2023 Volume 64 Issue 5 Pages 995-1001
To improve the stretch flange deformation limit of ultrahigh-strength steels, a partial edge heating method was investigated. The effect of the strain gradient on the stretch flange deformation limit when a partial edge heating method was applied was investigated by the hole expansion test and FEM analysis. The stretch flange formability was improved by application of a partial edge heating method because of the recovery of the shearing microstructure and the reduction of the hardness difference of microstructure. It was found that the improvement of the deformation limit strain for the strain gradient with edge heating depends on the strain gradient region. In the low strain gradient region, the improvement margin of the deformation limit strain with edge heating was small. On the other hand, in the high strain gradient region, the deformation limit strain was markedly improved compared with that without heating. It was confirmed that the strain gradient in the direction perpendicular to the maximum principal strain could be used for the prediction of the deformation limit and also the determination of the stretch flange fracture by the FEM analysis of actual parts even under edge heating conditions.
This Paper was Originally Published in Japanese in J. JSTP 63 (2022) 59–64.
Effects of strain gradient and edge heating on limit deformation strain.
In recent years, application of ultrahigh-strength steels (UHSS) to automobile parts has been advanced in order to reduce automobile body weight and improve collision safety.1) Among the problems in press forming of UHSS, stretch flange fracture has become apparent as a problem in mass production of automobile parts.2)
Stretch flange fracture is considered to be affected by the surface properties of the sheared edge of the material,3) the work hardening near the sheared edge,4) the hardness difference of the material microstructure,5) etc. It has been reported that stretch flange formability decreases as the roughness and work hardening quantity of the sheared edge surface and the hardness difference of the microstructure increase.
Conventionally, countermeasures for stretch flange fracture have been examined from the viewpoint of shearing technology as one of the methods for improving stretch flange formability. For example, methods such as “cut-off”6) to reduce work hardening near the sheared edge and “fracture surface smoothing”7) to improve the surface property of the sheared edge have been proposed. However, there seem to be issues in the stability of mass production with these methods. In particular, the increased labor and cost of die maintenance when these methods are applied to UHSS are concerns.
Therefore, Machida et al. proposed a method, mainly for mild steel sheets, in which the work hardening of the sheared edge is removed by partially heating the sheared edge.8) This method makes it possible to reduce the labor and cost of die maintenance because the equipment and workpiece do not come into direct contact during heating, and stable stretch flange formability is considered possible even if fluctuations occur in the shearing condition in mass production. Matsuki et al. clarified the mechanism of the effect of sheared edge heating on the stretch flange formability of a 980 MPa material with a dual-phase (DP) structure consisting of ferrite and martensite from the viewpoint of the recovery of the shearing structure and microstructural change, and confirmed that stretch flange formability was greatly improved even if only the region in vicinity of the sheared edge was partially heated with a laser for a short time.9)
Although, when the above-mentioned stretch flange formability improvement technology is applied to mass production, improved accuracy in formability evaluations by the forming simulation technique is also required in order to reduce the labor required in construction and optimization of die conditions. Various studies have examined the effect of the strain gradient on the deformation limit of stretch flanges with the aim of improving the accuracy of simulation. For example, Masuda et al. quantitatively examined the relationship between the impact of the strain gradient on the deformation limit and the plastic deformation characteristics of materials.10) Iizuka et al. quantitatively examined the effect of the strain gradient on the deformation limit of the stretch flange at a punched hole and shear edge and the effect of the material properties on the impact of the strain gradient.11) According to the results of that study, in formability evaluation of the stretch flange deformation part, it is necessary to consider the deformation limit strain of the stretch flange deformation part and the effect of the strain gradient in the direction perpendicular to the maximum principal strain, which has a large effect on the deformation limit strain. However, the relationship between sheared edge heating and the strain gradient was not investigated in the above-mentioned papers.8,9)
Therefore, in this study, the effect of the strain gradient on the deformation limit of the stretch flange when sheared edge heating was applied was examined in a 1180 MPa DP material which consist of ferrite and martensite. By applying them to an actual part, the effect of sheared edge heating on the stretch flange formability of the 1180 MPa material and the prediction of the stretch flange deformation limit by strain gradient were verified.
In this experiment, the punched edge of hole expanding specimens was heated with a laser, and a hole expanding test was carried out after air cooling. To examine the effect of the strain gradient, the hole expansion ratio was examined by changing the punched hole diameter and the punch shape.
2.1 Sample materialsThe mechanical properties of steel used in this study are shown in Table 1. A galvanized (GA) steel sheet (thickness: 1.4 mm) with a tensile strength of 1180 MPa class was used as the test steel, and a JIS No. 5 test piece taken from the direction perpendicular to the rolling direction was used in the tensile test. The hole expansion ratio λ, which is an index of stretch flange formability, was investigated according to the hole expanding test method provided in ISO 16630 (2009). The center of a sheet cut into a shape of 100 mm × 100 mm was punched to a diameter of φ10 mm with a punching clearance of 12.5% of the sheet thickness. The test was conducted 4 times using a conical punch with an apex angle of 60° with the burrs placed on the die side, and the results were summarized by the average value. The hole expansion ratio λ was determined by the following eq. (1).
| \begin{equation} \lambda = \frac{d_{1} - d_{0}}{d_{0}} \times 100 \end{equation} | (1) |
The test procedure for this study is shown in Fig. 1. Laser heating was chosen as the method for heating the sheared edge, as rapid local heating of the edge is possible and the heating temperature can be adjusted by the laser output. After punching the steel sheet, the sheared edge was heated with the laser. The test piece was then air-cooled, and stretch flange formability was evaluated by a hole expansion test in the cold.
Experimental procedure.
Figure 2 shows a schematic diagram of the laser heating process. Using a YAG laser fixed to the robot tip, heat treatment of the sheared edge of the hole expansion test piece was performed by scanning the laser beam irradiation position by the motion of the robot. The laser scanning speed was 10 mm/s, and the spot diameter was φ5 mm. Heating was carried out from the vertical direction of the sheet surface on the burr side of the punched edge with the laser beam offset by 3 mm to the inside of the punched hole, and the laser was moved in the circumferential direction (Fig. 3). Therefore, the heating range was 2 mm from the edge. In this test, the laser power was set at 3 conditions of 300 W, 400 W and 500 W in order to change the maximum achieved temperature during heating, and the effect of the laser power on the deformation limit strain was examined.
Schematic figure of laser heating device.
Edge heating conditions.
In order to examine the effect of the strain gradient on the stretch flange deformation limit in edge heating, the hole expansion ratio was investigated at 3 levels of the punched hole diameter and 2 levels of the hole expanding punch shapes, i.e., a 60° conical punch and a flat bottom punch. Table 2 and Table 3 show the test conditions for the 60° conical hole expansion test and flat bottom hole expansion test respectively. The 60° conical hole expansion test was carried out for 3 punched hole diameters conditions of φ5 mm, φ10 mm and φ25 mm, and the flat bottom hole expansion test was carried out under 2 punched hole diameter conditions of φ10 mm and φ25 mm. With the punching diameter of φ25 mm, a specimen shape of 200 mm × 200 mm was used because the holding margin between the die and the holder was considered insufficient with the 100 mm × 100 mm specimen shape. For the flat bottom hole expansion test, a punch shape in which the crack occurred from the hole edge was selected. The punching clearance in all hole expansion tests was 12.5%, and the lubrication condition was coated with steel sheet rust prevention oil. The burrs were placed on the die side and the test was carried out 4 times, and the results were arranged by the average value.
Figure 4 and Fig. 5 show the effect of the laser power on the hole expansion ratio in the 60° conical hole expansion test and the flat bottom hole expansion test respectively. The 60° conical hole expansion test is a high strain gradient region with a strain gradient Δε of approximately 0.04 mm−1 or more, whereas the flat bottom hole expansion test is a low strain gradient region with about 0.03 mm−1 or less. The strain gradient calculation method is described in the following section 3.2.
Effect of laser power on hole expansion ratio of 60° conical punches.
Effect of laser power on hole expansion ratio of flat-bottomed punches.
In the 60° conical hole expansion test shown in Fig. 4, improvement in λ was observed at the laser power of 300 W, and improvement in λ further progressed at 400 W, reaching a peak value, then decreased at 500 W. This tendency is consistent with the results reported by Matsuki et al.9) In addition, at laser power over 400 W, λ tended to decrease as the punching diameter increased. On the other hand, the effect of heating was not recognized in the flat bottom hole expansion test in Fig. 5.
3.2 Effect of strain gradient on stretch flange deformation limitIn order to investigate the effect of the strain gradient on the deformation limit strain of stretch flanges during edge heating, the strain gradient was calculated from the result of a forming analysis up to the limit shape in the hole expansion test by a Finite Element Method (FEM) analysis. The same shape as in the experiment was modeled for the FEM analysis. Since it has been confirmed that the strain gradient is virtually unaffected by material properties, the FEM analysis of the hole expansion test was carried out using the physical properties of the steel shown in Table 1, and the strain gradient was calculated.11) The FEM analysis was carried out by the dynamic explicit method with shell elements using the finite element software LS-DYNA version 971, and an isotropic hardening model was applied to the material model.
Figure 6 shows the effect of the strain gradient Δε on the deformation limit strain when edge heating is applied. The deformation limit strain was calculated by taking the λ obtained in the hole expansion test as the average strain at the center of sheet thickness. The deformation limit strain increased almost linearly with increasing strain gradient Δε. In the high strain gradient region, the deformation limit strain became larger as the laser power increased in comparison with the case without heating, while the effect of edge heating did not appear in the low strain gradient region.
Effect of strain gradient on limit deformation strain.
From Fig. 6, the deformation limit strain increases in the high strain gradient region when edge heating is performed. As the strain gradient increases, it is considered that the material outside the hole edge does not reach a strain localization condition even if that condition is achieved at the hole edge, and the deformation limit strain increases because the strain localization suppression effect of the remaining work harden ability increases and the necking growth suppression effect in the small strain region increases.12) The reason why the deformation limit strain increases in the high strain gradient region as a result of edge heating is considered to be the recovery of the punched edge and reduction of the hardness difference of the microstructure,9) which is expressed as softening of the hardness in the hardness test. In this section, hardness measurements were conducted under each heating condition, and the effect of the strain gradient on the stretch flange formability improvement effect was examined based on the hardness distribution measured under the laser heating condition of 400 W, in which λ took its peak value in Fig. 4.
The hardness test was performed by the Vickers hardness measurement test using a load of 0.98 N. As the laser was irradiated from the vertical direction of the sheet surface on the burr side, 0.1 mm inside from the surface on the burr side was measured at 0.1 mm from the edge and at intervals of 0.5 mm from 0.5 mm to 4.5 mm, as illustrated in Fig. 7. The measurement results are shown in Fig. 8.
Hardness measurement position.
Hardness distributions on the burr side surface.
Compared with the case without heating, softening occurs in the region up to 1.0 mm from the edge with heating at 400 W. In the high strain gradient region when edge heating is applied, it is considered that λ is improved by the softening that occurs near the edge.
However, in the low strain gradient region shown in Fig. 6, the deformation limit strain largely depends on the state of the sheared edge and the original elongation characteristics of the material. In this experiment, it is considered that λ did not improve because the edge was heated and softened locally, but the elongation of the material in the wider region, which affects the deformation limit in the low strain gradient region, did not change before and after heating. Therefore, in the low strain gradient region, it is suggested that λ can be improved by further widening the heating range from the edge.
4.2 Influence of heating conditionsFrom the results of the 60° conical hole expansion test in Fig. 4, it was found that λ improved with laser heating at 300 W, peaked at 400 W and decreased at 500 W. In this section, the mechanism of the change of λ under each heating condition is discussed based on the hardness distribution of the edge, the change in work hardening and the microstructural change.
First, from the hardness measurement results in Fig. 8, hardness decreased with increasing laser power up to 400 W. The hardness at 500 W increased up to 0.5 mm from the edge, while softening occurred in the region from 1.5 mm to 2.0 mm.
The influence of these heating conditions on hardness is discussed in terms of the change in work hardening of the sheared edge and the microstructural change of the material. First, to confirm the effect of the heating conditions on the work hardening of the sheared edge, the inner side 0.1 mm from the surface of the burr side was observed by SEM at a magnification of 5000×, and the state of work hardening was judged by the metal flow generated at the edge. The results of microstructural observation at the edge are shown in Fig. 9. The image without heating (Fig. 9(a)) shows that large plastic deformation (metal flow in the vertical direction) occurred as a result of punching. With the increase in laser power to Fig. 9(b) and (c), it is considered that recovery of the shearing structure progresses. It can also be inferred from the softening tendency of the edge that recovery has occurred. At Fig. 9(d), the metal flow disappeared and work hardening was removed. However, despite this removal of work hardening, the edge hardness in Fig. 8 increased and λ in Fig. 4 decreased from that at 400 W. In the past, the influence of the work hardening of the sheared edge and the hardness difference of the microstructure have been considered as factors that reduce stretch flange formability, and it is thought that the influence of the hardness difference of the microstructure also appears in this study at 500 W.
Material structure of the edge burr side.
Therefore, microstructural observation was confirmed the effect of the heating conditions on the microstructural change of the edge. As shown in Fig. 10, a position 0.1 mm from the edge was observed at a magnification of 5000× at 0.1 mm inside from the surface on the burr side and 0.1 mm inside from the surface on the rollover side and at the center of sheet thickness. Since it is considered that λ is also affected by the distribution of the microstructure in the thickness direction, observation was carried out in the above-mentioned three regions in the thickness direction.
SEM observation position.
The results of material microstructural observation on the burr side, at the center of thickness and on the rollover side are shown in Fig. 11, Fig. 12 and Fig. 13 respectively. The results of (a) in each figure show that the original structure is a two-phase structure consisting of hard martensite and very soft ferrite. In comparison with this original structure, up to 400 W, tempering of the martensite proceeded in all regions, i.e., the burr side, the center of thickness, and the rollover side. In accordance with this, it can be estimated that the softening of the martensite structure reduced the hardness difference with the ferrite phase.
Material structure inside the burr zone.
Material structure inside the thickness center.
Material structure inside the rollover zone.
However, at 500 W, very hard fresh martensite formed on the burr side, while the tempered martensite in the thickness center and on the rollover side remained substantially unchanged. Since the laser was irradiated from the burr side in this experiment, this means that the heating temperature exceeded the transformation point on the burr side, and it can be estimated that the hardness difference increased due to the two-phase structure of very hard fresh martensite with the very soft ferrite. Thus, distribution of the microstructure occurs in the thickness direction, and at 500 W, the difference in this microstructural distribution in the thickness direction is considered to have contributed to the reduction of λ.
Based on the results presented above, up to 400 W, it is considered that tempering of martensite and softening caused by the recovery of the shearing microstructure improved λ, and it is estimated that the temperature reached 600°C to 700°C with laser heating at 400 W. On the other hand, at 500 W, the large plastic deformation caused by shearing disappeared, but the hardening and increase in the hardness difference of the microstructure due to the fresh martensization of the burr side microstructure and the distribution of the microstructure in the thickness direction are considered to have reduced λ compared to that a 400 W, and it is estimated that the temperature reached 700°C to 800°C. These results are consistent with the results reported by Matsuki et al., which showed the largest λ in the temperature range just below the transformation point even in the 980 MPa material with a DP structure.9)
In this chapter, edge heating was applied to actual parts to verify its effect on stretch flange formability and the practicability of predicting the stretch flange deformation limit by the strain gradient.
5.1 Target part and evaluation methodThe target part was the floor cross shape automobile part shown in Fig. 14. The mechanical properties of the applied material are shown in Table 4. A GA steel sheet (thickness: 1.4 mm) with a tensile strength of 1180 MPa class was used, and a JIS No. 5 test piece was taken from the direction perpendicular to the rolling direction for the tensile test.
Model shape.
In the stretch flange fracture danger area of this part, stretch flange fracture tends to occur as the flange width increases. Therefore, the flange width in the stretch flange fracture danger area was changed, and the flange widths that could be formed under conditions without and with heating were predicted by FEM analysis from the relationship between the deformation limit strain and the strain gradient obtained in Chapter 3. The heating condition was 400 W, at which the peak value of λ was observed in Fig. 4. Based on the predicted results of stretch flange fracture, the blank edge was partially laser heated to verify the validity of the prediction.
5.2 Prediction of stretch flange formability by FEM analysis(1) FEM analysis conditions
The flange widths that can be formed under the conditions with and without heating were predicted by FEM analysis using the radial strain gradient and the maximum principal strain at the edge. The stretch flange fracture prediction was made based on the relationship between the strain gradient and the deformation limit strain without heating and with heating at 400 W obtained in Fig. 6. The flange widths of the stretch flange area were 4 types, i.e., the initial blank shape and flange widths of +3 mm, +6 mm and +9 mm with respect to the initial blank shape, as shown in Fig. 15, and the maximum principal strain and strain gradient of the edge part at each flange width were calculated by FEM analysis. The FEM analysis was carried out by the dynamic explicit method with shell elements using the finite element software LS-DYNA version 971, and an isotropic hardening model was applied to the material model. The same shape as in the experiment was modeled for this FEM analysis. A single process was used, in which form forming with a pad was applied, as shown in Fig. 16. The pad load was 196 kN.
Blank shape.
Tool shape.
(2) FEM analysis results
Figure 17 shows a graph of the strain gradient and the maximum principal strain at the edge obtained from the FEM analysis results plotted against the relationship between the deformation limit strain and strain gradient in Fig. 6. A judgment of fracture was given when the plot was above each limit line, a no-fracture judgment was given when the plot was below each limit line, and a plot on the prediction line was considered to be the fracture prediction. The predicted results showed that the initial blank shape was the limit without heating and the initial blank shape +6 mm was the limit with heating.
Stretch flange fracture prediction results.
(1) Experimental procedures
Flanges of the four kinds of blanks with each flange width, which were larger than the normal size, were prepared by laser blanking, and the blank shapes were then obtained by shearing only the stretch flange evaluation part by partial punching. The clearance was set at 12.5%. Next, two blank patterns without heating and with 400 W heating were prepared for the stretch flange area of each blank shape. As in Chapter 2, the sheared edge was heated with a laser from the burr side face with an output of 400 W and a scanning speed of 10 mm/s, and an area 2 mm wide from the edge was heated with a spot diameter of 5 mm. Finally, a press trial was performed using the prepared blanks. The forming method was the same as that in the FEM analysis, and the pad load was 196 kN.
(2) Experimental results
The results of the press trial are shown in Fig. 18. Without heating, a thickness penetrating crack occurred at +3 mm and a large fracture occurred at +6 mm and above. With heating, fracture did not occur up to +6 mm, and a large fracture occurred at +9 mm, which agreed with the predictions by the strain gradient and deformation limit strain during edge heating. These results confirmed that the effect of edge heating also appears in actual parts, and clarified the fact that the stretch flange deformation limit can be predicted by the strain gradient, even with edge heating.
Experimental results.
The effect of partial heating of the sheared edge on stretch flange formability and prediction of the stretch flange deformation limit by the strain gradient were examined for stretch flange fracture, which is one problem that occurs in press forming of ultrahigh-strength steels. The following conclusions were obtained.
