JP5545043B2 - Residual stress measurement method - Google Patents

Description

  The present invention relates to a method for measuring a residual stress of an object to be measured, and more particularly to a method for measuring a residual stress on the surface of the object to be measured.

  For example, a turbine blade of a jet engine is manufactured from a polycrystalline casting material having coarse crystal grains, a unidirectionally solidified material, or a single crystal material. In this turbine blade, if a tensile residual stress is generated on the surface of the dovetail fitted to the disk, there is a possibility that a problem in strength occurs. Therefore, in manufacturing the turbine blade, it is necessary to appropriately manage the machining process and, depending on the case, to apply shot peening to the surface of the dovetail so as to reliably apply a compressive stress. As described above, it is very important in managing the strength of the product to confirm whether or not the residual stress due to the processing is appropriately given near the surface of the product.

  As a method for measuring the residual stress on the material surface, a measurement method using a stress release method has been proposed. In this measurement method, with the strain gauge attached to the surface of the object to be measured for residual stress, the surface layer of the gauge application surface of the object to be measured is thinly peeled off, and the change in output of the strain gauge at this time indicates that the object to be measured is The surface residual stress is measured (for example, Patent Document 1).

JP 10-48069 A

  However, it is extremely difficult to measure the residual stress generated by machining near the surface of the dovetail of the turbine blade by the measurement method using the stress release method described above. The reason is that the main distribution region including the peak of the residual stress given to the dovetail surface by machining, shot peening or the like is limited to a depth of about 200 to 300 μm from the surface. That is, it is very difficult to peel off such an extremely thin surface layer portion from the object to be measured by the existing general-purpose processing technology.

  Even if it can be peeled off, if the peel thickness is thinner than 200 to 300 μm, it is extremely difficult to handle. In addition, if the peel thickness is such a thin thickness, the rigidity of the attached strain gauge itself may affect the output of the strain gauge, so the output that correctly reflects the residual stress of the object to be measured is applied to the strain gauge. It is impossible to ask.

As a method for measuring the stress on the surface of the material that can solve the problem of the stress release method, an X-ray diffraction method (sin ² ψ method) which is a non-contact measurement method can be given. In the X-ray diffraction method, the diffraction angle when the object to be measured is irradiated with X-rays of a specific wavelength is determined according to the stress of the object to be measured, sin ² ψ (ψ (psi)); This is a measurement method using the fact that it changes linearly with respect to the angle with the lattice plane normal.

When measuring the residual stress of an object to be measured by the X-ray diffraction method, an X-ray having a specific wavelength is irradiated to the object to be measured while changing the incident angle, and the diffraction angle (2θ) at each crystal grain having a different ψ (psi). Measure. Then, a plurality of combinations of the measured diffraction angle (2θ) and sin ² ψ are linearly approximated by the least square method. The residual stress of the object to be measured can be obtained by obtaining the slope of this linear approximation formula.

However, the X-ray diffraction method is suitable for the measurement of residual stress in polycrystalline materials with fine crystal grains, but it is suitable for the measurement of residual stress in turbine blades made of materials with coarse crystal grains (large crystal grain size). Not. In a material with coarse crystal grains, a large number of crystal grains are not simultaneously irradiated with X-rays. Therefore, even if the incident angle of X-rays is changed, a plurality of combinations of diffraction angles (2θ) and sin ² ψ cannot be obtained. It is.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a residual stress in a shallow portion of a surface layer of an article made of a material having coarse crystal grains such as a turbine blade of a jet engine. It is an object of the present invention to provide a residual stress measurement method capable of accurately measuring the stress including its distribution.

In order to achieve the above object, the residual stress measuring method of the present invention described in claim 1 is:
A method for measuring a residual stress applied to the surface of an object to be measured,
A test piece cut out from the surface side of the object to be measured including the portion to which the residual stress of the object to be measured is applied is polished in layers from the surface to be measured corresponding to the surface of the object to be measured. While measuring the strain on the back surface of the test piece facing the surface to be measured,
Analyzing the residual stress applied to the surface of the object to be measured (for example, by a finite element method) from the relationship between the layered polishing of the test piece and the strain measured on the back surface;
It is characterized by including.

  According to the residual stress measuring method of the present invention as set forth in claim 1, when the surface to be measured of the test piece cut out from the surface side of the object to be measured is polished stepwise, the back surface opposite to the surface to be measured is formed. Causes a strain corresponding to the residual stress applied to the portion to be measured of the test piece after being layered.

  Therefore, based on the strain generated on the back surface of the test piece measured while polishing the surface to be measured of the test piece in layers, the residual stress applied to the surface to be measured of the test piece and thus to the surface side of the object to be measured The depth distribution in the depth direction can be obtained by inverse analysis using an analysis model such as FEM (finite element method).

  Thus, the residual stress measurement method of the present invention is a measurement method for obtaining the distribution of residual stress in the depth direction by polishing the surface to be measured of the test piece to which the residual stress is applied. Residual stress cannot be measured by the diffraction method or the conventional stress relief method, even if it is the residual stress of a very shallow surface layer of an article made of a material with coarse grains, and the depth can be measured accurately. The residual stress distribution in the direction can also be known.

Moreover, the residual stress measuring method of the present invention described in claim 1 is:
The analyzing step comprises:
By adjusting the temperature or linear expansion coefficient (linear expansion coefficient ) on the measured surface side in the analysis model of the test piece, and the thickness between the measured surface side and the back surface in the analysis model of the test piece , the change of the strain output before and after removing the stepwise layered the specimen in the analysis model from the measurement surface, and the surface to be measured of the test piece was measured by the measuring step when the polishing layer the In accordance with the value of the change in strain output on the back surface of the test piece, a specific step of specifying the temperature or linear expansion coefficient of each layer to be removed stepwise from the measured surface side of the analytical model;
An estimation step of estimating a residual stress distribution in the depth direction of the test piece by an overall analysis model given to each layer corresponding to the temperature or linear expansion coefficient specified by the specifying step;
It is characterized by including.

According to the residual stress measuring method of the present invention described in claim 1 , by removing the analysis model in layers from the surface to be measured, a test piece in a state where the residual stress is applied to the surface to be measured is measured. A state of polishing in layers from the surface side is simulated.

  Then, the strain value generated on the back surface of the analytical model when the measured surface side of the analytical model is removed in layers is generated on the back surface of the test piece when the layer at the same position on the measured surface side of the test piece is polished. When the temperature or linear expansion coefficient on the measured surface side of the analytical model is set so as to match the strain value, the expansion and contraction that occurs on the measured surface side of the analytical model in accordance with the temperature or linear expansion coefficient is the test piece. This corresponds to the residual stress applied to the surface to be measured, in other words, the residual stress applied to the surface of the object to be measured.

  Therefore, when the measured surface side of the test piece is polished in layers, the strain value that matches the strain value generated on the back side of the test piece is analyzed when the measured surface side of the analysis model is removed in layers. By identifying the temperature or linear expansion coefficient on the measured surface side of the analytical model when it occurs on the back side of the model, the layer to be polished on the measured surface side of the test piece, and hence the surface side of the measured object The residual stress applied to the polished layer can be estimated by analysis. For this reason, even if the article is made of a material having coarse crystal grains, the residual stress of the surface layer can be accurately measured including its distribution.

In addition, when the parameter specified in the specific step in the residual stress measurement method of the present invention described in claim 1 is a linear expansion coefficient, the test piece and the analysis are taken into consideration by considering the material anisotropy of the test piece and the analysis model. The amount of elongation of the model can be changed depending on the direction. Therefore, if the strain change in the biaxial direction on the back surface of the test piece is obtained by a biaxial strain gauge, the residual stress distribution in the in-plane gauge biaxial direction as well as the depth direction can be estimated by analysis.

  According to the present invention, even in an article made of a material having coarse crystal grains, the residual stress in the shallow portion of the surface layer can be accurately measured including its distribution.

It is explanatory drawing which shows the cutout procedure of the test piece in the residual stress measuring method which concerns on one Embodiment of this invention. It is explanatory drawing which shows the procedure of the measurement preparation of the residual stress using the test piece of FIG. It is explanatory drawing which shows the measurement step of the distortion using the test piece of FIG. It is a graph which shows the residual stress given to the test piece of FIG. 1 by the relationship with the position from the back surface in the thickness direction of a test piece. It is a graph which shows the relationship between the distortion | strain of the test piece measured in the measurement step of FIG. 3, and the grinding | polishing thickness of a test piece.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiments, a case where residual stress applied to the dovetail surface of a turbine blade of a jet engine is measured will be described as an example.

  The dovetail 3 (corresponding to the object to be measured in the claims) of the turbine blade 1 shown in FIG. 1 is manufactured from a material such as a polycrystalline casting material with coarse crystal grains, a unidirectional solidified material, or a single crystal material. . In the turbine blade 1, in particular, if a tensile residual stress is generated on the surface 3 a of the dovetail 3 fitted to a disk (not shown), there may be a problem in strength. Therefore, when manufacturing the turbine blade 1, the machining process is appropriately managed, and in some cases, shot peening is applied to the surface 3 a of the dovetail 3 to reliably apply a compressive stress.

  When measuring the residual stress applied to the surface 3 a of the dovetail 3, first, the test piece 5 is cut out from the surface 3 a of the dovetail 3 of the turbine blade 1. At this time, the test piece 5 is cut into a thin plate shape by a method in which new residual stress such as wire cutting (a kind of electric discharge machining) is not applied as much as possible so as to include the surface layer portion of the dovetail 3 to which the residual stress is applied.

  Next, the strain gauge 7 is affixed to the back surface 5b facing the surface to be measured 5a which is a part of the surface 3a of the dovetail 3 in the cut out test piece 5. And it transfers to a measurement step.

  In the measurement step, as shown in FIG. 3, the test piece 5 is polished stepwise from the surface to be measured 5a by a method in which new residual stress such as electrolytic polishing is not applied as much as possible. In parallel with this polishing, the strain of the back surface 5b is measured using the output of the strain gauge 7.

As illustrated in the graph of FIG. 4, the test piece 5 (or the dovetail 3) has a surface layer portion (a portion near the Y coordinate = 2.5 in the drawing) in the vicinity of the measured surface 5a (the surface 3a of the dovetail 3). ), There is a residual stress given by machining or shot peening. Therefore, when the test piece 5 is sequentially polished in layers from the measured surface 5a side as shown in FIG. 3, as shown in the graph of FIG. 5, the output of the strain gauge 7 representing the strain of the back surface 5b of the test piece 5 is ε ₀ , ε ₁ , ε ₂ ,..., ε _n−2 , ε _n−1 , ε _n sequentially change.

Such output changes Δε ₁ (ε ₁ −ε ₀ ), Δε ₂ (ε ₂ −ε ₁ ),..., Δε _n−1 (ε _n−2 −ε _n−1 ), Δε _n ( ε _n-1 −ε _n ) is applied to the back surface 5b of the test piece 5 as the residual stress applied to the layer portion on the measured surface 5a side of the polished test piece 5 is released by polishing. This occurs when the amount of distortion that occurs changes. Therefore, the outputs ε ₀ , ε ₁ , ε ₂ ,..., Ε _n−2 , ε _n−1 , ε _n of the strain gauge 7 obtained by sequentially polishing the test piece 5 from the measured surface 5 a side in layers are This corresponds to the residual stress applied to the portion of the test piece 5 on the measured surface 5a side after polishing.

  Therefore, from the strain value of the back surface 5b of the test piece 5 measured by the strain gauge 7 while polishing the measured surface 5a of the test piece 5 in layers in the procedure shown in FIG. The distribution of the given residual stress can be obtained by inverse analysis.

  The residual stress analysis step will be described below. In this embodiment, an analysis model obtained by modeling the test piece 5 by the finite element method (FEM) is used for this analysis.

The analysis model described above is made of the same material as that of the test piece 5. Here, for ease of explanation, the surface 5a to be measured of the test piece 5 is layered, the first layer (thickness t ₁ ), the second layer (thickness t ₂ ), the third layer (thickness t ₃ ) It is assumed that the output of the strain gauge 7 is measured by polishing in three stages. The analysis procedure (step) will be described below.

  A specific step is performed as the first half of the analysis step. Here, for simplification of description, temperature is used as a parameter, but a linear expansion coefficient can also be used as a parameter instead of temperature.

In this specific step, first, the thickness t ₃ portion on the measurement surface side of the analysis model in which the thickness between the measurement surface side and the back surface is t ₃ + t ₀ is uniformly adjusted to the temperature T3. The third layer of the test piece 5 is placed at a position corresponding to the position where the strain gauge 7 is pasted on the back surface 5b of the test piece 5 on the back surface facing the measured surface side of the analysis model. the temperature T3 which appears strain changes [Delta] [epsilon] ₃ before and after the removal of, identified by repeating the FEM analysis (step 1). Specifically, the strain corresponding to the gauge calculated by the analysis model of t ₃ + t ₀ is set to coincide with Δε ₃ . At this time, another portion of the thickness t ₀ of the analysis model excluding the portions of the thickness t ₃ when adjusting the temperature T3 is assumed for example, room temperature. Thereby, the output change of the strain gauge 7 attached to the back surface 5b before and after the third layer of the measured surface 5a of the test piece 5 is polished is reproduced by the analysis model.

Next, the thickness t ₂ portion of the analytical model in which the thickness between the measured surface and the back surface is t ₂ + t ₃ + t ₀ is uniformly adjusted to the temperature T2, so that the gauge corresponding portion of the analytical model is obtained. , the temperature T2 of strain change [Delta] [epsilon] ₂ before and after removal of the second layer appears in the test piece 5 is specified by repeating the FEM analysis (Step 2). Specifically, the strain corresponding to the gauge calculated by the analysis model of t ₂ + t ₃ + t ₀ and the analysis model (the thickness between the measured surface side and the back surface is t ₍ Analysis model of ₃ + t ₀ ) is set so that the difference from the strain at the same position matches Δε ₂ . At this time, the temperature of the portion of the t ₃ at any model temperature of T3, t ₂ parts identified in step 1 and the temperature T2 identified in Step 2. Thereby, the output change of the strain gauge 7 attached to the back surface 5b before and after the second layer of the measured surface 5a of the test piece 5 is polished is reproduced by the analysis model.

Subsequently, the strain change Δε ₁ before and after the first layer of the test piece 5 is removed at the gauge corresponding portion of the analysis model by uniformly adjusting the surface to be measured, the back surface, and the thickness t ₁ to the temperature T1. Is identified by repeating the FEM analysis (procedure 3). Specifically, the strain corresponding to the gauge calculated by the analysis model of t ₁ + t ₂ + t ₃ + t ₀ and the analysis model (thickness between the measured surface side and the back surface) specified in the procedure 2 above. The difference between the strain at the same location in the analysis model in which t ₂ + t ₃ + t ₀ is equal to Δε ₁ . Thereby, the output change of the strain gauge 7 attached to the back surface 5b before and after the third layer of the measured surface 5a of the test piece 5 is polished is reproduced by the analysis model.

As described above, after the temperature setting for each layer is completed, the temperatures T3, T2, and T2 specified in the steps 1 to ₃ are measured for the portions of the thickness t ₁ , t ₂ , t ₃ on the measured surface side of the analysis model. An analysis is performed using the entire FEM model given T1 (procedure 4). The stress distribution obtained as an output of this analysis (FIG. 4 is an example) is the residual stress distribution generated on the measured surface 5a side of the test piece 5.

  The analysis of the procedure 4 can obtain the stress state in the test piece state, but in order to know the stress distribution in the actual machine state (the dovetail 3 of the turbine blade 1), the test piece model is incorporated into the actual product model. The stress distribution may be obtained by reconstructing the analysis model (constructing an analysis model corresponding to the dovetail 3 of the turbine blade 1) and analyzing the analysis model.

  When the linear expansion coefficient is used as a parameter instead of the temperature, three linear expansion coefficients satisfying the conditions described in steps 1 to 3 can be obtained instead of obtaining the temperatures T3, T2 and T1 in steps 1 to 3, respectively. That's fine. And in the said procedure 4, what is necessary is just to implement an analysis with the whole FEM model which gave three linear expansion coefficients calculated | required by the said procedures 1-3. When the parameter is a linear expansion coefficient, the elongation amount of the analysis model can be changed depending on the direction by considering the material anisotropy of the test piece 5 and the analysis model. Therefore, if the strain change in the biaxial direction on the back surface of the test piece 5 is acquired by a biaxial strain gauge, the residual stress distribution in the in-plane gauge biaxial direction as well as the depth direction can be estimated.

  From the FEM analysis described above, the residual stress applied to the surface 3a side of the dovetail 3 from which the test piece 5 is cut out is estimated from the output of the strain gauge 7 when the measured surface 5a of the test piece 5 is polished in layers. can do.

  As described above, the residual stress measurement method according to the present embodiment polishes the measured surface 5a side of the test piece 5 cut out from the surface 3a of the dovetail 3 to which the residual stress is applied, and thereby the depth direction of the test piece 5 is polished. This is a measurement method for obtaining the distribution of residual stress in the direction from the measured surface 5a to the back surface 5b of the test piece 5. Therefore, the residual stress cannot be measured by the X-ray method or the stress release method, and even the residual stress of the surface layer of an article made of a material having coarse crystal grains can be accurately measured, and the depth direction It is also possible to know the residual stress distribution. Moreover, although it depends on the accuracy of strain measurement, the estimation accuracy of the distribution of the residual stress to be obtained can be arbitrarily increased by reducing the polishing amount.

  In the present embodiment, the case where a model by FEM analysis is used for modeling the residual stress of the test piece 5 has been described, but the analysis may be performed by a method other than FEM.

  That is, the present invention is to measure a test piece cut out from the surface side of an object to be measured whose surface is stressed while stepwise polishing in layers from the surface to be measured corresponding to the surface of the object to be measured. The gist is to analyze the residual stress applied to the surface of the object to be measured from the relationship between the strain measured on the back surface of the test piece facing the surface and the position and polishing thickness of the polished layer on the test piece. Is.

  In this embodiment, machining, shot peening, etc. are performed on the surface 3a of the dovetail 3 of the turbine blade 1 manufactured from a cast material such as a polycrystalline material having coarse crystal grains, a unidirectionally solidified material, or a single crystal material. The case where the residual stress given by is measured is described as an example. However, the present invention is widely applicable when measuring the residual stress applied to the surface of the object to be measured regardless of the material or crystal structure.

1 Turbine blade 3 Dovetail 3a Surface 5 Test piece 5a Surface to be measured 5b Back surface 7 Strain gauge

Claims (1)

A method for measuring a residual stress applied to the surface of an object to be measured,
A test piece cut out from the surface side of the object to be measured including the portion to which the residual stress of the object to be measured is applied is polished in layers from the surface to be measured corresponding to the surface of the object to be measured. While measuring the strain on the back surface of the test piece facing the surface to be measured,
From the relationship between the layered polishing of the test piece and the strain measured on the back surface, an analysis step for analyzing the residual stress applied to the surface of the object to be measured ;
Including
The analysis step includes
By adjusting the temperature or linear expansion coefficient on the measurement surface side in the analysis model of the test piece and the thickness between the measurement surface side and the back surface in the analysis model of the test piece, from the measurement surface side The change in strain output before and after removing the test piece in layers in the analysis model stepwise is measured in the measurement step when the measured surface of the test piece is polished in layers. A specific step of adjusting the value of change in strain output and specifying the temperature or linear expansion coefficient of each layer to be removed stepwise from the measured surface side of the analytical model;
An estimation step of estimating a residual stress distribution in the depth direction of the test piece by an overall analysis model given to each layer corresponding to the temperature or linear expansion coefficient specified in the specifying step,
Residual Stress Measurement how, characterized in that.

Images