It is explanatory drawing which shows the example of a test condition input screen of the control apparatus of an ultrasonic torsional fatigue testing machine. 5 is a flowchart of details of a test process. It is a schematic diagram of the fatigue test result regarding how to determine the fatigue limit. It is a schematic diagram which shows that the intensity distribution in arbitrary load frequency | counts follows a normal distribution, and a standard deviation is the same. This is a method for obtaining a PSN diagram (in the case of a continuous decline type curve model, the destruction probability is 10%). It is a figure which shows the shear fatigue characteristic of the test piece of SNCM420 carburizing. It is a figure which shows the shear fatigue characteristic of the test piece of SUJ3 standard.

An embodiment of the present invention will be described with reference to the drawings. The following description also includes a description of a method for selecting a rolling bearing material for a railway vehicle. This method of estimating the fatigue limit surface pressure of a rolling contact metal material is a method of estimating the fatigue limit surface pressure P _{max 1im} of a metal material that will be a race or rolling element of a rolling bearing for a railway vehicle. As in A), a test process (S1), a shear fatigue strength determination process (S2), and a fatigue limit surface pressure calculation process (S3) are included. The rolling bearing of the railway vehicle is a bearing that supports an axle of the railway vehicle, for example. The metal material is, for example, rolling bearing steel that serves as a bearing ring or rolling element of a rolling bearing.

  The test process (S1) is a process of obtaining the relationship between the shear stress amplitude of the metal material and the number of loads by a complete double-sided ultrasonic torsional fatigue test. In this test, an ultrasonic torsional fatigue testing machine 2 that applies a complete double-sided ultrasonic torsional vibration to the metal material test piece 1 shown in FIG. This ultrasonic torsional fatigue testing machine 2 uses an extremely high-speed ultrasonic torsional fatigue test (full swing) with an excitation frequency of 20000 Hz. This ultrasonic torsional fatigue testing machine 2 cannot be used as it is, and has various improvements.

In the shear fatigue strength determination process (S2), the shear fatigue strength τ _1im in the ultra-long life region is determined according to a predetermined standard from the relationship between the shear stress amplitude and the number of loads obtained in the test process (S1). The above-mentioned “shear fatigue strength in the ultra-long life region” refers to the “shear fatigue limit”, but in this specification, it will be described as “shear fatigue strength in the ultra-long life region”. The above-mentioned “defined standard” in the shear fatigue strength determination process (S2) is, for example, a curve obtained by fitting the relationship between the shear stress amplitude of the test results and the number of loads to an established theoretical curve indicating the shear fatigue strength. And the shear fatigue strength is determined from the curve. Specifically, the SN diagram (fatigue strength diagram with 50% fracture probability) obtained by applying to the fatigue limit type polyline model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials Science (See FIG. 5C). The SN diagram may be obtained by applying not only to the fatigue limit type broken line model but also to a continuously decreasing curve model. However, in that case, it is necessary to define, for example, “τ _1im is a value on the SN diagram at 10 ¹⁰ times”.

ここで、Ａ、Ｂ、Ｄは定数である。
疲労強度、疲労寿命にはバラツキがある。本来、確率疲労特性は、複数の応力振幅で複数個の試験片を評価し、ある破壊確率におけるＰ−Ｓ−Ｎ線図を求めて評価する（非特許文献５参照）。しかしながら、Ｐ−Ｓ−Ｎ線図を求めるには多大な工数と時間を要する。金属材料疲労信頼性評価標準JSMS-SD-6-02では、Ｓ−Ｎ線図から任意の破壊確率におけるＰ−Ｓ−Ｎ線図を求める方法が提案されている。それは、図２２のように、任意の疲労寿命における強度分布は正規分布に従い、その標準偏差σは一定と仮定する。
得られたＳ−Ｎ線図を破壊確率５０％の疲労強度曲線とする。疲労限度型折れ線モデルでは時間強度部（傾斜直線部）の破損データ、連続低下型曲線モデルは全範囲の破損データを対象とする。図２３は連続低下型曲線モデルの例である。直線または曲線に沿って個々の破損データを任意の疲労寿命に平行移動し、それらが正規分布するとして標準偏差を求める。例えば、得られた標準偏差をｓとすると、破壊確率５０％の疲労強度曲線を１．２８２ｓだけ下に平行移動したものが破壊確率１０％のＰ−Ｓ−Ｎ線図となる。 The fatigue limit type polyline model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials Science, is regressed by applying the following equation.
σ = −Alog ₁₀ N + B (N <N _W )
σ = E (N ≧ N _W )
Here, A, B, E, and _Nw are constants. The fatigue limit (E in the above equation) is estimated as follows when there is one or more censored data at a load count of N = 5 × 10 ⁶ or more. The average value of the fracture data stress minimum value σ _{f min} and the censored data stress maximum value σ _{r max} lower than this is defined as the fatigue limit (see FIG. 21). If censored data exists at the same stress level as σ _{f min} and there is no censored data at a stress level lower than this, this σ _{f min} is used as the fatigue limit. After determining the fatigue limit in this way, this value is fixed and other parameters in the above equation are estimated from only the fracture data.
The continuously decreasing curve model is regressed by applying the following equation which is a basic equation of Stromeyer.

Here, A, B, and D are constants.
There are variations in fatigue strength and fatigue life. Originally, the probability fatigue characteristic is evaluated by evaluating a plurality of test pieces with a plurality of stress amplitudes and obtaining a PSN diagram at a certain fracture probability (see Non-Patent Document 5). However, it takes a lot of man-hours and time to obtain the PSN diagram. In the metal material fatigue reliability evaluation standard JSMS-SD-6-02, a method for obtaining a PSN diagram at an arbitrary fracture probability from an SN diagram is proposed. As shown in FIG. 22, it is assumed that the strength distribution in an arbitrary fatigue life follows a normal distribution and that the standard deviation σ is constant.
Let the obtained SN diagram be a fatigue strength curve with a fracture probability of 50%. In the fatigue limit type broken line model, the damage data of the time-strength part (inclined straight line part), and in the continuous decline type curve model, the damage data of the whole range are targeted. FIG. 23 is an example of a continuously decreasing curve model. Translate individual failure data along a straight line or curve to any fatigue life and determine the standard deviation as they are normally distributed. For example, when the obtained standard deviation is s, a P-S-N diagram with a fracture probability of 10% is obtained by translating a fatigue strength curve with a fracture probability of 50% downward by 1.282s.

In the fatigue limit surface pressure calculation process (S3), the contact dimensions of the object M1 made of the metal material (FIG. (D)) and the surface of the object M2 which is in rolling contact with the object M2 are in contact with each other When the load is applied such that the maximum alternating shear stress amplitude τ ₀ acting inside the surface layer of the object M1 of the metal material determined from the shape and size) and the load that gives the contact surface pressure is equal to the shear fatigue strength τ _1im The maximum contact surface pressure P _max is obtained by a predetermined calculation formula, and this maximum contact surface pressure P _max is _{used as} an estimated value of the fatigue limit surface pressure P _{max 1im} .
When the metal material is steel for rolling bearings, the object M1 manufactured with the metal material is a race or rolling element of a rolling bearing. This rolling bearing may be a ball bearing or a roller bearing.

The above-mentioned “determined calculation formula” used in the fatigue limit surface pressure calculation process (S3) is described in Non-Patent Document 3 described above. FIG. 5.13 of Non-Patent Document 3 is the circumferential distribution of the alternating shear stress at the depth at which the alternating shear stress acting below the contact surface in the line contact state is maximum, and is four times the maximum alternating shear stress τ ₀ . Is equal to the maximum contact surface pressure _Pmax . Therefore, in the case of line contact,
(Fatigue limit surface pressure P _{max 1im} ) = 4 × (shear fatigue strength τ _1im )
It becomes.
With respect to the major axis radius a and the uniaxial radius b of the contact ellipse, the line contact state is b / a = 0, and in this case, four times τ ₀ is equal to P _max as described above. The proportional constant between τ ₀ and P _max when b / a ≠ 0 is shown in FIG. 5.14 of Non-Patent Document 3.

According to the estimation method of this embodiment, since the fatigue test is performed by an ultrasonic torsional fatigue test, an extremely high-speed load is possible, and the relationship between the shear stress amplitude of the metal material and the number of loads can be obtained in a short time. The shear fatigue strength τ _1im in the ultra-long life region is determined from the relationship thus obtained, and the maximum alternating shear stress amplitude τ ₀ acting on the inside of the surface layer from the contact dimension specifications of the metal material becomes equal to the shear fatigue strength τ _1im. to estimate the maximum contact surface pressure _{P max} when the load acts as a fatigue limit surface pressure _{P max HM,} it is possible to precisely estimate the fatigue limit surface pressure _{P max HM} from the results of the torsional fatigue test. For this reason, when estimating the fatigue limit surface pressure P _{max 1im} of the rolling bearing steel, which is a material having a strong shear fatigue strength τ _1im , the effect that only a short test is required is more effectively exhibited. The Therefore, the shear fatigue characteristics of the metal material used as the rolling ring or rolling element of the rolling bearing for a railway vehicle can be obtained, and the fatigue limit surface pressure P _{max 1im} can be evaluated from the obtained shear fatigue characteristics. By using the metal material evaluated for the fatigue limit surface pressure P _{max 1im} for the bearing ring or rolling element of the rolling bearing, it is possible to realize a bearing life in an extremely long life region.

In this embodiment, as described above, the relationship between the shear stress amplitude of the rolling bearing steel and the number of loads is determined in a short period of time by an ultrasonic torsional fatigue test with an extremely high excitation frequency of 20000 Hz and a very high speed. The shear fatigue strength (or shear fatigue limit) τ _1im in the long-life region is determined, and the alternating shear stress amplitude τ ₀ acting on the inside of the surface layer from the contact dimension specifications of the rolling bearing
The maximum contact surface pressure _{P max} is estimated as fatigue limit surface pressure _{P max HM} when equal loads to shear fatigue strength tau _HM acts. For example, if the vibration continuous pressurization at 20000 Hz, and reaches the slight load times at ¹⁰⁹ half day or so. However, since the test piece 1 generates heat when continuously vibrated with a somewhat high shear stress amplitude, the test piece 1 needs to be cooled and forced air cooling is performed. If the heat generation of the test piece 1 is not sufficiently suppressed by forced air cooling alone, vibration and pause are alternately repeated. Although the actual load frequency is reduced by resting, if the tester 2 has an excitation frequency of 20000 Hz, even if the pause time is about 10 times the excitation time, it is still about 2000 Hz, and there is a week. if it reaches the 10 ⁹ times of load times.

It should be noted that the stress governing the fatigue fracture of the material is either normal stress or shear stress. Several years have passed since the ultrasonic axial load fatigue tester (full swing) was put on the market to evaluate fatigue characteristics due to normal stress at high speed. On the other hand, there has been little research on ultrasonic torsional fatigue tests to evaluate fatigue characteristics due to shear stress at high speed, and materials evaluated so far have a maximum shear stress amplitude (full swing) of 250 MPa or less. It is a mild steel or aluminum alloy that undergoes fatigue failure. On the other hand, the fatigue limit surface pressure of rolling bearings defined in ISO-281: 2007, which is the standard of dynamic load rating and rated life of rolling bearings, is 1500 MPa. The maximum alternating shear stress amplitude acting on τ ₀ = 375 MPa. Therefore, an ultrasonic torsion tester that can be evaluated with a maximum shear stress amplitude of 375 MPa or more is required, but there is no example of an ultrasonic torsion tester that can be evaluated with such a large maximum shear stress amplitude. Therefore, the present invention develops an ultrasonic torsion tester and determines the maximum contact surface pressure P _max when a load is applied in which the maximum alternating shear stress amplitude τ ₀ acting inside the surface layer is equal to the shear fatigue strength τ _1im. This is based on a comprehensive idea with the knowledge that the fatigue limit surface pressure P _max can be estimated as 1 _im .

According to this embodiment, the rolling bearing material for a railway vehicle is selected so that the fatigue limit surface pressure estimated by the fatigue limit surface pressure estimation method according to any one of the above configurations of the present invention is equal to or greater than a predetermined fatigue limit surface pressure. A certain metal material is used as a material for a bearing ring or rolling element of a rolling bearing for a railway vehicle.
According to the fatigue limit surface pressure estimation method of this embodiment, the fatigue limit surface pressure of a metal material for a rolling bearing can be accurately estimated from the results of a short-time fatigue test. Therefore, fatigue limit surface pressure can be adopted as one of the test items of the material used for the bearing ring or rolling element of the rolling bearing. By using only a material having a fatigue limit surface pressure obtained by an actual fatigue test that is equal to or higher than a predetermined fatigue limit surface pressure as a bearing material, it greatly helps to improve the reliability of rolling bearings for railway vehicles. Employing fatigue limit surface pressure as one of the test items for the materials used has traditionally required many years of testing and was too far from the actual situation, but this method can be put to practical use. By adopting it, it can be used to improve the reliability of the bearing. In addition, what is necessary is just to set "the defined fatigue limit surface pressure" used as a criterion suitably according to the objective. In addition, the fatigue limit surface pressure is estimated, for example, for each lot of material, for each purchased amount, for each supplier, and the like.

  FIG. 2 shows a conceptual configuration of a system for estimating a fatigue limit surface pressure of a rolling contact metal material used in the estimation method. This estimation system includes an ultrasonic torsional fatigue testing machine 2 and a fatigue limit surface pressure estimation device 5 that performs the processes of the shear fatigue strength determination process (S2) and the fatigue limit surface pressure calculation process (S3) of FIG. Is done.

  In FIG. 2, the ultrasonic torsional fatigue testing machine 2 includes a testing machine main body 3 and a testing machine control device 4. The testing machine main body 3 is attached to a torsional vibration converter 7 installed on the upper part of the frame 6 with an amplitude-amplifying horn 8 projecting downward, and a test piece 1 is detachably attached to the tip thereof, and The sound wave vibration is expanded and transmitted to the test piece 1 as vibration in the forward / reverse rotation direction around the axis of the amplitude expansion horn 8.

The test machine control device 4 includes a computer 10 and a test machine control program 11 that can be executed by the computer 10. The computer 10 is a desktop personal computer or the like, and includes a central processing unit 12, storage means 13 such as a memory, and an input / output interface 14. The tester control program 11 is stored in the storage means 13, and the remaining storage area of the storage means 13 becomes a data storage area 13a or a work area. In addition, an input device 15 such as a keyboard and a mouse, and a display device such as a liquid crystal display device and an output device 16 such as a printer are provided as a part of the computer 10 or connected to the computer 10. ing.
The test machine control device 4 is a device that controls the torsional vibration converter 7 of the test machine body 3, and a control output is given from the input / output interface 14 to the vibration converter 7 via the amplifier 17. The test machine control device 4 performs the following processing according to the test machine control program 11. First, as shown in a screen example in FIG. 19, a display device serving as an output device 16 displays a screen prompting input of test conditions (output, intermittent operation or continuous operation, test end condition, data collection condition, etc.). When the test condition is input from the input device 15 and a test start command is input, the tester body 3 is driven and controlled according to the input condition. Note that the value of the maximum shear stress amplitude is converted and displayed with respect to the input output P by the equation (9) described later.

  In FIG. 2, the fatigue limit surface pressure estimation device 5 includes a computer 10 and a fatigue limit surface pressure estimation program 19 executable by the computer 10. The computer 10 may be the same as or different from the computer constituting the tester control device 4, and includes a central processing unit 12, storage means 13 such as a memory, and an input / output interface 14. . The input device 15 and the output device 16 are provided as a part of the computer 10 or connected to the computer 10. FIG. 3 shows an example in which the tester control program 11 and the fatigue limit surface pressure estimation program 19 are stored in the same computer 10 and used as the tester control device / fatigue limit surface pressure estimation device 9.

The fatigue limit surface pressure estimation device 5 includes a computer 10 and the fatigue limit surface pressure estimation program 11 and each means shown in a conceptual configuration in FIG. The fatigue limit surface pressure estimation device 5 is a device for estimating the fatigue limit surface pressure P _{max 1im} of a metal material that is in rolling contact, and includes an input means 22, a shear fatigue strength determination means 23, and a fatigue limit surface pressure calculation means. 24, and also comprises a storage means 13 and an output means 4A.

The input means 22 is a means for storing the relationship between the shear stress amplitude of the metal material and the number of loads obtained by a complete double swing ultrasonic torsional fatigue test in a predetermined storage area of the storage means 13. The input means 22 is a file that summarizes the relationship between the shear stress amplitude of the metal material and the number of loads, for example, using an input device for manual input such as a keyboard, a reading device for a recording medium, a communication network, etc. Is stored so that a predetermined storage area or its storage location can be specified for later calculation.
The shear fatigue strength determining means 23 is means for determining the shear fatigue strength τ _1im in the ultra-long life region according to a predetermined standard from the relationship between the shear stress amplitude stored in the storage region and the number of loads. The specific processing performed by the shear fatigue strength determining means 23 is as described for the fatigue limit surface pressure calculation process (S3) in FIG.
The fatigue limit surface pressure calculating means 24 is determined by the shape and size of the surfaces of the object M1 made of the metal material and the object M2 that is in rolling contact with the object M1 and the load that gives the contact surface pressure. The maximum alternating shear stress amplitude τ ₀ acting inside the surface layer of the metal object M1 is obtained by a predetermined calculation formula for the maximum contact surface pressure P _max when the load is applied which is equal to the shear fatigue strength τ _1im. The maximum contact surface pressure P _max is a means for _setting the estimated value of the fatigue limit surface pressure P _{max 1im} . The specific processing performed by the fatigue limit surface pressure calculating means 24 is as described for the fatigue limit surface pressure calculation process (S3) in FIG.

  Next, details of the ultrasonic torsional fatigue testing machine 2 and details of a method for estimating the fatigue limit surface pressure will be described. This ultrasonic torsional fatigue testing machine 2 is designed as a complete double swing ultrasonic torsional fatigue testing machine that gives shear fatigue to rolling bearing steel at an extremely high speed. The excitation frequency range of the torsional vibration converter 7 is 20000 ± 500 Hz. Note that while there are various types of longitudinal vibration converters used in the ultrasonic axial load fatigue test, commercially available torsional vibration converters have only low output, and it is virtually impossible to make them yourself. It was. Therefore, it is necessary to optimize the shapes of the amplitude expanding horn 8 and the test piece 1 to give torsional fatigue to the high-strength rolling bearing steel.

The amplitude expansion horn 8 is of an exponential function type, and the diameter of the large-diameter side end face fixed to the torsional vibration converter 7 is 38 mm, and the diameter of the small-diameter side end face fixing the test piece 1 is 13 mm. It is designed and adjusted so that the enlargement ratio (ratio of the torsion angle on the small diameter side to the torsion angle on the large diameter side) is as large as possible and resonates in the vicinity of 20000 Hz. The large-diameter end face of the amplitude expanding horn 8 is provided with a male thread portion protruding in the axial direction for fixing to the torsional vibration converter, and the small-diameter end face is provided with a female thread for fixing the test piece. ing. The material of the amplitude expanding horn 8 is a titanium alloy. As a result of actually measuring Young's modulus E, Poisson's ratio ν, and density ρ, they were E = 1.16 × 10 ¹¹ Pa, ν = 0.27, and ρ = 4460 kg / m ³ , respectively. Using the FEM analysis software (Marc Mentat 2008 r1) (registered trademark), eigenvalue analysis of free torsional resonance was performed using the above E 1, ν, and ρ as physical property values. As a result, the enlargement ratio was 43.1 times.

FIG. 6 shows a schematic diagram of the test piece. One end of the actual test piece 1 is provided with a male screw portion for fixing to the tip of the amplitude expanding horn 8. In FIG. 6, the test piece 1 is a dumbbell type, and is determined by a shoulder length L ₁ , a half chord length L ₂ , a shoulder radius R ₂ , a minimum radius R ₁ , and an arc radius R.

In designing the test piece 1, a half chord length L ₂ , a shoulder radius R ₂ , and a minimum radius R ₁ are appropriately given (all units are m), resonance frequency f (= 20000 Hz), Young's modulus E, Poisson Ratio ν, density ρ (measured values of standard heat-treated bearing steel SUJ2 are E = 2.04 × 10 ¹¹ Pa, ν = 0.29, ρ = 7800 kg / m ³ ), and are substituted into the following equations (1) to (6) if the shoulder length _{L 1} is obtained (in m). The arc radius R is obtained from R ₁ , R ₂ and L ₂ .

Here, L ₂ = 0.0065 m, R ₂ = 0.0045 m, and R ₁ = 0.002 m, which have been studied in advance so that as much shear stress as possible acts on the surface of the minimum diameter portion of the test piece, are changed to the above f, E, ν, Substituting into the equations (1) to (6) together with ρ results in L ₁ = 0.00753 m. However, when a test piece with L ₁ = 0.00753 m was manufactured using standard hardened and tempered bearing steel SUJ2, it did not resonate. Therefore, using the FEM analysis software (Marc Mentat 2008 r1) (registered trademark), eigenvalue analysis of free torsional resonance was performed using the above-mentioned f, E, ν, and ρ as physical properties. As a result, the frequency of torsional resonance at L ₁ = 0.00753 m was 19076 Hz, which was outside the range of 20000 ± 500 Hz, which is the excitation frequency range of the torsional vibration converter. Therefore, as a result of obtaining L ₁ that torsionally resonates at 20000 Hz, L ₁ = 0.00677 m. When a specimen with L ₁ = 0.00677m was made with the standard hardened and tempered bearing steel SUJ2, it resonated around 20000Hz. FIG. 7 shows a test piece drawing (unit: mm).

FIG. 8 shows a torsion angle θ and a surface shear stress τ obtained by eigenvalue analysis of free torsional resonance using the test piece model of FIG. FIG. 8 shows the case where the _end surface twist angle θ _end is 0.01 rad, and the maximum shear stress τ _max acting on the surface of the minimum diameter portion of the test piece at this time is 526.18 MPa.
That is, in the category of linear elasticity, the relationship between the torsion angle θ _{end of the end} surface and the maximum shear stress τ _max of the surface at the minimum diameter portion of the test piece is expressed by equation (7). However, the unit of τ _max is MPa, and θ _end is dimensionless.
τ _max = 52618θ _end (7)

Using three specimens 1 made of standard hardened and tempered bearing steel SUJ2 having the shape shown in FIG. 7, the end face twist angle θ _end was measured while changing the amplifier output P (%). Table 1 shows the alloy components of the specimen material. The hardness was 722HV.

  A photograph of the lower end of the shoulder of the test piece during vibration was taken at 200 times with a digital microscope (VHX-900 manufactured by KEYENCE). Prior to that, emery polishing (# 500, # 2000) and diamond wrapping (1 μm) were applied to the shoulder of the test piece with a drilling machine to obtain a mirror surface state. After attaching the test piece to the testing machine, a color check developer was applied to the shoulder. FIG. 9 is a photograph at rest, where there are places where the developer is not applied. The behavior at the time of vibration was observed in the uncoated areas. In the case of FIG. 9, attention is paid to the behavior of the part with an arrow. The amplifier output P was changed from 10% to 90% in 5% increments, and was shaken for 1 second, during which time a photo was taken at a shutter speed of 1/15 sec. FIG. 10 is a photograph taken during vibration with P = 50%, and the range 2a is the locus of the point of interest in FIG.

From the range 2a measured by changing the amplifier output P (%), the end surface torsion angle θ _end was obtained as shown in FIG. As a result, as shown in FIG. 12, all the three test pieces 1 showed almost the same linear relationship between P and θ _end , and equation (8) was obtained as a regression line.

From the equations (7) and (8), the relationship between the amplifier output P and the maximum shear stress amplitude τ _max of the surface at the minimum diameter portion of the test piece is as shown in the equation (9). From equation (9), τ _max = 951 MPa at P = 90%, and it is fully expected that torsional fatigue can be given to high strength rolling bearing steel.

  The manufactured ultrasonic torsional fatigue tester 2 controls the amplifier 17 by the tester control device 4 configured by the personal computer 10 and the tester control program 11 described above with reference to FIG. FIG. 19 shows a screen for inputting test conditions of the ultrasonic torsional fatigue testing machine 2. FIG. 20 is a detailed flowchart of the test process. In the test process, according to the input test conditions, control of amplifier output, control for selecting continuous oscillation or intermittent oscillation, information acquisition (frequency and amplifier) as shown in FIG. (Acquisition of status), control of the end of the test, etc. are performed.

  In the example of the input screen shown in FIG. 19, the fact that the resonance frequency is displayed as 19.97 in the column for measurement preparation indicates that the test piece resonated at 19.97 kHz with an output of 10%, which is almost equal to the target 20000 Hz. According to this tester control device 4, when an amplifier output is input in the measurement condition column, the maximum shear stress amplitude is converted from the slope and intercept of the straight line of equation (9) input in advance on the initial setting screen. In the same column, select either continuous operation where vibration is continued or intermittent operation where vibration and pause are alternately repeated.

  When a crack occurs and grows to a certain size, the resonance frequency of the test piece 1 decreases. The reason why 50.00 is entered in the frequency fluctuation range in the same column is to stop the test because the fatigue failure occurs when the resonance frequency is lowered by 50 Hz or more from the time of the test. This value is variable, and an appropriate value should be input according to the test piece material. FIG. 13 shows an example of a test piece subjected to torsional fatigue failure. It shows that an axial shear crack occurred, grew to a certain length, then shifted to a tensile mold and deviated obliquely.

The bearing steel SUJ2, which was standard hardened and tempered in a normal temperature atmosphere, was evaluated by intermittent operation in which vibration and pause were alternately repeated. Regardless of the magnitude of the maximum shear stress amplitude, the excitation time was consistently 110 msec and the rest time was 1100 msec. The test piece is the same lot as that used for the above-mentioned end face twist angle measurement. The test was terminated if no damage occurred ¹⁰ times.

FIG. 15 shows the relationship between the shear stress amplitude obtained by the ultrasonic torsional fatigue test and the number of loads. The solid line in Fig. 15 is the SN diagram (fatigue strength diagram with 50% fracture probability) obtained by fitting to the fatigue limit type broken line model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials Science. The shear fatigue limit was τ _1im = 564 MPa. Considering the line contact state, if τ _1im = 564 MPa is equal to the maximum alternating shear stress amplitude τ ₀ ,
(Fatigue limit surface pressure P _{max 1im} ) = 4 × (shear fatigue strength τ _1im )
, The fatigue limit surface pressure is estimated to be P _{max 1im} = 2256 MPa.
Note that the SN diagram may be obtained by applying a continuous decline type curve model instead of the fatigue limit type line model. In this case, however, it is necessary to define, for example, “τ _1im is a value on the SN diagram at 10 ¹⁰ times”.

A torsional fatigue test piece (standard quenching) using the bearing steel SUJ2 in Table 1 as a raw material and providing a thin part with the same minimum diameter of 4 mm as the ultrasonic torsional fatigue test piece at a parallel part of 10 mm in diameter as shown in FIG. (Temperature unit in the figure is mm). The reason why the thinned portion is provided is to make the dangerous volumes substantially equal. The torsional fatigue test piece in FIG. 14 has R = 11.4 mm, whereas the ultrasonic torsional fatigue test piece has R = 9.7 mm. The reason for changing R is to make the stress concentration factor uniform. Prior to the torsional fatigue test, emery polishing (# 500, # 2000) and diamond wrapping (particle size 1 μm) were applied to the thinned portion for the purpose of eliminating the influence of the surface roughness. The torsional fatigue test was performed with a hydraulic servo type torsional fatigue tester with a complete swing and a load frequency of 10 Hz. As a result, a white circle plot in FIG. 15 was obtained, and the time strength of the hydraulic servo torsional fatigue test result was about 15% lower than that of the ultrasonic torsional fatigue test result. The ultrasonic torsional fatigue test tends to be evaluated with higher shear fatigue strength than the conventional torsional fatigue test. Therefore, 479 MPa (broken line in FIG. 15), which is 85% of the shear fatigue limit 564 MPa obtained in the ultrasonic torsional fatigue test, is defined as τ _1im . In that case, if the line contact state is considered and τ _1im = 479 MPa is equal to the maximum alternating shear stress amplitude τ ₀ , the fatigue limit surface pressure is estimated to be P _{max 1im} = 1916 MPa.

  In the torsional fatigue test, the shear stress is maximum on the specimen surface and zero on the shaft core. That is, a fatigue test with a stress gradient. Here, in the tensile compression fatigue test, it is known that in the axial load fatigue test, the vertical stress in the cross section of the smooth portion is uniform and shows a constant fatigue limit regardless of the diameter of the smooth portion. On the other hand, in the rotating bending fatigue test having a stress gradient, it is known that the fatigue limit decreases as the diameter of the smooth portion increases, and a dimensional effect that gradually approaches the fatigue limit in the axial load fatigue test is known. There is a report that the three steel types with different tensile strengths were subjected to axial load fatigue tests and rotary bending fatigue tests with various smooth part diameters, and the respective fatigue limits were determined (see Non-Patent Document 4 above). According to this, regardless of the steel type, the fatigue limit in the axial load fatigue test is about 80% of the fatigue limit in the rotating bending fatigue test with a smooth part diameter of 4 mm.

In the tensile and compression fatigue test, the fatigue limit in the axial load fatigue test without a stress gradient is the safety standard, but in the torsional fatigue test, the stress gradient is maintained no matter how large the diameter of the smooth part is. not exist. As long as it has a stress gradient, dimensional effects are unavoidable even in torsional fatigue tests. Therefore, it is assumed that the standard of the tensile compression fatigue test can be applied as it is to the torsional fatigue test. That is, since the minimum diameter of the ultrasonic torsional fatigue test piece is 4 mm, 383 MPa (dotted line in FIG. 15), which is 80% of the shear fatigue limit 479 MPa corrected for overestimation in the ultrasonic torsional fatigue test, is τ _1im . . In this case, if the line contact state is considered and τ _1im = 383 MPa is equal to the maximum alternating shear stress amplitude τ ₀ , the fatigue limit surface pressure is estimated to be P _{max 1im} = 1532 MPa.

The size effect that appears in fatigue tests with the above stress gradient is caused by a mechanical factor called stress gradient and a statistical factor that increases or decreases the volume subjected to a large load (dangerous volume). From the viewpoint of statistical factors, a PSN diagram may be obtained by evaluating multiple lines at multiple stress levels. However, it will often be difficult to implement due to time constraints. The metal material fatigue reliability evaluation standard JSMS-SD-6-02 of the Japan Society of Materials was used to determine the shear fatigue limit τ _1im in FIG. It has a function to obtain a PSN diagram with a small number of data. FIG. 16 is a PSN diagram (broken line in FIG. 16) with a fracture probability of 10% obtained thereby, and the 10% shear fatigue limit was 500 MPa. When the overestimation correction of the ultrasonic torsional fatigue test is performed on the value, 500 × 0.85 = 425 MPa (dotted line in FIG. 16). Furthermore, when the above-described dimensional effect correction is performed, 425 × 0.8 = 340 MPa (the chain line in FIG. 16). This value is the safest estimate of τ _1im . Considering a line contact state, _assuming that τ _1im = 340 MPa, which is equal to the maximum alternating shear stress amplitude τ ₀ , the fatigue limit surface pressure is estimated as P _{max 1im} = 1360 MPa. Here, the appropriate failure probability is set to 10%. However, a reasonable failure probability should be considered by comparing the dangerous volume of the ultrasonic torsional fatigue test piece with the dangerous volume of the actual rolling bearing.

As described above, the relationship between the shear stress amplitude of the rolling bearing steel and the number of loads is obtained by an ultrasonic torsional fatigue test (full swing), and then the shear fatigue strength (or shear fatigue limit) τ _1im in the _ultralong life region is _calculated . Determine the maximum contact surface pressure P _max when a load is applied in which the maximum alternating shear stress amplitude τ ₀ acting inside the surface layer is equal to the shear fatigue strength τ _1im from the contact dimension specifications of the rolling bearing, and the fatigue limit surface pressure P _{max The method} of estimating as 1 _im was shown.

Incidentally, FIG. 17 shows the distribution of the alternating shear stress τ _{yz in} the circumferential section below the contact surface and the vertical stress σ _{z in} the depth direction when P _max = 1500 MPa acts in the line contact state ( _y : circumferential direction, _z : depth direction). The coordinates are made dimensionless by the single axis radius b of the contact ellipse. The alternating shear stress τ _yz has a maximum absolute value at the depth of the dotted line. FIG. 18 shows a microcrack parallel to the surface seen near the depth at which the absolute value of the alternating shear stress is maximum when the rolling fatigue test was stopped before peeling occurred and the circumferential cross section was observed. . The driving force developed parallel to the surface is considered as alternating shear stress. In other words, the crack propagation mode is mode II (in-plane shear type). As shown in FIG. 17, since the normal stress σ _{z in the} direction perpendicular to the crack surface is compression, there is no mode I type (tensile type), and σ _z interferes between the crack surfaces, so mode II Acts to hinder progress.

On the other hand, for mode II cracks (shear cracks in FIG. 13) that are generated and propagated in the ultrasonic torsional fatigue test, compressive stress perpendicular to the crack surface does not act. Therefore, it _can be said that the fatigue limit surface pressure P _max1im estimated from the shear fatigue strength τ _{1im in the ultralong} life region obtained by the ultrasonic torsional fatigue test gives a value lower than the actual value and a value on the safe side.

<Example 1>: Fatigue limit surface pressure P _{max 1im} of a metal material used as a race or rolling element of a rolling bearing for a railway vehicle is estimated.
Examples of the metal material include SNCM420, SUJ2, SUJ3, and SCr420. In Example 3, the shear fatigue characteristics of each test piece are obtained using a test piece obtained by heat-treating the SNCM420 material and a test piece obtained by heat-treating the SUJ3 material. Estimated. The test piece shown in FIG. 7 was used as each test piece.

Table 2 shows the alloy components of the SNCM420 material and SUJ3 material used for the test pieces.

  The SNCM420 material shown in Table 2 above was sequentially turned, heat treated, and ground to produce test pieces. Heat treatment in this case is carburizing quenching, secondary quenching, tempering (carburization: 920 ° C. × 4 h, RX gas atmosphere, carbon potential is maintained at 1.2 → diffusion: 920 ° C. × 3 h, RX gas atmosphere → heating: 800 ° C. X 70 min. → oil quenching → tempering: 180 ° C. x 120 min.

  In addition, a test piece was manufactured by sequentially turning the SUJ3 material shown in Table 2 to turning → heat treatment → grinding. The heat treatment in this case is so-called quenching and tempering (heating: 810 ° C. × 80 min. → oil quenching → tempering: 180 ° C. × 180 min.) For quenching the entire SUJ3 material.

FIG. 24 is a diagram showing shear fatigue characteristics of SNCM420 carburized specimens. The solid line in the figure is an SN diagram obtained by applying to the fatigue limit type broken line model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials, and the shear fatigue limit τ _w0 is 526 MPa. became. Fracture probability correction (fracture probability 10%), size effect correction, and overestimation correction were performed on the shear fatigue limit τ _w0 to determine the fatigue limit surface pressure P _{max 1im} in the line contact state. Table 3 shows an estimation result of the fatigue limit surface pressure P _{max 1im} .

FIG. 25 is a diagram showing shear fatigue characteristics of SUJ3 standard test pieces. The solid line in the figure is an SN diagram obtained by applying to the fatigue limit type broken line model of JSMS-SD-6-02, a metal material fatigue reliability evaluation standard of the Japan Society of Materials, and the shear fatigue limit τ _w0 is 547 MPa. became. Fracture probability correction (fracture probability 10%), size effect correction, and overestimation correction were performed on the shear fatigue limit τ _w0 to determine the fatigue limit surface pressure P _{max 1im} in the line contact state. Table 4 shows an estimation result of the fatigue limit surface pressure P _{max 1im} .

According to Example 1, a metal material that is used as a rolling ring or rolling element of a rolling bearing for a railway vehicle can be subjected to an extremely high speed load by performing a fatigue test by an ultrasonic torsional fatigue test. The relationship between the shear stress amplitude of each metal material and the number of loadings can be obtained in half a day to one week). From this relationship, the fatigue limit surface pressure P _{max 1im} can be accurately estimated. Therefore, the fatigue limit surface pressure can be adopted as one of the test items of the material used for the bearing ring or rolling element of the rolling bearing for the railway vehicle. By using only a material whose fatigue limit surface pressure obtained by actual fatigue tests is equal to or higher than a predetermined fatigue limit surface pressure as a bearing material, it greatly helps to improve the reliability of rolling bearings for railway vehicles. Employing fatigue limit surface pressure as one of the test items for the materials used has traditionally required many years of testing and was too far from the actual situation, but this method can be put to practical use. By adopting it, it can be used to improve the reliability of the bearing.

DESCRIPTION OF SYMBOLS 1 ... Test piece 2 ... Ultrasonic torsional fatigue testing machine 3 ... Ultrasonic torsional fatigue testing machine main body 4 ... Tester control device 5 ... Fatigue limit surface pressure estimation device 7 ... Torsional vibration converter 8 ... Amplitude expansion horn 10 ... Computer 11 ... Tester control program 17 ... Amplifier 19 ... Fatigue limit surface pressure estimation program 22 ... Input means 23 ... Shear fatigue strength determination means 24 ... Fatigue limit surface pressure calculation means M1 ... Object made of metal material M2 ... Object in contact

Claims (12)

A method for estimating a fatigue limit surface pressure Pmax lim of a metal material used as a bearing ring or rolling element of a rolling bearing for a railway vehicle,
A test process to obtain the relationship between the shear stress amplitude of metal materials and the number of loads by ultrasonic torsional fatigue test,
The shear fatigue strength determination process for determining the shear fatigue strength τlim in the ultra-long life region from the relationship between the obtained shear stress amplitude and the number of loads, according to a predetermined criterion,
Maximum alternating force acting on the surface of the object of the metal material determined from the shape and size of the surface of the object made of the metal material and the object that is in rolling contact with the object, the size and the load that gives the contact surface pressure The maximum contact surface pressure Pmax when the load is applied with the shear stress amplitude τ ₀ equal to the shear fatigue strength τlim is determined by a predetermined calculation formula, and the maximum contact surface pressure Pmax is determined from the fatigue limit surface pressure Pmax lim. Including fatigue limit surface pressure calculation process as an estimated value,
A method for estimating fatigue limit surface pressure of rolling bearing materials for railway vehicles. In Claim 1, the defined calculation formula in the fatigue limit surface pressure calculation process is:
(Fatigue limit surface pressure Pmax lim) = 4 × (shear fatigue strength τlim)
A method for estimating the fatigue limit surface pressure of rolling bearing materials for railway vehicles.   3. The test result according to claim 1, wherein the predetermined criterion for determining the shear fatigue strength τlim in the ultralong life region in the process of determining the shear fatigue strength is a fatigue limit type broken line model indicating the shear fatigue strength. A method for estimating the fatigue limit surface pressure of rolling bearing materials for rolling stock, which is a process for obtaining a curve in which the relationship between the shear stress amplitude and the number of loads is applied and obtaining the shear fatigue strength from the curve.   4. The continuous decreasing curve according to claim 1, wherein the predetermined criterion for determining the shear fatigue strength τlim in the ultralong life region in the process of determining the shear fatigue strength is a continuously decreasing curve indicating the shear fatigue strength. 5. A method for estimating the fatigue limit surface pressure of rolling bearing materials for rolling stock, which is a process for obtaining a curve in which the relationship between the shear stress amplitude and the number of loads in the test result is applied to a model and obtaining the shear fatigue strength from the curve.   5. The method according to claim 1, wherein in the test process, the ultrasonic torsional fatigue test is performed a plurality of times to obtain a plurality of relationships between the shear stress amplitude of the metal material and the number of loads, and the shear fatigue is performed. In the strength determination process, a PSN diagram having an arbitrary fracture probability is obtained from the relationship between the shear stress amplitude and the number of loads obtained in the plurality of test processes, and from the PSN diagram, A method for estimating the fatigue limit surface pressure of rolling bearing materials for rolling stock that determines the shear fatigue strength τlim in the long-life region.   6. The process according to claim 1, wherein a value of 85% with respect to the shear fatigue strength determined according to the predetermined standard is used in the fatigue limit surface pressure calculation process in the shear fatigue strength determination process. A method for estimating a fatigue limit surface pressure of a rolling bearing material for a railway vehicle having a shear fatigue strength τlim.   The shear fatigue according to any one of claims 1 to 6, wherein a value of 80% of the shear fatigue strength in the ultralong life region determined in the shear fatigue strength determination process is used in the fatigue limit surface pressure calculation process. A method for estimating the fatigue limit surface pressure of a rolling bearing material for rolling stock with a value of strength τlim.   5. The shear stress amplitude of a metal material according to claim 1, wherein, in order to safely estimate an absolute value of shear fatigue strength, the ultrasonic torsion fatigue test is performed a plurality of times in the test process. A plurality of relationships between the number of times of loading and the number of loading times, and in the process of determining the shear fatigue strength, a PSN diagram of an arbitrary fracture probability is obtained from the relationship between the shear stress amplitude and the number of times of loading obtained in the plurality of testing processes, From this PSN diagram, the fracture probability correction, which is a correction for determining the shear fatigue strength τlim in the ultra-long life region, and the shear fatigue strength determined according to the predetermined criteria in the process of determining the shear fatigue strength The overestimation correction, which is a correction to set the value of 85% to the value of the shear fatigue strength τlim used in the fatigue limit surface pressure calculation process, and the value determined in the shear fatigue strength determination process Any two of the three corrections including the size effect correction, which is a correction in which the value of 80% of the shear fatigue strength in the ultra-long life region is set to the value of the shear fatigue strength τlim used in the fatigue limit surface pressure calculation process. A method for estimating the fatigue limit surface pressure of rolling bearing materials for railway vehicles, in which the breaking fatigue strength τlim obtained by combining the above corrections is regarded as an absolute value.   The fatigue limit surface of a rolling bearing material for a railway vehicle according to any one of claims 1 to 8, wherein the metal material is any one of SNCM420, SUJ2, SUJ3, and SCr420, and the test piece is a heat-treated product. Pressure estimation method.   A metal material for which a fatigue limit surface pressure estimated by the fatigue limit surface pressure estimation method according to any one of claims 1 to 9 is equal to or greater than a predetermined fatigue limit surface pressure is used for a railway vehicle. A method for selecting a rolling bearing material for a rolling stock used as a material for a bearing ring or rolling element of a rolling bearing. An apparatus for estimating a fatigue limit surface pressure Pmax lim of a metal material that is in rolling contact with a rolling ring or rolling element of a rolling bearing for a railway vehicle,
An input means for storing the relationship between the shear stress amplitude of the metal material and the number of loads obtained by a complete double-swing ultrasonic torsional fatigue test in a predetermined storage area;
A shear fatigue strength determining means for determining the shear fatigue strength τlim in the ultra-long life region from the relationship between the stored shear stress amplitude and the number of loads, according to a predetermined criterion;
Maximum alternating force acting on the surface of the object of the metal material determined from the shape and size of the surface of the object made of the metal material and the object that is in rolling contact with the object, the size and the load that gives the contact surface pressure The maximum contact surface pressure Pmax when the load is applied with the shear stress amplitude τ ₀ equal to the shear fatigue strength τlim is determined by a predetermined calculation formula, and the maximum contact surface pressure Pmax is determined from the fatigue limit surface pressure Pmax lim. An apparatus for estimating a fatigue limit surface pressure of a rolling bearing material for a railway vehicle, comprising a fatigue limit surface pressure calculation means for an estimated value.   An ultrasonic torsional fatigue testing machine body that performs a complete double-sided ultrasonic torsional fatigue test on a test piece of a metal material used as a bearing ring or rolling element of a rolling bearing for railway vehicles, and this ultrasonic torsional fatigue testing machine body A system for estimating a fatigue limit surface pressure of a rolling bearing material for a railway vehicle, comprising: a testing machine control device that controls the test device according to an input test condition; and a fatigue limit surface pressure estimation device according to claim 11.

Images