Method for predicting fracture mode of welded member

Abstract

To correctly predict a fracture form when two or more overlapped steel plates are welded by resistance welding and a load is applied to a welded member having a crack in the steel plate.SOLUTION: Finite element analysis is performed using an analysis model of a welded member in which shapes of a steel sheet, a nugget, and a corona bond, and a position and a shape of one or more cracks generated in the steel sheet are reproduced, it is determined which of limit loads corresponding to four types of fracture modes is the smallest, and it is predicted that a fracture mode corresponding to the smallest limit load occurs first.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a method for predicting the fracture mode of a welded member. [Background technology]

[0002] In recent years, the automotive industry has been moving toward thinner and stronger steel sheets in order to improve fuel efficiency by reducing vehicle weight and ensure passenger safety in the event of a collision. Automobile bodies are typically assembled by welding structural components made of thin steel sheets. In order to increase the strength of a vehicle body, it is necessary to increase the strength of the steel sheets themselves, and also to ensure the strength of the welded joints of the steel sheets to ensure safety.

[0003] The strength of a welded component having a weld is affected by the chemical composition of the steel sheet, the strength of the steel sheet, the structure of the weld, the type of load applied to the weld, and the fracture mode of the weld. It is generally known that when the fracture mode of a weld is brittle fracture, the tensile strength is significantly reduced compared to when the fracture mode is ductile fracture. Therefore, in order to ensure the strength of welded components, a technology is needed to predict and prevent the occurrence of brittle fracture.

[0004] As a test method for evaluating the strength and fracture mode of welded components, the International Standard ISO 14272 and the Japanese Industrial Standard JIS Z 3137 specify a method for performing a cross tension test using a test piece made by overlapping two metal plates perpendicularly and welding them. In this test method, the cross tension strength (CTS) is measured by applying a tensile load to the test piece using a tensile testing device until the test piece breaks. The fracture mode is then determined by observing the test piece after fracture.

[0005] Patent Documents 1 and 2 describe methods for predicting which fracture will occur first in a situation where a fracture of a weld, which is a type of brittle fracture, and a fracture of a corona bond, which is a type of ductile fracture, compete with each other by performing finite element analysis on a test specimen used in a cross tension test. The finite element analysis described in Patent Document 1 uses an analytical model in which a crack is provided due to delamination of the corona bond. The finite element analysis described in Patent Document 2 uses an analytical model in which a crack is provided at the end of the sheet separation. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2018-128299 [Patent Document 2] Japanese Patent Application Publication No. 2018-128302 Summary of the Invention [Problem to be solved by the invention]

[0007] In recent years, with the increasing strength of thin steel sheets, there have been an increasing number of cases where minute cracks have occurred near the welds of welded components. For example, when plated steel sheets are used as thin steel sheets, the temperature rise associated with welding can cause part of the plated coating to melt, and cracks caused by liquid metal embrittlement (hereinafter referred to as "LME cracks") can appear near the welds. Furthermore, when welded components are mechanically restrained and used in a hydrogen environment, cracks caused by hydrogen embrittlement cracks can occur.

[0008] The presence of these microcracks does not necessarily result in a decrease in the strength of the welded components, but the tip of the crack may become the initiation point for brittle or ductile fracture. Therefore, when a load exceeding the design load is applied to the welded components, the presence of cracks may affect the fracture mode of the weld.

[0009] However, the analytical models used in the finite element analyses described in Patent Documents 1 and 2 did not take into account the presence of cracks caused by LME cracks, etc. Therefore, the conventional techniques were unable to correctly predict the fracture mode of welded components containing such cracks.

[0010] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a method capable of correctly predicting the fracture mode of a welded member having a crack in a weld. [Means for solving the problem]

[0011] The present inventors have developed a new method for predicting the fracture mode of a welded member in order to solve the above-mentioned problems and achieve the object. The gist of the present invention is as follows.

[0012] [1] A method for predicting the fracture mode of a welded member formed by resistance welding two or more overlapping steel plates when a load is applied to the welded member, comprising: Observing the welded component to measure the shapes of the steel plate, nugget, and corona bond, as well as the positions and shapes of one or more cracks generated in the steel plate; constructing an analytical model of the welded component, including the steel plate, the nugget, the corona bond, and the crack, on a computer based on the measurements; setting a critical stress intensity factor and a critical equivalent plastic strain of the welded member; When the magnitude of the load applied to the welded member is set as a common independent variable by finite element analysis using the analytical model, a first function having a stress intensity factor at the edge of the nugget as a dependent variable; a second function having the equivalent plastic strain in the corona bond as a dependent variable; one or more third functions with the stress intensity factor at the tip of the crack as a dependent variable; and One or more fourth functions with the equivalent plastic strain at the tip of the crack as a dependent variable Seeking a first limit load when the value of the first function reaches the limit stress intensity factor; a second limit load when the value of the second function reaches the limit equivalent plastic strain; one or more third limit loads when the value of the third function reaches the limit stress intensity factor; and One or more fourth limit loads when the value of the fourth function reaches the limit equivalent plastic strain. Determine which of the following is smallest: When the first limit load is smallest, brittle fracture occurs at the edge of the nugget, When the second limit load is smallest, ductile fracture occurs in the corona bond. When any one or more of the third limit loads is smallest, brittle fracture occurs at the crack corresponding to the third limit load. When one or more of the fourth limit loads are the smallest, ductile fracture occurs at the crack corresponding to the fourth limit load. predicting the fracture mode that will first occur in the welded member; A method for predicting fracture morphology of welded components.

[0013] [2] At least one of the cracks is located on the electrode contact surface of the steel sheet; The method described in [1] above.

[0014] [3] At least one of the cracks is located in the corona bond. The method according to [1] or [2] above.

[0015] [4] The chemical composition of at least one of the steel plates satisfies, in mass percentage, the following: carbon 0.05% or more and 0.40% or less, phosphorus 0.100% or less, and sulfur 0.100% or less, and the carbon equivalent C calculated by the following formula (1) eq is 0.40% or more and 0.90% or less, A method according to any one of [1] to [3] above.

number

[0016] [5] At least one of the steel sheets is a zinc-plated steel sheet having a tensile strength of 590 MPa or more. A method according to any one of [1] to [4] above. [Effects of the Invention]

[0017] According to the method of the present invention, in a welded component having a crack in the weld, it is possible to correctly predict the fracture mode of the welded component while taking into account the influence of the presence of the crack. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a flow chart illustrating an example of a method according to the present invention. [Figure 2] 1 shows the shape of a test piece used in a cross-tensile test, where (A) is a cross-sectional view and (B) is a top view. [Figure 3] FIG. 1 is a schematic diagram showing a main part of a welded member having a crack at the electrode contact surface. [Figure 4] FIG. 10 is an enlarged top view of the stamped portion of the welded member. [Figure 5] FIG. 1 is a schematic diagram showing a main part of a welded member having a crack at the position of a corona bond. [Figure 6] FIG. 1 is a perspective view showing an example of an analytical model of a test piece used in a cross tension test. [Figure 7] FIG. 2 is an enlarged cross-sectional view of a main part of an analytical model of a welded member having a crack. [Figure 8] 10 is a graph of four functions obtained for the welded component of sample No. 8. [Figure 9] 1 is a graph showing the relationship between the carbon content contained in a welded member and the stress intensity factor. [Figure 10]1 is a graph showing the relationship between the carbon equivalent and the equivalent plastic strain of a welded member. DETAILED DESCRIPTION OF THE INVENTION

[0019] A method for predicting the fracture mode of a welded component according to the present invention will be described in detail below. The method for predicting the fracture mode of a welded component having a crack in a weld due to LME cracking or the like is a method for predicting the fracture mode of a welded component, taking into account the effect of the crack on the fracture mode. More specifically, by performing finite element analysis using an analytical model in which a crack is introduced near the weld, it is possible to predict the fracture mode of a welded component while taking into account the effect of the crack on the fracture mode. Note that the present invention is not limited to the embodiments described below.

[0020] Fig. 1 is a flowchart showing an example of a method for predicting a fracture mode according to the present invention. As shown in Fig. 1, the method according to the present invention predicts the fracture mode of a welded component by performing finite element analysis according to steps (1) to (6). First, the chemical composition and mechanical properties of the steel plates constituting the welded component, as well as the structure of the welded component, will be described.

[0021] [Chemical composition of steel plate] The phenomenon to be predicted in the method according to the present invention is the fracture mode of a welded component formed by resistance welding two or more overlapping steel plates when a load is applied to the welded component. There are no particular restrictions on the steel plates that make up the welded component, and any steel plates that can be welded by resistance welding may be used.

[0022] In a preferred embodiment, the chemical composition of at least one of the steel plates constituting the welded component to which the method according to the present invention is applied satisfies, by mass percentage, the carbon content of 0.05% or more and 0.40% or less, the phosphorus content of 0.100% or less, and the sulfur content of 0.100% or less, and the carbon equivalent C calculated by the following formula (1) eq is 0.40% or more and 0.90% or less.

[0023]

number

[0024] Here, [C], [Si], [Mn], [Ni], [Cr], [Mo], and [V] are symbols indicating the mass percentage of the elements represented by the respective element symbols in the chemical composition of the steel sheet, and if the element is not contained in the steel sheet, zero is substituted. In this specification, the chemical composition and carbon equivalent of the steel sheet are expressed as mass percentages.

[0025] Carbon: 0.05% or more and 0.40% or less Among the chemical components of steel plate, if the carbon content is less than 0.05%, the fracture morphology may be ductile regardless of the prediction results. If the carbon content exceeds 0.40%, the fracture morphology may be brittle regardless of the prediction results. Therefore, it is preferable that the carbon content be 0.05% or more and 0.40% or less. It is more preferable that the carbon content be 0.06% or more and 0.39% or less.

[0026] Phosphorus less than 0.100% If the phosphorus content of the steel plate is 0.100% or less, the possibility of promoting brittle fracture of welded components is reduced. Therefore, it is preferable that the phosphorus content is 0.100% or less. It is more preferable that the phosphorus content is 0.080% or less. There is no particular lower limit for the phosphorus content, and the phosphorus content may be below the detection limit.

[0027] Sulfur less than 0.100% If the sulfur content of the steel plate's chemical components is 0.100% or less, the possibility of promoting brittle fracture of the welded component is reduced. Therefore, it is preferable that the sulfur content is 0.100% or less. It is more preferable that the sulfur content is 0.080% or less. The lower limit of the sulfur content is not particularly limited, and the phosphorus content may be below the detection limit. It is preferable that at least one steel plate of the two or more steel plates constituting the welded component satisfies the above-mentioned chemical compositions of carbon, phosphorus, and sulfur, and it is more preferable that all steel plates satisfy the above-mentioned chemical compositions.

[0028] Carbon equivalent: 0.40% or more, 0.90% or less Among the chemical components of a steel plate, if the carbon equivalent is less than 0.40%, the fracture morphology may be ductile regardless of the prediction results. If the carbon equivalent exceeds 0.90%, the fracture morphology may be brittle regardless of the prediction results. Therefore, the carbon equivalent is preferably 0.40% or more and 0.90% or less. The carbon equivalent is more preferably 0.45% or more and more preferably 0.85% or less. It is preferable that the carbon equivalent determined for at least one steel plate of two or more steel plates constituting a welded component satisfies the above chemical components, and it is more preferable that the carbon equivalent determined for all steel plates satisfies the above chemical components.

[0029] Silicon: 0.05% or more and 2.50% or less Silicon is a useful element that contributes to improving the strength of steel and suppressing carbides through solid solution strengthening. If the silicon content of the chemical components of steel sheet is less than 0.05%, these effects may not be obtained. If the silicon content exceeds 2.50%, the steel sheet may become embrittled. In either case, this may cause a decrease in the accuracy of predicting the fracture morphology of welded components. Therefore, it is preferable that the silicon content be 0.05% or more and 2.50% or less. It is more preferable that the silicon content be 0.10% or more and more preferably 2.00% or less.

[0030] Manganese: 0.5% or more, 4.5% or less Manganese is an element that improves hardenability and contributes to improving the strength of the HAZ. If the manganese content of the steel sheet is less than 0.5%, the HAZ may be excessively softened locally, resulting in ductile fracture regardless of the predicted fracture mode. If the manganese content exceeds 4.5%, the fracture mode may be brittle fracture regardless of the predicted fracture mode. Therefore, it is preferable that the manganese content be 0.5% or more and 4.5% or less. It is more preferable that the manganese content be 1.0% or more and 4.0% or less.

[0031] Aluminum: 0.005% or more, 0.100% or less Aluminum acts as a deoxidizer and is also a solid-solution strengthening element. If the aluminum content of the steel sheet is less than 0.005%, these effects may not be obtained. If the aluminum content exceeds 0.100%, the quality of the slab during steelmaking may deteriorate. In either case, this causes a decrease in prediction accuracy, so the aluminum content is preferably 0.005% or more and 0.100% or less. It is more preferable that the aluminum content be 0.010% or more and 0.080% or less.

[0032] Nitrogen is 0.010% or less Nitrogen forms coarse nitrides, which can become the starting point for void formation during deformation, which can reduce the accuracy of prediction. Therefore, it is preferable that the nitrogen content of the steel plate be as low as possible. However, from the viewpoint of manufacturing costs, the nitrogen content is preferably 0.010% or less, and more preferably 0.006% or less. The lower limit of the nitrogen content is not particularly limited, but the lower limit currently industrially feasible is approximately 0.0003%, and is substantially higher. It is preferable that at least one steel plate of the two or more steel plates constituting the welded component satisfy the above-mentioned chemical compositions of silicon, manganese, aluminum, and nitrogen, and it is more preferable that all steel plates satisfy the above-mentioned chemical compositions.

[0033] Titanium 0.5% or less, Boron 0.010% or less, Nickel 0.5% or less, Chromium 1.0% or less, Molybdenum 0.5% or less, Vanadium 0.5% or less, Antimony 0.5% or less, Calcium 0.5% or less, Niobium 0.5% or less, Total rare earth elements (REM) 0.5% or less The steel sheet may optionally contain these elements. The inclusion of these optional elements makes it possible to strengthen the steel sheet and control precipitates. If the content of these optional elements exceeds the upper limit of each element, the effect of addition may saturate or grain boundary embrittlement may occur.

[0034] The balance of the chemical components of the steel sheet is iron and unavoidable impurities.

[0035] [Mechanical properties of steel sheets, etc.] In carrying out the method according to the present invention, the values ​​of the yield stress and tensile stress, which are among the mechanical properties of the steel plates constituting the welded components, must be known in advance. This is because these values ​​must be input into the analytical model used in the finite element analysis. The values ​​of the yield stress and tensile stress can be determined by actual measurements in a tensile test using a test piece. Alternatively, when a standard steel grade is used for the steel plate, standard values ​​or literature values ​​of the yield stress and tensile stress for that steel grade may be used.

[0036] In a preferred embodiment, at least one of the steel sheets constituting the welded component targeted by the method of the present invention is a zinc-based plated steel sheet having a tensile strength of 590 MPa or more. As described above, when plated steel sheets are welded, the temperature rise associated with welding can cause cracks due to LME cracking near the weld. Furthermore, in so-called high-tensile steels with a tensile strength of 590 MPa or more, the heat-affected zone (hereinafter referred to as "HAZ") is prone to hardening, so consideration must be given to fracture strength in welded components made of high-tensile steel. By using the method of the present invention, it is possible to accurately predict the fracture mode of a welded component made of zinc-based plated steel sheets having a tensile strength of 590 MPa or more and having a crack. It is preferable that at least one of the two or more steel sheets constituting the welded component is a zinc-based plated steel sheet having a tensile strength of 590 MPa or more, and more preferably that all of the steel sheets are zinc-based plated steel sheets having a tensile strength of 590 MPa or more.

[0037] In a preferred embodiment, a steel sheet can be used that has been coated with a zinc-based plating by various methods such as hot-dip galvanizing, electroplating, vapor deposition, thermal spraying, etc. Specifically, hot-dip galvanized steel sheet (GI), alloyed hot-dip galvanized steel sheet (GA), hot-dip Zn-5 mass% Al alloy-plated steel sheet (GF), hot-dip Zn-55 mass% Al alloy-plated steel sheet (GL), electrogalvanized steel sheet (EG), electrogalvanized zinc-Ni alloy-plated steel sheet (Zn-11 mass% Ni), etc. can be used, but the present invention is not limited to these, and all known zinc-based plated steel sheets containing zinc can be used.

[0038] [Welded material structure] The welded component targeted by the method according to the present invention is a welded component formed by resistance welding two or more overlapping steel plates. A weld in a welded component is formed where two or more steel plates overlap, and is composed of a portion where molten metal has solidified. In this specification, the portion where molten metal has solidified (weld) will hereinafter be referred to as a "nugget." It is preferable that the nugget is formed across all of the steel plates that make up the welded component.

[0039] The welded portion of the welded member is formed by resistance welding. Resistance welding is a welding method in which a large current is passed between steel sheets using electrodes, heating the contact points by Joule heat and simultaneously applying pressure to join the steel sheets. In the present invention, the resistance welding may be any type of resistance welding that forms a nugget. The resistance welding is typically preferably spot welding, but may also be seam welding or projection welding.

[0040] The shape and size of the welded members used in the method according to the present invention are not particularly limited, and the overall structure of the welded members may be of any size and shape. Furthermore, the thickness of the steel plate is also not particularly limited. However, when comparing the results with test results using actual test pieces, it is preferable to use welded members having a standardized size and shape from the standpoints of ease of producing test pieces and reproducibility of the test.

[0041] Figure 2 shows the shape of the test specimen used in the cross-tensile test specified in the aforementioned International Standard ISO 14272 and Japanese Industrial Standard JIS Z 3137. This test specimen consists of two identical metal plates, an upper plate 2a and a lower plate 2b, each 150 mm long and 50 mm wide. These plates are overlapped so that their longitudinal directions intersect at right angles, and the upper plate 2a and the lower plate 2b are welded together using spot welding or projection welding. As shown in the cross-sectional view (A), a nugget 3 is formed across the upper and lower plates 2a and 2b. The cross-tensile strength of the test specimen can be measured by mounting the test specimen on a jig using the fixing holes shown in the top view (B). A tensile load is applied in the Z-axis direction along which the nugget 3 peels using a tensile testing machine. The fracture morphology can also be determined by observing the test specimen after fracture.

[0042] In the above explanation, a test piece used in a cross tension test has been used as an example of a standardized welded component structure, but the method according to the present invention is not limited to a specific type of load. In addition to test pieces used in cross tension tests, the present invention can also be applied to welded components having other structures in which different types of loads are applied to the weld, such as test pieces used in shear tension tests and L-tensile tests. In the case of a shear tension test, the type of load applied to the nugget is shear stress. In the case of an L-tensile test, the type of load applied to the nugget is tensile stress accompanied by bending.

[0043] [(1) Observation of welded components] Next, each step of the method according to the present invention will be described. In step (1) of the method according to the present invention, the welded component is observed to measure the shapes of the steel plate, nugget, and corona bond, as well as the position and shape of one or more cracks that have occurred in the steel plate. The purpose of these measurements is to construct an analytical model on a computer in the next step (2) based on the measurements performed in step (1).

[0044] The method for measuring the position and shape in observing a welded component is not particularly limited, but for example, the position and shape can be measured by cutting the welded component at a cross section that includes the center of the nugget and is parallel to the plate thickness direction, polishing the cross section, taking a metallographic photograph of the polished cross section, and quantifying the dimensions of each part of the photographed image by image analysis. It is preferable to ensure that the magnification of the metallographic photograph is sufficient to observe the crack and accurately measure its shape.

[0045] FIG. 3 is a schematic diagram showing a portion of a welded component with a crack at the electrode contact surface. In the welded component 1, the overlapping portion of an upper sheet 2a and a lower sheet 2b, each made of steel sheets 2, is welded by resistance welding. A nugget 3 is formed in the hatched area. A gap called a sheet separation 6 exists at a portion of the boundary between the upper sheet 2a and the lower sheet 2b, formed by thermal expansion of the steel sheets during welding. A joint called a corona bond 4, which is formed by solid-state welding, is formed between the end 3a of the nugget and the end of the sheet separation 6. The corona bond 4 exists in a ring shape around the periphery of the nugget 3. A mark 7 is formed on the electrode contact surface 2c of the upper sheet 2a and the lower sheet 2b due to pressure applied by the electrodes during resistance welding. Around the mark 7, a transition portion of the sheet thickness, called an electrode shoulder 8, exists. In this specification, the term "electrode contact surface" refers to the entire surface of the two surfaces of the steel plate 2 that comes into contact with the electrode during resistance welding, that is, the surface on which the engraved mark 7 is present.

[0046] In a preferred embodiment, at least one of the one or more cracks generated in the steel sheet is located on the electrode contact surface of the steel sheet. In the example shown in FIG. 3, one crack 5 is present at the edge of the electrode shoulder 8 on the electrode contact surface 2c. The crack 5 propagates from the crack opening 5b on the electrode contact surface 2c toward the crack tip 5a located inside the steel sheet 2. When the crack is generated by LME cracking, the crack propagates normally in a direction perpendicular to the electrode contact surface 2c, i.e., along the Z axis in FIG. 3, as shown in FIG. 3. On the other hand, when the crack is caused by hydrogen embrittlement cracking, the crack propagates normally in the in-plane direction of the steel sheet 2, i.e., along the XY plane in FIG. 3. The crack may be located not only at the positions shown in FIG. 3 but also in the embossed mark 7, the electrode shoulder 8, or the electrode contact surface 2c near the electrode shoulder 8.

[0047] Figure 4 is an enlarged top view of the dents on the welded parts. When resistance welding is performed by spot welding, the tip of the electrode is usually circular, so the dents 7 are circular. As shown in Figure 4, the crack 5 may be present on a portion of the outer circumference of the electrode shoulder 8, or multiple cracks 5 may be present intermittently along the outer circumference. The length of the crack opening 5b measured along the outer circumference of the electrode shoulder 8 is usually several millimeters to several tens of millimeters. Although extremely rare, long cracks 5 may form semicircularly along the outer circumference of the electrode shoulder 8, as shown in the right half of Figure 4. The distribution of the crack openings 5b in the XY plane cannot be observed from the cross-sectional view shown in Figure 3, but can be measured by observing the electrode contact surface 2c from the Z-axis direction as shown in Figure 4. Even if the cracks 5 occur in the dents 7, the electrode shoulder 8, or the electrode contact surface 2c near the electrode shoulder 8, as described above, the distribution of the cracks 5 in the XY plane can be observed using the above method.

[0048] In a preferred embodiment, at least one of the cracks occurring in the steel sheet is located at the corona bond. Figure 5 is a schematic diagram showing a main portion of a welded component with a crack at the corona bond. Cracks 5 may occur not only at the electrode contact surface 2c but also at the corona bond 4, which is not visible from the outside of the welded component 1, as illustrated in Figure 5. When one or both of the upper and lower sheets 2a and 2b are plated steel sheets, traces of the plated coating are present at the corona bond 4, which may lead to the occurrence of crack 5 due to LME cracking. In the example shown in Figure 5, crack 5 propagates from the crack opening 5b located at the corona bond 4 toward the crack tip 5a located inside the steel sheet 2. The direction of crack 5 propagation is perpendicular to the electrode contact surface 2c, i.e., along the Z axis in Figure 5. To observe the distribution of cracks 5 occurring at the corona bond 4 in the XY plane, for example, the fracture surface of the corona bond 4 after fracture in a cross-tensile test can be observed.

[0049] Specific examples of parameters obtained by measuring the shapes of the steel plate, nugget, and corona bond observed in step (1) include the thickness t of the steel plate, the radius of the nugget 3, the radius of curvature of the edge 3a of the nugget, and the width a of the corona bond 4. C , the radius of the indentation 7, the indentation 7a which is the depth of the indentation 7 based on the electrode contact surface 2c, and the width 8a of the electrode shoulder. Specific examples of parameters obtained by measuring the position and shape of the crack include the position and length a of the crack 5, L The position of the crack 5 can be identified, for example, by the positions of the crack tip 5a and the crack opening 5b on the observation surface. L can be identified by the distance from the tip 5a of the crack to the opening 5b of the crack on the observation surface. The position of the crack 5 is not limited to one location, and may be two or more locations.

[0050] As mentioned above, depending on the type of steel of the steel plate 2, the HAZ may be harder than the base material. In such cases, it is preferable to measure the position where the HAZ is formed and its hardness in step (1), and reflect the parameters obtained by these measurements in the analysis model. Specifically, the position and hardness of the HAZ formed around the nugget 3 can be measured by measuring the distribution of Vickers hardness in the cross section of the welded component shown in Figure 3. Generally, three regions are formed, in order from the region closest to the nugget 3: a hard region, a region of intermediate hardness, and a soft region.

[0051] (2) Construction of analytical model In step (2) of the method according to the present invention, an analytical model of the welded component, including the steel plate, nugget, corona bond, and crack, is constructed on a computer based on the measurements made in step (1). In this specification, the term "analytical model" refers to a virtual numerical analysis model used in the finite element analysis performed in steps (3) and (4), in which the shape of the welded component is divided into a finite number of elements called a mesh. Such analytical models are sometimes called solid models. In step (2), it is preferable to construct an analytical model that reproduces as faithfully as possible the shape of the actual welded component and the position and shape of the crack, based on the parameters obtained by the measurements made in step (1).

[0052] FIG. 6 is a perspective view showing an example of an analytical model of a test piece used in the cross tension test shown in FIG. 2. The origin of the XYZ coordinate system in the analytical model shown in FIG. 6 is set at the center of the nugget 3. The welded component 1 made up of this test piece is plane-symmetric with respect to a plane passing through the center of the nugget 3 and perpendicular to the X axis, and is also plane-symmetric with respect to a plane passing through the center of the nugget 3 and perpendicular to the Y axis. In constructing the analytical model, a quarter-solid model shown in FIG. 6 is first created, and then this is expanded to the opposite side of the plane of symmetry, thereby constructing an analytical model of the entire welded component 1. In this way, by constructing an analytical model with attention to the symmetry of the welded component, work efficiency can be improved.

[0053] Based on the parameters of the crack position and shape measured in step (1), a crack can be introduced into the analytical model by creating a region where mesh integration is not performed at the same position in the constructed analytical model. However, as mentioned above, the distribution of the cracks observed in step (1) on the XY plane may not necessarily be highly symmetric. Therefore, when expanding the 1 / 4 solid model, it is necessary to introduce cracks into the analytical model based on the measured parameters, while paying attention to the symmetry of the cracks.

[0054] For example, suppose the cracks 5 observed in step (1) are distributed over half of the circumference of the electrode shoulder 8 in the region where the X coordinate of the quarter solid model is positive, as shown in the right half of Figure 4. In this case, when the quarter solid model shown in Figure 6 is expanded symmetrically with respect to the YZ plane, cracks are introduced by not integrating the mesh in the region where the X coordinate is positive with respect to the YZ plane, and cracks are not introduced by maintaining the mesh integrated state on the negative side. This allows us to construct a half-symmetric solid model with cracks that faithfully reflects the observation results. Furthermore, when symmetry is low, such as when multiple cracks 5 are intermittently distributed on the circumference of the electrode shoulder 8, as shown in the left half of Figure 4, cracks can be introduced one by one into the analytical model based on the measured parameters.

[0055] As mentioned above, corona bonds are solid-phase welded, so they can be treated as a continuum in the analysis model. However, when a load is applied to a welded component, it is expected that the corona bond will separate at the beginning of deformation, and then the bond will be released and a crack will form, affecting the fracture mode of the entire welded component. Therefore, in this invention, mesh integration is not performed on corona bonds, just like cracks. In this case, the length of the region where mesh integration is not performed can be set based on the measurement in step (1), and in its longest case, it will be the width a of corona bond 4 shown in Figure 3. C The nugget end 3a is one end of the region where mesh integration is not performed.

[0056] Fig. 7 is an enlarged cross-sectional view of a main part of an analytical model of a welded component having a crack. As shown in Fig. 7, by making the mesh size near the nugget end 3a and the crack tip 5a finer than the mesh size at other positions, the accuracy of calculations in finite element analysis can be improved. The mesh size near the nugget end 3a and the crack tip 5a is preferably 25 µm or less. Furthermore, the area where the fine mesh is provided is preferably an area with a radius of 0.2 mm or more centered on the nugget end 3a and the crack tip 5a.

[0057] In the example of the analysis model shown in FIG. 7, three regions are defined, in order from the position closest to the nugget 3, based on the hardness measurement in step (1): HAZ (hard) 2d, HAZ (intermediate) 2e, and HAZ (soft) 2f. The thresholds defining the boundaries of these regions can be determined, for example, by dividing the numerical range from the minimum to maximum measured hardness into three. The mesh size in the HAZ can be the finest in the HAZ (hard) 2d, the coarsest in the HAZ (soft) 2f, and an intermediate size in the HAZ (intermediate) 2e. This improves the accuracy of calculations in the HAZ (hard) 2d region, where brittle fracture is likely to occur. It is preferable that the mesh size in the region of the nugget 3 adjacent to the HAZ (hard) 2d be as fine as the mesh size in the HAZ (hard) 2d.

[0058] As explained above, in step (2), cracks are basically introduced into the analytical model based on the parameters obtained by measuring the crack position and shape in step (1). However, depending on the purpose of implementing the method of the present invention, while based on the observation results, cracks of arbitrary positions, shapes, and numbers different from those that faithfully reproduce the positions and lengths of cracks actually observed may be introduced into the analytical model. By conducting such numerical experiments, the effect of the presence of cracks on the strength of welded components can be predicted under a wider range of conditions.

[0059] [(3) Setting of limit stress intensity factor and limit equivalent plastic strain] In step (3) of the method according to the present invention, the critical stress intensity factor K f and limit equivalent plastic strain ε f Set the critical stress intensity factor K f and limit equivalent plastic strain ε f are physical quantities specific to the steel plates that make up the welded components.

[0060] The stress intensity factor K is a physical quantity that represents the strength of the stress distribution near the tip of a crack. There are three modes of crack deformation depending on the direction of the load applied to the crack, and the stress intensity factor K is defined individually for each mode. In this invention, the stress intensity factor K corresponding to mode I, which is an in-plane opening type, is used. The critical stress intensity factor K f is the limit value of the stress intensity factor K at the tip of the crack when brittle fracture occurs at the crack. As will be described later, the limit stress intensity factor K f has a correlation with the carbon content of the steel plate.

[0061] Equivalent plastic strain ε is a physical quantity that converts the strain of a material that undergoes plastic deformation under multiaxial stress into an amount equivalent to the axial strain under uniaxial tension. f is the limit value of the equivalent plastic strain ε at the tip of the crack when ductile fracture occurs at the crack. As will be described later, the limit equivalent plastic strain ε f has a correlation with the carbon equivalent of the steel plate.

[0062] These limit values ​​may be determined by a combination of experiments and numerical calculations, or by numerical calculations alone based on the physical properties of the steel plates constituting the welded components. When combining experiments and numerical calculations, first, steel pieces for cross tension tests are prepared from multiple steel types with different carbon contents and carbon equivalents. Next, using the resulting steel pieces, multiple test pieces for cross tension tests with different nugget radii are prepared under crack-free conditions by varying the resistance welding conditions. Next, cross tension tests are performed on these test pieces, and the fracture stroke when the test piece fractures is measured. Furthermore, the test piece is observed after fracture to identify the fracture location, and the fracture surface is observed to determine whether the fracture occurred due to brittle or ductile fracture. Generally, when the nugget radius is small, brittle fracture occurs at the nugget, while when the nugget radius is large, ductile fracture tends to occur at the corona bond.

[0063] Next, for the test specimens in which brittle fracture occurred and the test specimens in which ductile fracture occurred, analytical models without introducing cracks were constructed according to the above-mentioned procedures (1) and (2). Next, finite element analysis was performed using the constructed analytical models. For the test specimens in which brittle fracture occurred, the stress intensity factor K at the fracture position in the fracture stroke was calculated, and this value was used as the critical stress intensity factor K. f For test pieces that have undergone ductile fracture, the equivalent plastic strain ε at the fracture position during the fracture stroke is calculated, and this value is called the limit equivalent plastic strain ε f The above-mentioned experiments and finite element analysis are repeated for a plurality of test pieces made of steel plates with different carbon contents and carbon equivalents and with different nugget radii, and data is accumulated. Details of the finite element analysis using the analytical model will be described later.

[0064] Next, by analyzing the data of the specimens that had undergone brittle fracture among the accumulated data, the relationship between the carbon content and the critical stress intensity factor K f The constants a and b included in the following equation (2) that expresses the relationship between: The two constants included in equation (2) can be determined experimentally, for example, by the least squares method.

[0065]

number

[0066] Here, "[C]" is the carbon content of the steel plates that make up the welded component, and "ln" is the natural logarithm. If the carbon content of the two or more steel plates that make up the welded component is not the same, the highest value is used.

[0067] Next, by similarly sorting out the accumulated data on test specimens that underwent ductile fracture, the carbon equivalent and limit equivalent plastic strain ε f The constants c and d included in the following equation (3) that expresses the relationship between: The two constants included in equation (3) can also be determined experimentally by, for example, the least squares method.

[0068]

number

[0069] where C eq is the carbon equivalent calculated by formula (1). If the carbon equivalents of the two or more steel plates that make up the welded component are not the same, the highest value is used. ε f In the case where the maximum principal strain is less than 0, ε f =0.01.

[0070] Finally, the critical stress intensity factor K is calculated by substituting the carbon content of the welded components observed in step (1) into equation (2). f and then by substituting the carbon equivalent of the welded material into equation (3), the limit equivalent plastic strain ε f These values ​​are set as limit values ​​specific to the welded component.

[0071] When the constants in equations (2) and (3) are calculated only by numerical calculation, instead of experimentally determining the fracture stroke by a cross tension test, the fracture stroke is calculated by finite element analysis using the physical properties of the steel plate, such as the yield stress and tensile stress.Then, by performing the same data processing as above, the constants in equations (2) and (3) can be calculated.

[0072] When carrying out the method according to the present invention, it is not necessary to accumulate data and determine the constants included in equations (2) and (3) in step (3) each time, and they may be omitted. For example, if the constants included in equations (2) and (3) have already been determined by statistical analysis based on previously accumulated data, there is no need to accumulate new data, and step (3) is carried out using the previously determined constants to determine the critical stress intensity factor K. f and limit equivalent plastic strain ε f can be calculated.

[0073] [(4) Function calculation] In step (4) of the method according to the present invention, a finite element analysis is performed using the analytical model constructed in step (2) to determine a first function with the stress intensity factor at the nugget end as a dependent variable, a second function with the equivalent plastic strain in the corona bond as a dependent variable, one or more third functions with the stress intensity factor at the crack tip as a dependent variable, and one or more fourth functions with the equivalent plastic strain at the crack tip as a dependent variable, when the magnitude of the load applied to the welded members is used as a common independent variable.

[0074] Table 1 shows the common independent variables, dependent variables, and relational expressions for each of the four functions, from the first function f1 to the fourth function f4, using symbols.

[0075] [Table 1]

[0076] The "node position" shown in the first row of Table 1 refers to the node position of the mesh, which is the element that makes up the analytical model used in finite element analysis. In other words, each of the four functions has as its dependent variable the stress intensity factor K or equivalent plastic strain ε found at the node position shown in Table 1. On the other hand, the load W, which is a common independent variable, is a quantity that indicates the magnitude of the load applied to the test piece, and the load W does not change depending on the node position.

[0077] Of the three types of node positions shown in Table 1, both brittle and ductile fracture can occur at the crack tip. Therefore, the third function f3 uses the stress intensity factor K, which governs brittle fracture, as its dependent variable, and the fourth function f4 uses the equivalent plastic strain ε, which governs ductile fracture, as its dependent variable. Since the present invention deals with one or more cracks, in step (4) a number of third and fourth functions equal to the number of cracks are calculated. At the nugget tip, the molten metal that constitutes the nugget is hard and brittle, so if fracture occurs, it is always brittle fracture, and ductile fracture need not be considered. In contrast, in corona bonding, if fracture occurs, it is basically ductile fracture, and brittle fracture need not be considered.

[0078] The finite element analysis performed in step (4) can be performed using finite element analysis software. The analysis begins by inputting stress-strain curve data for each part of the analytical model that constitutes the welded component. For the base metal region of the welded component, excluding the nugget and HAZ, the stress-strain curve obtained by a tensile test on an uncracked steel plate can be input. For the elastic region of the stress-strain curve in the nugget and HAZ below the yield stress, the stress-strain curve obtained by a tensile test on an uncracked steel plate multiplied by the ratio of the hardness of the nugget or HAZ region to the hardness of the base material can be input. For the plastic region above the yield stress, the data can be input so that it is continuous with the curve below the yield stress using the Swift equation shown in Equation (4) below.

[0079]

number

[0080] where σ is the true stress, σ Y is the yield stress, ε P is the true plastic strain, A is the material constant, and N is the hardening coefficient.

[0081] Since it is cumbersome to input the above data for all regions of the HAZ, as mentioned above, the data input work can be simplified by dividing the HAZ into three regions, for example, HAZ (hard), HAZ (intermediate), and HAZ (soft), and inputting a common representative stress-strain curve for each region.

[0082] Once the stress-strain curves have been input into the analysis model, a simulation of the cross tension test is performed using finite element analysis software, and four functions are calculated with the load W at that time as a common independent variable. The stress intensity factor K at the node positions of the crack tip and nugget end can be calculated using the following equation (5), assuming a plane strain state.

[0083]

number

[0084] Here, J is the J integral, E is Young's modulus, and ν is Poisson's ratio. The J integral is obtained by performing the J integral on the integral path surrounding the node position. As mentioned above, in step (2), the mesh size near the crack tip and nugget end is made finer than usual, so a stable solution for the J integral can be obtained. However, if the maximum principal stress at the node position is less than 0, the value of K is uniformly set to 0 (zero).

[0085] The equivalent plastic strain ε at the corona bond and crack tip is automatically calculated using finite element analysis software based on the principal strain and Poisson's ratio of each component. Referring again to Figure 7, the equivalent plastic strain ε at the corona bond tends to be greatest at the strain concentration point 4a, located a short distance from the nugget edge 3a toward the corona bond 4. Therefore, for the second function f2, the node at the strain concentration point 4a where the equivalent plastic strain ε is greatest is identified in advance among the multiple nodes located on the corona bond 4. The relationship between the load W and the equivalent plastic strain ε at the identified node is then calculated as the second function f2. By performing this procedure, the fracture mode of the welded component can be predicted more accurately.

[0086] [(5) Calculation of limit load] In step (5) of the method according to the present invention, it is determined which of the following is smallest: the first limit load when the value of the first function reaches the limit stress intensity factor; the second limit load when the value of the second function reaches the limit equivalent plastic strain; one or more third limit loads when the value of the third function reaches the limit stress intensity factor; and one or more fourth limit loads when the value of the fourth function reaches the limit equivalent plastic strain.

[0087] Table 2 shows the threshold values ​​and relational expressions with symbols for each of the four types of limit loads, from the first limit load W1 to the fourth limit load W4.

[0088] [Table 2]

[0089] The four functions, from the first function f1 to the fourth function f4 shown in Table 2, are known by procedure (4). f and limit equivalent plastic strain ε f is also known from step (3). Therefore, by using these known functions and calculated values, it is possible to determine which of the four types of limit loads, from the first limit load W1 to the fourth limit load W4, is the smallest. Specifically, the four types of limit loads are determined by plotting the four types of functions, from the first function f1 to the fourth function f4, obtained in step (4), on a graph and calculating the critical stress intensity factor K f or limit equivalent plastic strain ε f Alternatively, the four types of limit loads can be calculated by multiplying the inverse functions of the four types of functions above by the limit stress intensity factor K f or limit equivalent plastic strain ε f Alternatively, the output may be calculated by inputting the value of

[0090] As mentioned above, the second function f2 at the node position of the corona bond is calculated as a function at the node located at the strain concentration point where the equivalent plastic strain ε is maximized among the multiple nodes located on the corona bond. Therefore, the value of the second function f2 is the limit equivalent plastic strain ε f The second limit load W2 when the load reaches the limit is also the maximum value among the limit loads at the nodes located in the corona bond. As with the function described in step (4), in step (5), the third limit load W3 and the fourth limit load W4 are calculated in numbers equal to the number of cracks. On the other hand, for the edge of the nugget, it is sufficient to calculate the first limit load W1 at the edge 3a of the nugget.

[0091] (6) Prediction of fracture morphology In step (6) of the method of the present invention, based on the smallest limit load determined in step (5), it is predicted that the fracture mode that will first occur in the welded component will be brittle fracture at the edge of the nugget if the first limit load is the smallest, ductile fracture at the corona bond if the second limit load is the smallest, brittle fracture at a crack corresponding to the third limit load if any of one or more third limit loads is the smallest, or ductile fracture at a crack corresponding to the fourth limit load if one or more fourth limit loads are the smallest.

[0092] Table 3 shows the minimum limit load in the method according to the present invention, and the fracture modes of the welded member predicted when the minimum limit load is reached, along with their abbreviations.

[0093] [Table 3]

[0094] The prediction of fracture mode in the present invention shown in Table 3 is based on the principle that when the load applied to the welded component is increased and the load reaches the smallest of the four limit loads determined in step (5), fracture will occur at the node corresponding to that limit load. According to the method of the present invention, when multiple possible fracture locations are assumed, such as the edge of the nugget, the corona bond, and the tips of one or more cracks, it is possible to predict the location where fracture will first occur, and if the location where fracture will occur is a crack, it is also possible to predict whether the fracture mode will be brittle fracture or ductile fracture.

[0095] Although Tables 2 and 3 list four types of limit loads, from 1 to 4, it is not always possible to find all of these limit loads in step (5). For example, if the increase in the function value K or ε with respect to the increase in load W in the function found in step (4) is slow, the value of the function may not reach the critical stress intensity factor K within the range of the calculated load W. f or limit equivalent plastic strain ε fIn such cases, the critical load for the function cannot be calculated, so in step (6), it is sufficient to determine which is the smallest critical load among those calculated in step (5).

[0096] Furthermore, if there are two or more minimum values ​​among the first to fourth limit loads, it is not possible to narrow down the predicted fracture mode to one. In such cases, it is sufficient to predict that one of the multiple fracture modes that shows the minimum value will occur first, or that these fracture modes will occur simultaneously. [Example]

[0097] The present invention will be described in more detail below with reference to examples, but the present invention is not limited to the following examples in any way.

[0098] [Preparation of sample pieces] Thirteen types of steel plates, designated by steel plate symbols A through M, were prepared, with the chemical compositions (the balance being Fe and unavoidable impurities), carbon equivalents, and thicknesses shown in Table 4. All steel plates had the shapes of upper plate 2a and lower plate 2b shown in Figure 1, and were 100 mm long and 50 mm wide. A 20 mm diameter hole was provided at each end of the steel plate in the longitudinal direction to secure the steel plate to a jig. Tensile tests were conducted on these steel plates to measure their yield strength and tensile strength. The obtained yield strength and tensile strength values ​​are shown in Table 4.

[0099] [Table 4]

[0100] Next, one upper and one lower plate were selected from steel plates designated A to M, and the two selected steel plates were spot welded together with their centers perpendicular to each other to prepare 14 types of test specimens for cross tension tests. The steel plate combinations used in the prepared test specimens are shown in Table 5.

[0101] [Table 5]

[0102] As shown in Table 5, the test piece for sample No. 7 was made by combining steel sheets with different chemical compositions, while the other test pieces were made by combining steel sheets of the same type.

[0103] Next, in addition to the steel plates shown in Table 4, several types of steel plates with different carbon contents and carbon equivalents were prepared. Using these steel plates, test specimens for cross tension tests were prepared in addition to the 14 types of test specimens for cross tension tests shown in Table 5. These test specimens were prepared, together with the test specimens shown in Table 5, for the purpose of using them in step (3).

[0104] [Step (1)] Next, cross-tensile tests were performed on all test specimens, including the 14 types shown in Table 5 and other test specimens prepared separately, to measure the stroke (uniaxial displacement) at fracture. The test specimens were also visually inspected after fracture to identify the fracture location and fracture morphology. The fractured test specimens were then cut along a cross section parallel to the plate thickness direction, including the center of the nugget. The cross section was then polished and observed using an optical microscope to measure the shapes of the steel plate, nugget, and corona bond. Parameters measuring the shape of the steel plate included the radius of the mark 7 created by spot welding, the indentation 7a, the width 8a of the electrode shoulder, and the radius of curvature of the tip of the sheet separation 6, as shown in Figure 3. For the nugget 3, the radius and the radius of curvature R of the nugget edge 3a were measured. For the corona bond 4, the length a C As part of these measurement results, only the parameters measured for 14 types of test specimens are shown in Table 5. However, for test specimens No. 1 to 14 shown in Table 5, the radius of curvature of the tip of the sheet separation 6 was 0.15 mm, and the width 8a of the electrode shoulder was constant at 1.5 mm, so these parameters were omitted from Table 5.

[0105] Next, the positions of the crack tip 5a and the crack opening 5b are identified using the same observation surface for all test pieces, and the crack length a LIn all test pieces, only one crack was present at the electrode contact surface 2c or the corona bond 4. The measured crack length a L The results are shown in Table 5. The electrode contact surface 2c was also observed from a direction perpendicular to the Z axis shown in Figure 4, and the distribution of cracks 5 was observed and identified. The locations of the identified cracks are also shown in Table 5. Regarding the crack locations shown in Table 5, those marked "electrode shoulder" indicate that cracks 5 were formed at the electrode shoulder 8, with a length of several millimeters to several tens of millimeters. Furthermore, those marked "stamp mark" indicate that cracks 5 were formed in a semicircular shape at the outer circumferential position of the stamp mark 7.

[0106] [Step (2)] Next, analytical models were constructed for all test specimens on a computer based on the obtained measurements. The finite element analysis software ABAQUS Ver. 6-12-1 was used to construct the analytical models and perform the finite element analysis described below. The analytical models were constructed according to the method described above, so as to reproduce as faithfully as possible the shapes of the steel plate, nugget, and corona bond of the actual welded components.

[0107] [Step (3)] Next, for all test pieces, based on the results of the cross tension test, the stress intensity factor K or equivalent plastic strain ε at the node at the fracture location when a fracture stroke was applied to the welded member was calculated by finite element analysis. However, no cracks were introduced into the analytical model used in the calculation of step (3). The calculated value of the stress intensity factor K was used as the limit stress intensity factor K. f The equivalent plastic strain ε is set to the limit equivalent plastic strain ε f The critical stress intensity factor K f and limit equivalent plastic strain ε f In theory, if the material is the same, it is considered that the value will be invariant regardless of whether or not there is a crack. f and ε fWhen calculating by numerical calculation, it is not necessary to introduce cracks into the analytical model. Also, as in the calculation of step (2), in step (3), mesh integration is not performed for the corona bond, just as it is for the crack.

[0108] Next, the obtained data were sorted and the four constants included in equations (2) and (3) were determined by the least squares method, resulting in the following equations (6) and (7).

[0109]

number

[0110]

number

[0111] Next, for the 14 types of test specimens shown in Table 5, the carbon content [C] shown in Table 4 was substituted into equation (6) to obtain the critical stress intensity factor K f Determine the carbon equivalent C shown in Table 4 eq Substituting into equation (7) gives the limit equivalent plastic strain ε f For the test piece of sample No. 7, which used a steel plate with steel plate designation G as the upper plate and a steel plate with steel plate designation A as the lower plate, the carbon content [C] and carbon equivalent C of both plates were determined. eq The higher value of the steel plate with the steel plate symbol G was used. The limiting stress intensity factor K f and limit equivalent plastic strain ε f The values ​​are shown in Table 6.

[0112] [Table 6]

[0113] [Step (4)] Next, finite element analysis was performed using the 14 analytical models constructed above, and four functions were obtained when the load W applied to the welded member was used as a common independent variable. Cracks were introduced into all analytical models used in the calculations in step (4) according to the method described above.

[0114] Figure 8 shows graphs of the four functions obtained for the welded component of sample No. 8. In this analytical model, the following differences were observed among the four functions in the way the stress intensity factor K and equivalent plastic strain ε changed when a load W was applied to the welded component. For the first function f1, the value of the stress intensity factor K increased from the beginning, but the change became slower once the load W exceeded 300 kN. For the second function f2, the value of the equivalent plastic strain ε increased sharply once the load W exceeded 150 kN. For the third function f3, the value of the stress intensity factor K remained near zero regardless of the value of the load W. For the fourth function f4, the value of the equivalent plastic strain ε increased sharply once the load W exceeded 300 kN.

[0115] [Step (5)] Next, the obtained function and the critical stress intensity factor K shown in Table 6 f and limit equivalent plastic strain ε f For example, in the example shown in Figure 8, the value of the fourth function f4 is the limit equivalent plastic strain ε of sample No. 7 shown in Table 6. f The load W when it reaches the value of 2.28 can be read from the graph as being approximately 470kN, and this value was set as the fourth limit load W4. For the other three functions, none of the function values ​​reached the limit value within the calculated range.

[0116] [Step (6)] Next, based on the smallest limit load determined in step (5), the fracture mode was predicted according to the criteria shown in Table 3. The predicted fracture mode is shown in Table 6 in comparison with the results of the actual cross tension test. As shown in Table 6, the fracture mode predicted by the method according to the present invention is in good agreement with the actual experimental results.

[0117] Figure 9 shows the curve of Equation (6) on a graph with the carbon content [C] on the horizontal axis and the stress intensity factor K on the vertical axis, and also shows the critical stress intensity factor K shown in Table 6. fAs shown in Fig. 9, in the first fracture mode indicated by the black triangles and the third fracture mode indicated by the black circles, both of which are brittle fractures, the critical stress intensity factor K f The value of was consistent with the curve of equation (6).

[0118] In Figure 10, the horizontal axis is the carbon equivalent C eq The line of Equation (7) is plotted on a graph with the vertical axis representing the equivalent plastic strain ε, and the limit equivalent plastic strain ε shown in Table 6 is plotted. f As shown in Fig. 10, in the second fracture mode indicated by the open triangle and the fourth fracture mode indicated by the open circle, both of which are ductile fractures, the critical equivalent plastic strain ε f The value of was consistent with the curve of equation (7).

[0119] According to the results shown in Fig. 9 and Fig. 10, in the finite element analysis, the stress intensity factor K is f At the node where the equivalent plastic strain ε reaches the limit equivalent plastic strain ε f Conversely, the equivalent plastic strain ε does not reach the limit equivalent plastic strain ε f At the node where the stress intensity factor K reaches the critical stress intensity factor K f In other words, it is predicted that the fracture mode will occur depending on which of the stress intensity factor K and equivalent plastic strain ε reaches its limit value first, and according to Table 6, these predictions are in good agreement with the fracture mode observed in the experiment. [Industrial Applicability]

[0120] The method according to the present invention makes it possible to accurately predict the fracture mode of a welded component having a small crack by numerical simulation, thereby reducing the number of actual experiments or omitting some of the experiments, thereby improving the efficiency of various research and development projects targeting welded components, welding methods, steel plates suitable for welding, etc. [Explanation of symbols]

[0121] 1 Welding materials 2 steel plate 2a Upper board 2b Lower plate 2c Electrode contact surface 2d HAZ (hard) 2e HAZ (intermediate) 2f HAZ (soft) 3. Nuggets 3a Nugget edge 4. Coronabonds 4a Strain concentration point 5 Crack 5a Crack tip 5b Crack opening 6 Seat separation 7 Stamp marks 7a Indentation 8 Electrode shoulder 8a Electrode shoulder width

Claims

1. A method for predicting a fracture mode of a welded member formed by resistance welding two or more overlapping steel plates when a load is applied to the welded member, comprising: Observing the welded component to measure the shapes of the steel plate, nugget, and corona bond, as well as the positions and shapes of one or more cracks generated in the steel plate; constructing an analytical model of the welded component, including the steel plate, the nugget, the corona bond, and the crack, on a computer based on the measurements; setting a critical stress intensity factor and a critical equivalent plastic strain of the welded member; When the magnitude of the load applied to the welded member is set as a common independent variable by finite element analysis using the analytical model, a first function having a stress intensity factor at the edge of the nugget as a dependent variable; a second function having the equivalent plastic strain in the corona bond as a dependent variable; one or more third functions with the stress intensity factor at the tip of the crack as a dependent variable; and One or more fourth functions with the equivalent plastic strain at the tip of the crack as a dependent variable Seeking a first limit load when the value of the first function reaches the limit stress intensity factor; a second limit load when the value of the second function reaches the limit equivalent plastic strain; one or more third limit loads when the value of the third function reaches the limit stress intensity factor; and One or more fourth limit loads when the value of the fourth function reaches the limit equivalent plastic strain. Determine which of the following is smallest: When the first limit load is smallest, brittle fracture occurs at the end of the nugget, When the second limit load is smallest, ductile fracture occurs in the corona bond. When any one or more of the third limit loads is smallest, brittle fracture occurs in the crack corresponding to the third limit load. When one or more of the fourth limit loads are the smallest, ductile fracture occurs at the crack corresponding to the fourth limit load. predicting the fracture mode that will first occur in the welded member; A method for predicting fracture morphology of welded components.

2. At least one of the cracks is located on the electrode contact surface of the steel plate. The method of claim 1.

3. At least one of the cracks is located in the corona bond.

3. The method according to claim 1 or 2.

4. The chemical composition of at least one of the steel plates satisfies, in mass percentage, the following: carbon 0.05% or more and 0.40% or less, phosphorus 0.100% or less, and sulfur 0.100% or less; and the carbon equivalent C calculated by the following formula (1) ｅｑ is 0.40% or more and 0.90% or less, 3. The method according to claim 1 or 2. [Equation 1] Here, [C], [Si], [Mn], [Ni], [Cr], [Mo] and [V] are symbols representing the mass percentages of the components represented by the respective element symbols in the chemical composition of the steel sheet.

5. At least one of the steel sheets is a zinc-based plated steel sheet having a tensile strength of 590 MPa or more.

3. The method according to claim 1 or 2.

Images