KR20160134372A - Bearing life prediction method on run-out - Google Patents

Abstract

SUMMARY OF THE INVENTION It is a technical object of the present invention to provide a method of predicting a bearing shudder reference life which can more accurately predict the life of a bearing. Particularly, since the precision of the bearing used in precision machining of machine tools means precision of forming the product, there is a need for a method of predicting the bearing shake reference life for the user. According to this necessity, the bearing shaking reference life prediction method of the present invention includes: selecting a bearing wear model which is a cause of shaking which is a lifetime reference of a bearing; An acceleration life test step of measuring a degree of shaking of the bearing by applying a load and a rotation speed selected as acceleration conditions to the bearing; Analyzing the shaking by applying the load and the rotational speed to the selected bearing wear model as a boundary condition of a numerical analysis; Making a part life curve based on the number of revolutions data that reaches the life criterion at the load; And estimating the life of the bearing in the load by calculating the number-of-rotations data as a regression analysis curve.

Description

Bearing vibration prediction method on run-out

The present invention relates to a method for predicting the life of a bearing.

Generally, the bearing is provided between the stationary member and the rotary member and serves to induce smooth rotation of the rotary member with respect to the stationary member. Such bearings include, as is well known, an outer race provided on the fixed member, an inner race provided on the rotary member, and a plurality of rollers provided between the outer race and the inner race .

These bearings suffer from defects in rollers and races, such as wear, spalling, and fitting (pitting). Since defects have a direct impact on the service life of the bearings, it is very important to predict the defects in advance.

The service life of the currently specified bearings is calculated according to the nominal life equation of the ISO 281 standard. This specification defines the life of a bearing by determining that the point of occurrence of spalling on the roller or race of the bearing is a failure. The nominal life is determined by geometric factors such as the contact angle between the roller and the race, and the ratio.

However, the existing bearing life prediction method uses the point of occurrence of spalling as a judgment of the life of the bearing. Even if spalling occurs, if the influence of the bearing shake is not influenced, the use of the bearing after the rated life of the ISO 281 standard There is a problem of judging a failure even if it is possible.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method of predicting the bearing shudder reference life which can more accurately predict the life of a bearing.

Particularly, since the precision of the bearing used in precision machining of machine tools means precision of forming the product, there is a need for a method of predicting the bearing shake reference life for the user.

According to an aspect of the present invention, there is provided a method for estimating a bearing shudder reference life according to an embodiment of the present invention includes: selecting a bearing wear model that causes shudder, which is a life span of a bearing; An acceleration life test step of measuring a degree of shaking of the bearing by applying a load and a rotation speed selected as acceleration conditions to the bearing; Analyzing the shaking by applying the load and the rotational speed to the selected bearing wear model as a boundary condition of a numerical analysis; Making a part life curve based on the number of revolutions data that reaches the life criterion at the load; And estimating the life of the bearing in the load by calculating the number-of-rotations data as a regression analysis curve.

Wherein the step of selecting the bearing wear model includes a step of selecting a bearing wear model by using a progressive wear model considering that the bearing wears by contact and an non-viscous wear model considering that the bearing is damaged by repeated tensile stress step; And deriving a wear model constant for each of the progressive wear model and the non-lubrication wear model.

The progressive wear model may use an Archard wear model that shows a linear relationship between load, speed, and hardness at the contact of the bearing, and the Achad wear model is defined by the following equation,

는 마모 높이,

는 수직압력,

는 미끄럼 속도,

는 경도,

는 마모 계수이고,

,

,

는 마모 상수로 마모 정도에 영향을 미치는 인자들의 자유도고,

와

는 재료 표면에 대한 표면 거칠기 상수이다.here,

Wear height,

Vertical pressure,

The sliding speed,

The hardness,

Is the wear factor,

,

,

Is the wear modulus, the degree of freedom of the factors affecting the degree of wear,

Wow

Is the surface roughness constant for the material surface.

, 상수

,

,

와 표면 거칠기 상수

,

는 핀 온 디스크 시험으로 구할 수 있다.In the step of deriving the wear model constant, as the constant of the progressive wear model,

, a constant

,

,

And surface roughness constant

,

Can be obtained by pin-on-disk testing.

The non-lacquer wear model may use a normalized cockcroft-latham fix model to simulate non-lacquer wear, and the normalized ck-croft-lath fix model may be calculated by the following equation Defined,

는 임계값,

는 최대 주응력,

는 유효응력이고,

는 손상 유효변형률,

는 전체 변형률이고,

는 손상 상수이다.here,

Lt; / RTI >

Is the maximum principal stress,

Is an effective stress,

The damage effective strain,

Is the total strain,

Is a damage constant.

와 상수

는 역공학 방법을 사용하여 구할 수 있다.In the step of deriving the wear model constant, as the constant of the non-lattice wear model,

And constant

Can be obtained using a reverse engineering method.

In the acceleration life test step for measuring the degree of shaking of the bearing, the degree of shaking in the axial direction and the radial direction of the bearing can be measured using a dial gauge.

The step of applying the load and the rotation speed as boundary conditions of numerical analysis to analyze the shaking may be performed by applying the load and the rotation speed under the same conditions as the acceleration life test step to the progressive wear model and the non- Interpreting the numerical value for < / RTI > And updating the shape to the progressive wear model and the non-lacquer wear model based on the numerical value for the wear.

, 상기 미끄럼 속도

, 상기 경도

값이 계산되고, 상기 핀 온 디스크를 통해 구한 마모 모델 상수를 적용하면 상기 마모 높이

가 계산될 수 있다.In the step of analyzing the numerical value, the numerical analysis of the progressive wear model calculates the degree of wear at each node as a height reference, and in the numerical analysis, the vertical pressure

, The sliding speed

, The hardness

Value is calculated, and applying the wear model constant obtained through the pin-on disk, the wear height

Can be calculated.

The step of updating the wear shape may update the wear shape computed in a form that deducts the position of the node by the height of the wear in the outward direction of the vector to the progressive wear model.

과 상기 유효 응력

, 상기 유효 변형률

, 상기 전체 변형률

값이 계산되고, 그 비율에 따라 손상 누적치가 계산될 수 있다.In the step of analyzing the numerical value, the numerical analysis of the non-lattice wear model indicates spalling caused by accumulation of damage, so that abrasion occurs on a per element basis, and the maximum principal stress

And the effective stress

, The effective strain

, The total strain

A value is calculated, and a damage accumulation value can be calculated according to the ratio.

In the step of updating the wear shape, when the accumulated value of the damage reaches a threshold value set in the model, the non-losing wear model can update the wear shape by removing the element.

As described above, the bearing shudder reference life prediction method according to an embodiment of the present invention can have the following effects.

According to an embodiment of the present invention, the life of the bearing can be predicted based on the shaking of the bearing, so that the life of the bearing can be predicted more accurately than the technique of determining the failure at the time of the existing spalling. In particular, it is possible to predict quantitatively bearing life, predict bearing life in a short time using accelerated life testing and analytical models, and predict wear life of bearings of different sizes and types using wear life models .

FIG. 1 is a flowchart schematically illustrating a method for predicting a bearing shudder reference life according to an embodiment of the present invention.
2 is a graph of the test response at (a) at 1,020 kgf, (b) at 893 kgf, (c) at 765 kgf and (d) at 638 kgf in acceleration conditions.
FIG. 3 is a graph comparing the response of the test and analysis at the acceleration conditions (a) at 1,020 kgf, (b) at 893 kgf, (c) at 765 kgf and (d) at 638 kgf.
4 is a graph showing an example of a bearing part life curve according to the method of predicting the bearing shudder reference life in FIG.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art to which the present invention pertains. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

FIG. 1 is a flowchart schematically illustrating a method for predicting a bearing stagnation reference life according to an embodiment of the present invention. FIG. 2 is a graph showing the relationship between a bearing stiffness and a bearing stiffness at an acceleration condition of 1,020 kgf, (c) is the test response graph at 765 kgf and (d) is the test response graph at 638 kgf, FIG. 3 is a graph showing the test response graph at (a) at 1,020 kgf, (b) at 893 kgf, ) Is a graph comparing the response of the test and analysis at 638 kgf, and Fig. 4 is a graph showing an example of the bearing part life curve according to the bearing shudder reference life prediction method of Fig.

The method of predicting bearing shudder reference life according to an embodiment of the present invention proceeds as shown in FIG. The first step is a bearing wear model selection step, the second step is a step of deriving a wear model constant, the third step is a step of testing an accelerated life, the fourth step is a step of interpreting numerical values, Is the step of making the part life curve, and the sixth step is the step of predicting the bearing life.

Hereinafter, each step will be described in more detail.

1. Selection of bearing wear model

1.1 Incremental wear model

The progressive wear model uses the Archard wear model which shows the linear relationship between the load, speed, and hardness at the contact of the material, and the wear of the material can be predicted through this model. This specification uses the Archard wear modification model which increases the degree of freedom of each variable according to the characteristics of the material in the existing Archard model.

-----------------------------------(1-1)

----------------------------------- (1-1)

는 마모 높이,

는 수직압력,

는 미끄럼 속도,

는 경도,

는 마모 계수 이다.

,

,

는 마모 상수로 마모 정도에 영향을 미치는 인자들의 자유도다. 이 표준은 추가적으로 표면 상태를 고려하여, 마모 모델에 적용하였다.

와

는 재료 표면에 대한 표면 거칠기 상수이다. 표면에 열처리 및 코팅이 되어있는 부품은 초반에 발생하는 마모량이 적다. 마모될수록 표면은 거칠어지며, 마모 속도는 더 증가할 것이다.
here

Wear height,

Vertical pressure,

The sliding speed,

The hardness,

Is the wear factor.

,

,

Is the wear factor and the freedom of factors affecting wear. This standard was further applied to the wear model, taking into account the surface condition.

Wow

Is the surface roughness constant for the material surface. Parts that are heat treated and coated on the surface have a small amount of wear occurring in the early stage. The more wear, the rougher the surface, the faster the wear rate will be.

1.2 Non-Viscous Wear Model

Non-coherent wear models are caused by cumulative damage due to repeated tensile stresses at the bottom of the contact surface, such as spalling. To simulate non-point worn wear, a normalized cockcroft-latham correction model is used.

-------------------------------------------(2-1)

------------------------------------------- (2-1)

는 임계값을 나타낸다.

는 최대 주응력,

는 유효응력이다. 그리고

는 손상 유효변형률,

는 전체 변형률이다. 이 표준에서는 비선형 마모 경향을 적용하기 위해서, 손상 상수

를 추가하였다. 손상 누적이 임계

값에 도달하면 부품은 파단에 이른다.

Represents a threshold value.

Is the maximum principal stress,

Is the effective stress. And

The damage effective strain,

Is the total strain. In order to apply a nonlinear wear trend in this standard,

Respectively. Damage accumulation is critical

When the value is reached, the part is broken.

2. Derivation of wear model constants

2.1 Incremental wear model

, 상수

,

,

와 표면 거칠기 상수

,

는 핀 온 디스크 시험으로 구한다. 핀과 디스크는 모두 베어링강의 시편으로 제작한다. 점진 마모 모델은 다음과 같이 정립하였다.In a progressive wear model,

, a constant

,

,

And surface roughness constant

,

Is obtained by the pin-on-disk test. Both pins and discs are made of bearing steel specimens. The progressive wear model was established as follows.

----------------------------(1-2)

---------------------------- (1-2)

2.2 Non-viscous wear model

와 상수

는 역공학 방법을 사용하여 구한다. 실험과 해석결과의 차이를 최소화하는 값으로 최적화하여, 다음과 같은 비점진 마모 모델이 정립하였다.In the case of the non-coherent wear model constant,

And constant

Is obtained using a reverse engineering method. The following non-lattice wear model was established by optimizing the values to minimize the difference between the experiment and the analysis results.

-----------------------------------------(2-2)

----------------------------------------- (2-2)

3. Accelerated Life Test

3.1 Selecting Test Response

a) Bearing life is determined by the falling of a part of the material such as spalling. In the case of bearings used in machine tools, spalling can continue to be used if there is little shaking. In particular, large bearings are costly and cumbersome to replace, so the bearing life is judged based on shaking rather than spalling.

b) Bearing life is judged to have reached the end of its useful life when the shaking of the used bearing exceeds the allowable shaking degree.

3.2 Selection of acceleration condition

a) Factors affecting the service life of the bearing usage mechanism include excessive load, rotational speed, lack of lubrication, and contamination. Among them, the load and the rotational speed are set as severe conditions, and the acceleration test is carried out to derive the test response result according to the acceleration condition.

b) Load conditions shall be selected based on the rated load. For example, the acceleration conditions are selected based on 200%, 175%, 150%, and 125% of the rated load.

( Example ) Accelerated life test condition of 50mm inner diameter cross roller bearing

Table 1 - Selection of accelerated life test conditions

3.3 Measurement

Shaking indicates the degree of shaking when the product is in use. Using a dial gauge, measure the degree of shaking in the axial and radial directions of the rotating bearing.

3.4 Test Response Graph

The shaking in the axial direction and the radial direction according to the number of revolutions is derived as a graph. When the test is performed under four acceleration conditions, the results shown in FIG. 2 can be obtained.

For reference, FIG. 2 illustrates a test response graph of an inner diameter 50 mm cross roller bearing.

4. Numerical Analysis of Wear

4.1 Numerical Analysis of Wear

a) Proceed by applying the load and rotation speed of the same conditions to the numerical analysis model.

, 미끄럼 속도

, 경도

값이 계산된다. 핀 온 디스크를 통해 구한 마모 상수를 적용하면 마모 높이

가 계산된다.b) Gradual wear The numerical analysis computes the degree of wear at each node as a height criterion. In the numerical analysis, the vertical pressure of the progressive wear model

, Sliding speed

, Hardness

The value is calculated. Applying the wear constant obtained through the pin-on-disk, the wear height

Is calculated.

과 유효 응력

, 유효 변형률

, 전체 변형률

값이 계산되고, 그 비율에 따라 손상 누적치가 계산된다.
c) Non-Kinematic Wear The numerical analysis shows the spalling that occurs due to the accumulation of damage, so wear occurs on a per-element basis. Maximum principal stress

And effective stress

, Effective strain

, Total strain

The value is calculated, and the cumulative damage value is calculated according to the ratio.

4.2 Update wear geometry

a) Progressive wear progresses in the form of subtracting the position of the node by the height of the calculated abrasion in the outward direction of the vector.

b) Non-pointing wear updates the wear shape by removing the element when the cumulative value of the damage reaches the threshold set in the model.

4.3 Measurement

a) The shaking is a geometrical problem that comes from the roller, inner race, and outer race gap due to wear of the roller surface. We apply the shape of the updated model to the numerical analysis, so that the gap occurs and the actual phenomenon can be simulated.

b) Obtain data on the degree of shaking by measuring the axial and radial displacements of the bearings during the test as in the test.

For reference, FIG. 3 illustrates a test and analysis response graph of an inner diameter 50 mm cross roller bearing.

5. Life prediction

a) The component life curve is based on the number of revolutions data that reaches the life criterion at each test load. Calculating this data as a regression curve, we can see the life of the bearing shake criterion at a specific load.

b) In order to verify the procedure, additional tests and interpretations were carried out at 75% of rated load and rated load.

4, The part life curve of the cross roller bearing shows that the bearing life is 285.45 cycles [x10,000] at a load of 400 kgf.

6. Terms and Definitions

6.1 Bearing life

It is the number of revolutions until the life standard is exceeded when using bearings. In this standard, the total number of revolutions until the degree of shaking is exceeded.

6.2 Bearing run-out

When using bearings, clearance is generated by wear of inner race and outer race or roller. And the degree of shaking of the bearing in the axial direction and the radial direction due to the clearance.

6.3 Accelerated life test

This test is carried out under harsh conditions to shorten the test period. The test is conducted by setting the factors affecting the parts life to harsh conditions.

6.4 Numerical simulation

Describe the finite element analysis. Generally, a finite element model is composed of nodes and elements.

6.5 Progressive wear

It is the wear which gradually occurs by the friction generated by the contact on the material surface.

6.6 Non-progressive wear

It is the wear that the material locally falls off progressively due to accumulation of damage from the bottom of the material surface.

6.7 Wear geometry update

The amount of wear is applied to the numerical analysis model shape. This is to create a realistic gap and to obtain the shaking result by numerical analysis.

6.8 Node

It is a discrete point where a line or plane in a continuum of numerical analysis is interconnected.

6.9 Elements

It is a partition variable that constitutes a continuum of numerical analysis, and is interconnected with nodes.

6.10 Displacement

A displacement of a point within an element. The space is set according to the numerical analysis model coordinates, and the displacement direction is set.

6.11 Regression analysis

It is the causal relationship between independent variables and dependent variables.

As described above, the bearing shudder reference life prediction method according to an embodiment of the present invention can have the following effects.

According to an embodiment of the present invention, the life of the bearing can be predicted based on the shaking of the bearing, so that the life of the bearing can be predicted more accurately than the technique of determining the failure at the time of the existing spalling. In particular, it is possible to predict quantitatively bearing life, predict bearing life in a short time using accelerated life testing and analytical models, and predict wear life of bearings of different sizes and types using wear life models .

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, Of the right.

Claims (12)

Selecting a bearing wear model which causes vibration, which is a lifetime reference of the bearing;
An acceleration life test step of measuring a degree of shaking of the bearing by applying a load and a rotation speed selected as acceleration conditions to the bearing;
Analyzing the shaking by applying the load and the rotational speed to the selected bearing wear model as a boundary condition of a numerical analysis;
Making a part life curve based on the number of revolutions data that reaches the life criterion at the load; And
Calculating the number of revolutions data as a regression analysis curve to predict the life of the bearing in the load. The method of claim 1,
The step of selecting the bearing wear model includes:
The bearing wear model includes a progressive wear model considering that the bearing wears by contact, and an non-viscous wear model considering that the bearing is damaged by repeated tensile stress; And
And deriving a wear model constant for each of the progressive wear model and the non-losing wear model. 3. The method of claim 2,
In the progressive wear model,
Using an Archard wear model showing a linear relationship between load, speed, and hardness at the contact of the bearing,
The Achad wear model is defined by the following equation,

here,

Wear height,

Vertical pressure,

The sliding speed,

The hardness,

Is the wear factor,

,

,

Is the wear modulus, the degree of freedom of the factors affecting the degree of wear,

Wow

Is a surface roughness constant for a material surface. 4. The method of claim 3,
In deriving the wear model constant,
As the constant of the progressive wear model,

, a constant

,

,

And surface roughness constant

,

In the step of selecting the bearing wear model obtained by the pin-on-disk test, 3. The method of claim 2,
In the non-losing wear model,
To simulate non-dense wear, a normalized cockcroft-latham correction model was used,
The normalized kerkcroft-ramp correction model is defined by the following equation,

here,

Lt; / RTI >

Is the maximum principal stress,

Is an effective stress,

The damage effective strain,

Is the total strain,

A method for predicting the bearing shake reference life which is a damage constant. The method of claim 5,
In deriving the wear model constant,
As a constant of the non-lattice wear model,

And constant

A method for predicting bearing shudder fatigue life using a reverse engineering method. The method of claim 1,
In the acceleration life test step of measuring the degree of shaking of the bearing,
Wherein the degree of shaking in the axial direction and the radial direction of the bearing is measured using a dial gauge. 3. The method of claim 2,
The step of applying the load and the rotational speed as boundary condition of numerical analysis to analyze the shaking,
Applying a load and a rotation speed under the same conditions as the accelerated life test step to the progressive wear model and the non-lubrication wear model to analyze a value for wear; And
And updating the wear profile on the progressive wear model and the non-losing wear model based on the value for the wear. 5. The method of claim 4,
The step of applying the load and the rotational speed as boundary condition of numerical analysis to analyze the shaking,
Applying a load and a rotation speed under the same conditions as the accelerated life test step to the progressive wear model and the non-lubrication wear model to analyze a value for wear; And
And updating the wear profile on the progressive wear model and the non-losing wear model based on the value for the wear,
In interpreting the numerical value,
In the numerical analysis of the progressive wear model, the degree of wear at each node is calculated on the basis of height,
In the numerical analysis, the vertical pressure of the progressive wear model

, The sliding speed

, The hardness

Value is calculated, and applying the wear model constant obtained through the pin-on disk, the wear height

A method for predicting the bearing shake reference life. The method of claim 9,
Wherein updating the wear profile comprises:
And updating the wear shape calculated in the form of subtracting the position of the node by the height of the wear in the outer direction of the vector to the progressive wear model. The method of claim 6,
The step of applying the load and the rotational speed as boundary condition of numerical analysis to analyze the shaking,
Applying a load and a rotation speed under the same conditions as the accelerated life test step to the progressive wear model and the non-lubrication wear model to analyze a value for wear; And
Updating the marble shape in the progressive wear model and the non-lacquer wear model based on the value for the wear,
In interpreting the numerical value,
Numerical analysis of the non-lattice wear model shows spoiling occurring due to accumulation of damage, so that abrasion occurs on a per element basis,
The maximum principal stress

And the effective stress

, The effective strain

, The total strain

And calculating a damage accumulation value based on the ratio. 12. The method of claim 11,
In the step of updating the wear profile,
And the non-losing wear model is updated by removing the element when the cumulative value of the damage reaches a threshold value set in the model.

Images