JPH11191098A - Method for predicting molding defect in spinning machining - Google Patents

Abstract

PROBLEM TO BE SOLVED: To develop a spinning-machined product in a short development period at small development cost by calculating and predicting molding defects in spinning molding. SOLUTION: Meshes for FEM analysis representing a base material shape and a metal mold shape are prepared (11) and used, molding conditions are set, and drawing simulation of spinning machining by the FEM analysis by a dynamic explicit solution method is performed (13); and hydrostatic stress, equivalent stress, and equivalent strain for every unit time in a molding process are found (14), the value I of a decision parameter is found from the found hydrostatic stress, equivalent stress, and equivalent strain for every unit time by using an Oyane's theoretical crack decision expression (15), and on the basis of the found theoretical crack decision parameter value I, the occurrence of a molding defect is predicted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for predicting the occurrence of molding defects in spinning.

[0002]

2. Description of the Related Art Spinning is a process in which a disk-shaped material is mounted on a mandrel at the tip of a main shaft of a spinning machine and rotated by the main shaft, while the surface of the material rotates around an axis parallel to the main shaft. By pressing the mold and moving the mold forward in the direction in which the axis of the main shaft extends, the material is subjected to drawing and ironing along the mandrel shape, for example, a clutch housing of an automatic transmission for a vehicle. In the spinning process, there is a case where a molding defect such as a crack or a wrinkle is generated in the material during the drawing process.

[0003]

Therefore, as a countermeasure against the above-mentioned molding failure, molding conditions such as a material diameter, a material plate thickness, a spindle rotation speed, a mold advance speed, and a mold sectional shape are usually changed by trial and error. However, such a trial and error method has a problem that the development period is long and the development cost is increased. However, there has hitherto not been any method capable of accurately predicting a molding defect in a drawing process in spinning.

[0004] That is, as a method of predicting a crack in a material by analysis, a finite element method (FEM) of sheet forming has conventionally been used.
A prediction method based on the analysis described in (1) is known. In this method, cracks are predicted from the reduction rate of the sheet thickness. However,
Since there is no feature that the cracking phenomenon of the spinning molded product occurs in accordance with the thickness distribution, the above-described method of judging from the thickness reduction rate cannot be applied. On the other hand, in the forging and deep-drawing processes, which are the fields of plastic working other than spinning, there are many research reports that use "Oyane's formula" as a theoretical crack judgment formula to predict material cracking by forming analysis with FEM. It is also known that prediction accuracy is high. The formula of Oyane is as follows [Equation 1].

(Equation 1)

[0005] The basic concept of the Oyane decision formula is as follows.
Under positive hydrostatic stress, cracks (cracks) occur when voids in the deforming material grow to a certain size or more.
The case where the value of the theoretical crack determination parameter I exceeds 1 corresponds to a state where the crack occurs.

However, until now, this determination formula has a static assumption, such as when machining is performed in a region where the material is not significantly deformed, or when the equivalent stress and equivalent strain do not greatly vary with time in the machining process. It is applied only when it holds (Journal of the Japan Society of Mechanical Engineers 75-639 (1972), 596
Of Moriya Oyane and "Plasticity and Processing", 36-416 (1
995), 985), and in that case, the static implicit method (a kind of solution of the equation of motion in the finite element method) is used at time t + Δt to satisfy the equilibrium at time t + Δt. A technique of performing a solution operation of a large-scale simultaneous linear equation of a motion equation) has been applied. On the other hand, spinning is a processing method in which the material rotates at a high speed of, for example, about 300 rpm, the mold and the material repeatedly contact and separate violently in a short time, and the material is largely deformed. However, the prediction of cracking by the above-mentioned determination formula in such processing methods having completely different deformation behaviors has not been performed so far.

Similarly, as a theoretical crack judgment formula which has been applied only when a static assumption is satisfied, the formula of Cockcroft et al. Shown in the following [Equation 2] and the Brozzoo shown in [Equation 3] are shown below. ), And the formula of Clift et al. Shown in [Equation 4] are known, but the prediction of cracking in spinning by these formulas has not been performed so far. Cockcroft's formula is

(Equation 2) Where I is a theoretical crack determination parameter, σmax
Is the maximum normal stress, ε is the considerable strain, and C is a constant. The equation of Brozzo is

(Equation 3) Where I is the theoretical crack determination parameter, σh is the hydrostatic stress, σmax is the maximum normal stress, ε is the equivalent strain, C
Is a constant. Clift's formula is

(Equation 4) Where I is a theoretical crack determination parameter, σ is equivalent stress, ε is equivalent strain, and C is a constant.

[0008]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method for predicting a molding defect that solves the problems of the prior art in spinning in view of the theoretical crack determination formula. According to the method for predicting the occurrence of molding defects in spinning according to the present invention, a mesh for FEM analysis representing a base material shape and a die shape is created, and the meshes for FEM analysis are used and the molding conditions are set to obtain a dynamic explicit method. Simulation of draw forming of spinning by FEM analysis, and the stress and strain per unit time during the forming process are obtained from the analysis results, and the theoretical crack judgment formula is used from the obtained stress and strain per unit time. To determine the value of the theoretical crack determination parameter, and predict the occurrence of molding defects based on the calculated value of the theoretical crack determination parameter It is an feature.

That is, the inventor of the present application pays attention to the above-mentioned theoretical crack judgment formula which has been applied only when the static assumption is satisfied, while the analysis of spinning processing is performed by a dynamic explicit method (a motion equation in the finite element method). Is a method of solving the equation of motion without performing the solution operation of the simultaneous equations, and solving the equation of motion as it is, and approximately obtaining the solution of the equation of motion at time t + Δt based on the equation of motion at time t).
Having arrived at what can be performed by a molding simulation based on EM analysis, the material is quantitatively evaluated for the damage of the material due to the impact or load received from the mold for each element by the theoretical crack determination parameter value obtained by the theoretical crack determination formula. In other words, a mesh for FEM analysis that represents the shape of the base material and the shape of the mold was created, and the mesh for the FEM analysis was used. The forming conditions were set, and the drawing forming simulation of spinning by FEM analysis using the dynamic explicit method was performed. Performed, the stress and strain per unit time during the molding process are determined from the analysis result, and the value of the theoretical crack determination parameter is determined using the theoretical crack determination formula from the obtained stress and strain per unit time, The occurrence of molding failure was predicted based on the obtained values of the theoretical crack determination parameters. Verification confirmed that the prediction accuracy was actually high.

Therefore, according to the method of predicting the occurrence of molding defects according to the present invention, the occurrence of molding defects (cracks and wrinkles) can be predicted with high accuracy in a short time in a desk study, and the molding conditions (material diameter) , Material thickness, spindle speed, mold shape, number of molds, etc.) can be determined in advance without trial-and-error moldability tests. Therefore, unnecessary moldability tests can be omitted, and good quality can be obtained. Spinning products can be developed with shorter development time and lower development costs than before.

In the present invention, when performing the drawing forming simulation of the spinning process by the FEM analysis in the dynamic explicit method, the speed boundary condition is slid to a speed range approximately 20 to 30 times as high as the actual forming speed range. The drawing forming simulation may be performed in such a manner. When the forming speed range is increased in this way, the time required for the analysis calculation is 1/20 to 1/30 as compared with the case of solving in the actual speed range.
(Product name) CRAY
A molding simulation can be performed within one hour by using a so-called supercomputer such as T90. By the way, if the molding speed range is not increased, the calculation for the molding simulation will take several hours to several tens of hours even with the above-mentioned supercomputer.

Moreover, as a result of verification by the present inventor, if the speed boundary condition is increased to 40 times the actual molding speed, the undeterminable region in which good and bad judgment results are mixed becomes too wide, and the prediction accuracy is too low. That is, if the molding speed is up to about 30 times the actual molding speed, even if the molding simulation is performed by sliding the speed boundary condition to a high speed region, the prediction accuracy is excessively reduced because the undetermined region is not so wide. It turned out that there was no.

Further, in the present invention, when calculating the theoretical crack determination parameter value, a prediction result based on the previously determined value of the theoretical crack determination parameter is compared with an experimental result, and based on the result, the theoretical crack determination parameter is determined. The unit time for obtaining the time function related to the stress in the judgment formula is subdivided into a plurality of units, the time function is obtained for each of the subdivided times, and the maximum value, minimum value or average value of the obtained time functions is calculated as the unit time. It may be applied to the theoretical crack determination formula as a time function for each time, in this case, the theoretical crack determination parameter value can be fine-tuned according to the difference of the material, and the calculation accuracy can be further improved. it can.

Further, in the present invention, when creating an FEM analysis mesh representing the base material shape, the material shape is represented by two or three layers, more preferably a two-layer solid element, in the thickness direction. It may be good,
Further, the material shape may be radially divided into 7 to 9 parts, more preferably 9 parts, and the innermost element may be set to a range constrained by a mandrel and represented by a solid rigid element, and The material shape may be roughly divided into 60 in the circumferential direction.

As a result of verification by the inventor of the present application, when the material shape is represented by one layer of solid elements, the calculation result is easily diverged in FEM analysis. When the material shape is four or more layers, the calculation time is too long, and the material shape is changed in the radial direction. If it is divided into 6 or less, the calculation result tends to diverge. If it is divided into 10 or more, it takes two digits to specify the elements, it takes too much calculation time, and if you divide the material shape into about 60 ± 10 in the circumferential direction, It has been found that high-precision analysis can be performed without too long calculation time. Therefore, if the material shape is divided into elements as described above, it is possible to shorten the calculation time while maintaining high analysis accuracy.

[0016]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a process chart showing an embodiment of a method for predicting the occurrence of forming defects in spinning according to the present invention. In the method of this embodiment, first, in step 11, the finite element method (FE) is used.
A universal file of a mesh (MESH) used for the analysis in M), that is, a mesh-like basic shape model (representing a material shape, a mold shape, and a mandrel shape) divided into a number of square elements is created.

FIG. 2 is an explanatory view showing a method of dividing an element of a basic shape model in the above embodiment. In this embodiment, as shown, a disk-shaped spinning process having a hole at the center is shown. Material 1 is represented by a solid element, the thickness direction of material 1 has a two-layer structure, and the radial direction of material 1 is divided into nine parts.
The circumferential direction of the material 1 is roughly divided into 60 at a pitch of about 6 °, and the innermost element of the material 1 constrained by the mandrel 2 of the spinning machine is a solid rigid element because it is a part that is not related to deformation. And the roller-shaped mold 3 used for spinning is represented by a rigid element of the shell.

Next, here, in step 12 of FIG. 1, the molding conditions (the number of revolutions of the main shaft of the spinning machine, the advance speed of the mold 3 in the direction in which the main shaft extends, the material data of the raw material 1, etc.) are input. Then, FEM data for executing the dynamic explicit method is created. In addition, when the material data of the input material 1 is obtained, in this embodiment, since the material constants a and b in the theoretical crack determination formula (1) are also used, the stress of the material is calculated as follows.
To formulate the strain relationship into the plastic deformation equation σ = Fε ⁿ ,
F-value, n-value (work hardening index) and r-value (Rankford value) were measured by performing a tensile test of the material 1 based on the IS standard, and two types of a uniaxial tensile test and a plane strain test were performed. performing material tests measuring the fracture strain epsilon ₁ in each case. [Table 1] shows SP221 and SCR4
The measurement example of the above-mentioned value about three kinds of materials of 20 and S35C is shown.

[0019]

[Table 1]

In this embodiment, the measurement results of the above-described tensile test methods are substituted into the above-mentioned equation (1) of the theoretical crack determination of Oyane. To solve the simultaneous equations to obtain the material constants a and b. [Table 2] shows SP2
The calculation results of the above-mentioned material constants a and b based on the measurement results of [Table 1] for three kinds of materials 21, SCR420 and S35C are shown.

[0021]

[Table 2]

Next, here, in step 13 of FIG. 1, the drawing forming simulation of the spinning process by the FEM analysis by the dynamic explicit method is performed based on the mesh for FEM analysis and the forming conditions by using the super computer as described above. I do. 3 and 4 show the material SP2.
21, material diameter 240mm, material thickness 6.0mm, drawing ratio 7
0%, spindle speed 366rpm, die advance speed 300mm
3 shows the results of the draw forming simulation at / min. FIG. 3 shows the results of the draw forming simulation when three molds 3 were used, and FIG. 4 shows the results of the draw forming simulation when one mold 3 was used. Is shown. In each figure, the left side shows a state before molding and the right side shows a state after molding, and reference numeral 4 denotes a molded product.

Next, in step 14 of FIG. 1, among the results of the FEM analysis by the dynamic explicit method, there are three types shown in [Table 3] for each time increment (unit time) Δt during the molding process. Obtains data and outputs it to a file.

[0024]

[Table 3]

Next, here, in step 15 of FIG. 1, the three types of data shown in Table 3 above are read from the file, respectively, and based on these data, (2) to (4) of the following [Equation 5] are obtained. ), The time functions y ₁ , y ₂ , Δy ₂ are determined, the accumulated time t _n is determined by the equation (5), and from these and the values of the material constants a and b determined earlier,
The I value (theoretical crack determination parameter value) is numerically calculated by the equation (6) which is a modification of the Ohane theoretical crack determination equation (1). However, the case of y ₁ <0 is not included in the I value calculation.

(Equation 5)

Examples of changes in the time functions y ₁ and y ₂ obtained by the above equations (2) and (3) are shown in the relationship diagrams of FIGS. 5 and 6, respectively. Here, considerable time function y ₂ regarding distortion increases relatively stable, as shown in FIG. 6, but the time function y ₁ about the normal stress and the equivalent stress is tend to vary greatly as shown in FIG. 5 is there. Using this tendency, the method of this embodiment performs a tuning measure for improving prediction accuracy as described later.

Finally, in step 16 in FIG. 1, the formability is evaluated based on the I value obtained by the above equation (6). That is, when I ≧ 1, cracks or wrinkles are theoretically generated. FIG. 7 is a relational diagram showing the distribution of the I value in the radial direction in the molded product, obtained by the above-described calculation, and a table explaining the analysis results. In this analysis result, three molds 3 are used. In this case, the region where I ≧ 1 (crack) exists outside the molded product, but when one mold 3 is used, the region where I ≧ 1 (crack) does not exist. did.

FIG. 8A is a perspective view showing the whole drawn product as a result of actually performing spinning molding using three dies under the same conditions as the calculation of the occurrence of molding failure shown in FIG. FIG. 8B is an enlarged perspective view showing the outer peripheral portion of the drawn product shown in FIG. 8A, and FIG.
FIG. 9 is a perspective view showing the whole drawn molded product as a result of actually performing spinning molding with one mold under the same conditions as the molding defect occurrence prediction calculation shown in FIG. 7, and FIG. It is a perspective view which expands and shows the outer peripheral part of the draw-formed product shown to Fig.9 (a).

As shown in FIGS. 8 and 9 and the table on the right side of FIG. 7 which summarizes the actual molding results, when three dies 3 are used, as shown in FIG. Although a crack occurred in the outer peripheral portion of the molded product, when one mold 3 was used, no crack occurred in the molded product as shown in FIG. 9 (b). Supported.

Also, wrinkles may occur in the material in the process of drawing in spinning. In such a case, the above-mentioned I value is 1 or more. As shown in 10, the occurrence of wrinkles can be clearly predicted. In FIG. 10, the upper side is a state before molding and the lower side is a state after molding, and reference numeral 4 indicates a molded product. FIG. 11 is a perspective view showing the entire drawn molded product as a result of actually performing spinning molding under the same conditions as the molding simulation shown in FIG. 10, and as is clear from FIG.
The above-described molding simulation and the result of the prediction of the occurrence of molding failure based on the simulation are in good agreement with the actual wrinkle generation phenomenon.

By the way, if the above-described analysis is performed in the actual mold forward speed range, the calculation time can be from several hours to several tens of hours even with a supercomputer, as described above. If a long calculation time is required, it is practically inconvenient to use it as a prediction method. Therefore, the applicant of the present application proposes that the F
In performing the drawing forming simulation of the spinning process by the EM analysis, as shown in Table 4, the speed boundary condition of the mold advance speed Vz where the actual speed range is about 300 mm / min is set to 20 times the actual forming speed range. The draw forming simulation was performed by sliding the slide into three different speed ranges, ie, a double speed (speed boundary condition A), 30 times (speed boundary condition B) and 40 times (speed boundary condition C).

[0032]

[Table 4]

If the molding speed range is increased as described above, 2
Even when the speed is increased about 0 to 30 times, the time required for the analysis calculation can be reduced to about 1/20 to 1/30 as compared with the case of solving in the actual speed range. The molding simulation can be performed within the time.

FIGS. 12, 13 and 14 show [Table 5]
The velocity boundary condition A for the 15 types of molding conditions shown in
(20 times), speed boundary condition B (30 times), and speed boundary condition C (40 times). The results of the calculation of the I value based on the draw forming simulation and the actual forming were performed under three types of speed boundary conditions. Here, the setting range of the molding conditions is as follows: material material SP221, material diameter 220 to 27
0mm, material thickness 3.2-6.0mm, drawing ratio 70-80
%, Spindle rotation speed Vθ264 to 486 rpm, and mold forward speed Vz 240 to 360 mm / min. The numbers given below each molding result in the figure indicate the condition No. in Table 5.

[0035]

[Table 5]

The value of A shown in FIGS. 12 to 14 is
This is for quantitatively evaluating the prediction accuracy of molding failure, and indicates the width of a non-determinable region in which the determination result of molding OK and the determination result of molding NG (defective) are mixed. That is, A = IMax.−IMin., And as shown in the symbol table on the right side in FIGS. 12 to 14, IMax. Is the maximum value of I that is OK, and Imin. Is the minimum value of I that is NG. It is.

Under the speed boundary condition A (20 times) shown in FIG. 12, A = 0.10, and the I value error is within ± 5%. Therefore, in this case, it can be said that the prediction accuracy is sufficiently high. Also, the speed boundary condition B shown in FIG. 13 (30 times)
In this case, A = 0.13, and the I value error is + 0% to −13.
%It turns out that. Therefore, also in this case, it can be said that the prediction accuracy is high. However, under the velocity boundary condition C (40 times) shown in FIG. 14, A = 0.32, and the I value error is −
4% to -36%. Therefore, in this case, although the calculation time can be greatly reduced, the prediction accuracy is generally not high. Therefore, in the method of the above embodiment, analysis is performed in a speed range of about 20 to 30 times the actual speed range.

As a feature of the dynamic explicit method, the time function y ₁ relating to the hydrostatic stress and the equivalent stress is compared in a fine time range Δt as shown in FIG. 5 and FIG. Tend to fluctuate significantly.
Using this tendency, in the method of the above-described embodiment, the following tuning is performed in the calculation of the I value in order to improve the prediction accuracy when a prediction error occurs due to a difference in the material of the material.

That is, here, Δ in the calculation of the I value
t is further subdivided, and the value of the time function y ₁ is calculated for each of the subdivided Δt _1. If the I value is lower than the experimental result, the above Δt ₁ is advanced K times. The maximum value of the time function y ₁ during (Δt ₁ elapses K times) is used for the calculation of the expression (1), while the I value is higher than the experimental result. Is the Δ
The minimum value or the average value of the values of the time function y ₁ during the advance of t ₁ K times (when Δt ₁ has elapsed K times) is used for the calculation of the equation (1). Here, K is an arbitrary integer. For example, in FIG. 15, K = 4. By such change in accordance with time function y ₁ experimental results, it is possible to the I value is positioned in the vicinity of 1.0 to cracking occurs, improving the prediction accuracy. Note that K
The calculation of the I value can also be made appropriate by appropriately setting and Δt ₁ .

In order to improve the prediction accuracy, in the method of the above embodiment, as a method of processing the distribution of the I value in the circumferential direction and the radial direction in the material, as shown in FIG. When the distribution of the I value in the circumferential direction varies on the circumference, an average value may be obtained in the circumferential direction, and the average value of the I value may be evaluated. That is, for example, here, I values are obtained at six locations at a pitch of 60 ° in the circumferential direction, and the average value is obtained and evaluated. In the radial direction, element division is 9
Since the innermost circumference is a rigid element as a division, the I values of the eight element groups may be obtained, and the formability may be evaluated using the maximum I value among them.

Thus, according to the method of the above embodiment, the occurrence of molding defects (cracks and wrinkles) can be predicted with high accuracy in a short time in a desk study, and the molding conditions are determined by trial and error. It can be pre-indexed without a performance test, thus eliminating wasteful formability testing and developing a better quality spinning product with a shorter development time and lower development costs.

Further, according to the method of the above embodiment, when performing the drawing forming simulation of the spinning process by the FEM analysis in the dynamic explicit method, the speed boundary condition is set to a speed range of 20 to 30 times as fast as the actual forming speed range. , And the draw forming simulation is performed, so that the time required for the analysis calculation is reduced by half compared to the case of solving in the actual speed range.
The molding simulation can be performed within one hour, which is sufficiently practical if a normal supercomputer can be used, and if it is within the above range, the undetermined area is very wide. Because
The prediction accuracy can be kept high.

Further, according to the method of the above embodiment, the prediction result based on the theoretical crack determination parameter value I obtained above is compared with the experimental result, and based on the result, the hydrostatic stress in the theoretical crack determination formula is calculated. subdividing the unit time for determining the time function y ₁ and to a corresponding stress in a plurality, the calculated the time function y ₁ for each subdivided time Delta] t _1, the maximum value thereof determined time function y _1, the minimum value or Average value per unit time Δ
Since the theoretical crack determination parameter (I) is applied to the theoretical crack determination formula (6) as a time function for each t, the theoretical crack determination parameter value I can be fine-tuned according to the difference between the materials, and the calculation accuracy can be further improved.

Further, according to the method of the above-described embodiment, when creating a mesh for FEM analysis representing the shape of the base material, the shape of the material 1 is represented by a solid element having a two-layer structure in the thickness direction and in the radial direction. It is divided into 9 parts, the innermost element is set in the range constrained by the mandrel and is represented by a solid rigid element, and it is divided into about 60 parts in the circumferential direction, so while maintaining high analysis accuracy, Calculation time can be reduced.

Although the present invention has been described based on the illustrated example, the present invention is not limited to the above-described example.
Instead of the above-mentioned formulas of Cockcroft et al., The formulas of Blozzo et al., And the formulas of Clift et al., It is also possible to use other theoretical crack judgment formulas. Thus, occurrence of molding failure can be predicted. Further, the method of dividing the element of the material shape can be appropriately changed according to the necessity for analysis, etc., other than the dividing method of the above embodiment, and the molding shape is not limited to the cup shape of the above embodiment, but may be appropriately Can be changed.

[Brief description of the drawings]

FIG. 1 is a process chart showing one embodiment of a method for predicting the occurrence of molding defects in spinning according to the present invention.

FIG. 2 is an explanatory diagram showing a method of element division of a basic shape model in the embodiment.

FIG. 3 is an explanatory diagram showing a result of a drawing forming simulation in a case where three dies are used in the embodiment.

FIG. 4 is an explanatory diagram showing a result of a drawing forming simulation when one mold is used in the embodiment.

FIG. 5 is a relationship diagram showing a change state of a time function with respect to hydrostatic stress and equivalent stress obtained from the result of the drawing forming simulation in the embodiment.

FIG. 6 is a relationship diagram showing a change state of a time function with respect to an equivalent distortion obtained from a result of a drawing forming simulation in the embodiment.

FIG. 7 is a relation diagram showing a distribution state of I values in a radial direction in a molded product, obtained by calculation in the above embodiment, and a table for explaining the analysis result.

8 (a) is a perspective view showing the entire drawn molded product as a result of actually performing spinning molding with three dies under the same conditions as the molding defect occurrence prediction calculation shown in FIG. 7;
(B) is an enlarged perspective view showing an outer peripheral portion of the drawn product shown in (a).

9 (a) is a perspective view showing the entire drawn molded product as a result of actually performing spinning molding with one mold under the same conditions as the molding defect occurrence prediction calculation shown in FIG. 7;
(B) is an enlarged perspective view showing an outer peripheral portion of the drawn product shown in (a).

FIG. 10 is an explanatory diagram showing a result of a forming simulation when wrinkles occur in a material in the process of drawing in spinning forming.

11 is a perspective view showing the entire drawn molded product having wrinkles as a result of actually performing spinning molding under the same conditions as the molding simulation shown in FIG.

FIG. 12 is an explanatory diagram showing a result of calculation of an I value based on a drawing forming simulation under a speed boundary condition A (20 times) and actual forming.

FIG. 13 is an explanatory diagram showing a result of calculation of an I value based on a drawing forming simulation and an actual forming under a speed boundary condition B (30 times).

FIG. 14 is an explanatory diagram showing a result of calculation of an I value based on a drawing forming simulation under a speed boundary condition C (40 times) and actual forming.

FIG. 15 is an explanatory diagram illustrating one of tuning methods for improving prediction accuracy.

FIG. 16 is an explanatory diagram illustrating another tuning strategy for improving prediction accuracy.

[Explanation of symbols]

 1 Material 2 Mandrel 3 Mold 4 Molded product

Claims (6)

[Claims] 1. An FEM analysis mesh representing the shape of a base material and a mold is created, and the meshes for the FEM analysis are used and the molding conditions are set to form a draw by spinning by FEM analysis using a dynamic explicit method. A simulation is performed, and the stress and strain per unit time during the forming process are obtained from the analysis results, and the values of the theoretical crack judgment parameters are obtained from the obtained stress and strain per unit time using a theoretical crack judgment formula. A method for predicting the occurrence of a molding defect in spinning, comprising predicting the occurrence of a molding defect based on the value of the theoretical crack determination parameter obtained. 2. A drawing forming simulation by spinning a speed boundary condition to a speed range approximately 30 times or less of an actual forming speed range when performing a drawing forming simulation of spinning processing by FEM analysis in the dynamic explicit method. The method for predicting the occurrence of molding defects in spinning according to claim 1, wherein the method is performed. 3. A method for calculating a theoretical crack determination parameter value, comprising: comparing a prediction result based on the value of the theoretical crack determination parameter obtained earlier with an experimental result; The unit time for calculating the time function is subdivided into a plurality of units, the time function is obtained for each of the subdivided times, and the maximum value, minimum value or average value of the obtained time functions is used as the time function for each unit time. 3. The method according to claim 1, wherein the method is applied to the theoretical crack determination formula. 4. An FE representing the shape of the substrate and the shape of the mold.
The forming defect in the spinning process according to any one of claims 1 to 3, wherein, when creating the mesh for M analysis, the material shape is represented by a solid element having a two-layer or three-layer structure in a thickness direction. Occurrence prediction method. 5. An FE representing the shape of the substrate and the shape of the mold.
When creating a mesh for M analysis, the material shape is divided into 7 to 9 in the radial direction, the innermost element is set in a range constrained by a mandrel, and is represented by a solid rigid element. A method for predicting the occurrence of molding defects in spinning according to any one of claims 1 to 4. 6. An FE representing the shape of the substrate and the shape of the mold.
The method according to any one of claims 1 to 5, wherein when forming the M analysis mesh, the material shape is roughly divided into 60 in the circumferential direction.