Materials and Design 30 (2009) 2410–2419

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

The study of spring-back of CK67 steel sheet in V-die and U-die bending processes
M. Bakhshi-Jooybari *, B. Rahmani, V. Daeezadeh, A. Gorji
Faculty of Mechanical Engineering, Babol University of Technology, Babol, P.O. Box 484, Mazandaran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: One of the most sensitive features of sheet metal forming processes is the elastic recovery during unload-
Received 15 August 2008 ing, called spring-back, which leads to some geometric changes in the product. This phenomenon will
Accepted 18 October 2008 affect bend angle and bend curvature, and can be influenced by various factors. In this research, the
Available online 30 October 2008
effects of significant parameters on spring-back in U-die and V-die bending of CK67 anisotropic steel
sheet were studied by experiments and numerical simulations. Comparison between the experimental
Keywords: and the finite element simulation results were also performed.
C. Forming
Ó 2008 Elsevier Ltd. All rights reserved.
F. Plastic behaviour
F. Elastic behaviour

1. Introduction the orientation of anisotropic sheet in U-die bending process. Pape-

leux and Ponthot [9] investigated the effect of blank holder force
In sheet metal forming industry, especially in sheet bending and friction coefficient on spring-back in U-die bending. Thipprak-
process, spring-back has a very significant role. In this process, mas and Rojananan [10] examined the spring-back and spring-go
the dimension precision is a major concern, due to the considerable phenomena on the V-bending process using the finite element
elastic recovery during unloading which leads to spring-back. Also, method (FEM).
under certain conditions, it is possible for the final bend angle to be In this paper, experimental and numerical studies of the effects
smaller than the original angle. Such bend angle is referred to as of significant parameters including sheet thickness, sheet anisot-
spring-go or spring-forward [1]. The amount of spring-back/ ropy and punch tip radius on spring-back/spring-go in V-die and
spring-go is influenced by various process parameters, such as tool U-die bending processes of CK67 (DIN 17222) steel sheet have
shape and dimension, contact friction condition, material proper- been conducted. The results of the experiments were also com-
ties, sheet anisotropy, and sheet thickness [2]. pared with those of the finite element simulations. In addition,
During the past two decades, number of researchers have inves- the results obtained from the two bending processes were also
tigated and attempted to obtain a basic understanding of spring- compared.
back behaviour. Tekiner [3] examined the effect of bending angle
on spring-back of six types of materials with different thicknesses 2. Experimental procedure
in V-die bending. Moon et al. [4] experimentally showed the effect
of combined hot die and cold punch on reduction of spring-back of 2.1. Material characterization
aluminium sheets. Li et al. [5] also showed that the accuracy of
spring-back simulation is directly affected by the material-harden- To characterize the material properties and anisotropy, differ-
ing model. In addition, Cho et al. [6] carried out numerical studies ent specimens of 25  100 mm CK67 steel sheets were cut at dif-
on the effects of some parameters such as punch and die corner ra- ferent orientations to the rolling directions (0°, 45°, and 90°).
dii, punch-die clearance, and coefficient of friction on spring-back Tensile specimens were used to determine the stress–strain
in U-die bending process. Xu et al. [7] investigated the influence curves and the sheet anisotropy parameters, r-values [11]. More-
of some sensitive factors on the accuracy and efficiency of spring- over, rectangular specimens (40  120 mm) cut from the same
back simulation. Gomes et al. [8] simulated material models based sheets were used in the die bending experiments. During the
on various anisotropic models and compared their results with the bending tests, the sheets with thicknesses 0.5, 0.7 and 1 mm,
experimental outcome to show the variation of spring-back with were examined.
A universal Denison Mayes Group (DMG) testing machine
with a capacity of 600 KN was used for the experiments. Fig. 1
shows the stress–strain curve of the sheet obtained from the ten-
* Corresponding author. Tel./fax: +98 111 3234205.
sile test.
E-mail addresses: [email protected], [email protected] (M. Bakhshi-Jooybari).

0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2008.10.018
 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419 2411

Table 2
Data used for simulation.

Material data Quantity

Grade CK67
Yield stress, r (MPa) 265
Young’s modulus, E (GPa) 210
Poisson’s ratio, m 0.3
Friction coefficient, l 0.1
Material density (kg/m3) 6900
Punch tip radius (mm) Variable
V-die tip radius (mm) 1
U-die tip radius (mm) 5
Sheet thickness, t Variable
Punch velocity, v (mm/s) 20
Sheet orientation Variable

Fig. 1. Stress–strain curve of specimen.

Fig. 3. The finite element meshes used.

Table 1 2.2. Sheet anisotropy

R-values for the sheet with different thicknesses.

t (mm) R22 R33 R12 The plastic properties of rolled sheets differ from the through-
0.5 0.9318 0.8871 0.8566 thickness direction, normal anisotropy, and vary with orientation
0.7 0.9731 0.8452 1.0121 in the plane of the sheet, planar anisotropy. At a given angle, h,
1 1.009 0.7384 0.9012 to the rolling direction, the sheet anisotropy is defined by the plas-
tic strain ratio, r-value [12], which is

Fig. 2. Die sets used.

2412 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419

h i
ew ln ww0
rh ¼ ð1Þ
et rh ¼ h i ð3Þ
ln wwl
0 l0
where ew and et are the width and thickness strains of a uniaxial
tension specimen cut at an angle, h, to the rolling direction, respec-
tively. It should be noted that for thin sheets it is difficult to mea-
sure the thickness strain. It is concluded from the constancy of
volume that [12]
ew
rh ¼ ð2Þ
el þ ew
where el ¼ ln ll0 and ew ¼ ln ww0 . Thus:

Fig. 4. Experimental illustration of spring-back and spring-go in V-die and U-die Fig. 5. Simulation illustration of spring-back and spring-go in V-die and U-die
bending. bending.
 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419 2413

Table 4
Spring-back angle in U-die bending (in degree).

t = 0.5 mm t = 0.7 mm t = 1 mm
Orientation Orientation Orientation
0 45 90 0 45 90 0 45 90
Punch tip radius, 2 mm
Experimental 3.23 3.54 3.89
Simulation 1.82 1.56 1.92
Punch tip radius, 4 mm
Experimental 4.8 5.4 6 4.22 4.96 5.55 3.95 4.33 4.64
Simulation 4.04 4.1 4.45 3.86 3.91 4.57 3.24 3.35 3.85
Punch tip radius, 6 mm
Experimental 4.85 5.2 5.52
Simulation 3.86 3.47 4.11

2.3. Die description and performing the experiments

Fig. 2 shows the schematic of the die sets. V-die set contains a
die and a punch with 60° bend angle. Alternative punches with dif-
Fig. 6. A typical simulation result: part 1, the SET location before unloading; part 2, ferent tip radii 2, 3, 4 and 5 mm were used to investigate the effect
location after unloading. of punch tip radius in this die set. U-die set contains a die, a blank
holder and a punch. Alternative punches with various tip radii 2, 4
However, for simulation of anisotropic sheets by using the FE
and 6 mm were also used in this die set. The clearance between the
code, it is necessary to introduce the yield stress ratios, R-values
punch and die, for the three thicknesses of sheet was calculated
[13], as
and applied in U-die set. The die sets were made of steel St-52.
r11 r22 r33 r12 To install the V-die set on the machine, a shoe set with 130 mm
R11 ¼ ; R22 ¼ ; R33 ¼ ; R12 ¼ ; ð4Þ
r r r r guide pillars was used.
where r11 ; r12 ; r22 and r33 are the yield stresses of directions 0°, The tests were performed at a constant velocity. After placing
45°, 90° and thickness orientation, respectively, and r = r11 is the the blank on the die (under the blank holder in U-die bending),
yield stress at orientation 0°. Meanwhile, obtaining the precise yield the upper shoe which was attached to the ram of the machine,
stress ratios is tedious; therefore the plastic strain ratios can be con- moved against the lower shoe. The bending process was divided
verted by the following expressions: into two stages; in the first stage, called loading, the punch moved
down until its stroke reached to a specific value, 40 mm. In the sec-
R11 ¼ 1 ond stage, named unloading (spring-back), the punch moved up.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r90 ðr 0 þ 1Þ Experiments were repeated for various sheet thicknesses, punch
R22 ¼ tip radii and orientations in V-die bending; Thus 36 experiments
r0 ðr 90 þ 1Þ
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi were totally performed. In U-die bending, the effect of punch tip
r90 ðr 0 þ 1Þ ð5Þ radius on spring-back was studied for the sheet thickness 1 mm
R33 ¼ at different orientations. Also, the effect of sheet thickness was
r 0 þ r 90
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi examined for punch tip radius 4 mm at various orientations; thus
3ðr 0 þ 1Þr90 15 tests were totally performed for this die set. Therefore, 51 tests
R12 ¼
2ðr 45 þ 1Þðr0 þ r90 Þ were totally done for the two die sets.
An optical profile projector, Baty R14, was used to measure the
Table 1 lists the R-values acquired for the sheet with different
bend angles.
thicknesses.

Table 3
Spring-back and spring-go angle in V-die bending (in degree).

t = 0.5 mm t = 0.7 mm t = 1 mm
Orientation Orientation Orientation
0 45 90 0 45 90 0 45 90
Punch tip radius, 2 mm
Experimental 6.8 7 7.2 5.5 6 6 2.5 2.5 2
Simulation 6.3 7 7.6 4 4.2 4.8 1.5 1.3 1
Punch tip radius, 3 mm
Experimental 0 0.5 0.8 2.5 3.5 3.5 0 0.2 0.8
Simulation 0 0.1 0.5 3 3.2 3.4 0.8 0.9 1.1
Punch tip radius, 4 mm
Experimental 0.8 1.5 1.5 0 1 1.4 1.3 0.7 1.2
Simulation 1 2 2 0.6 1.1 1.5 2 1 1.3
Punch tip radius, 5 mm
Experimental 2.8 3.2 3.5 1.1 2 2.5 0 1.2 1.7
Simulation 2.5 3 3.5 1.4 1.7 2 0.3 1.4 1.5
2414 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419

3. Finite element simulation Hill’s yield function for anisotropic materials is widely used in
FE modeling due to its relative ease of formulation. In plane stress
3.1. Sheet anisotropy conditions, appropriate for thin shell elements, Hill’s yield criteria
would be defined by the following expression [15]:
The simulation of spring-back was performed by using the FE
h i1=2
code, ABAQUS 6.4. Material properties obtained from the tensile
ðG þ HÞr2x  2Hrx ry þ ðF þ HÞr2y þ 2Nr2xy ¼ r0 ð6Þ
tests, were introduced in the software, and the Hill’s anisotropy
parameters [14] were implemented into the code to model where F, G, H and N are material constants characterizing the cur-
the sheet anisotropy. Table 2 shows the data used in the rent state of anisotropic yield behavior, and r0 is the uniaxial tensile
simulations.

Fig. 7. Effect of sheet thickness for punch tip radius 2 mm; V-die bending. Fig. 8. Effect of sheet thickness for punch tip radius 4 mm; V-die bending.
 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419 2415

yield stress. These parameters are determined from the r-values punch, die and blank holder were assumed as rigid bodies, and
measured at 0°, 45° and 90° to the rolling direction given by the blank was assumed deformable.
r0 1 r0 ðr0 þ r90 Þð1 þ 2r 45 Þ In the simulations, the quadrangle 4 node S4R shell elements
F¼ ; G¼ ; H¼ ; N¼
r 90 ð1 þ r 0 Þ 1 þ r0 1 þ r0 2r 90 ð1 þ r 0 Þ were used for the sheet modeling. These elements were also used
ð7Þ by researchers, for instance Ragai et al. [2] implemented these ele-
ments in bending simulations. This type of element is used in
stress–strain nonlinear problems with large deformation.
3.2. Bending simulation As presented in Fig. 3, in V-die bending simulation, the number
of elements in the model is 80; 5 elements across the width and 16
Due to symmetry of the die sets, only half of the die, punch, elements across the length of the blank. In U-die bending simula-
blank, and blank holder in U-die bending, were modeled. The

Fig. 9. Effect of sheet thickness for punch tip radius 4 mm; U-die bending. Fig. 10. Effect of punch tip radius for orientation 90°; V-die bending.
2416 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419

tion, the number of elements in the model is 40; 2 elements across the blank bend angles at the end of the loading step (die angle) and
the width and 20 across the length of the blank. In the simulations, after unloading.
the blank meshing was ascending; the elements located in the
bending zone are smaller than those farther from it. Fig. 3 shows 4. Results and discussion
the meshes produced on half of the surface of the sheet.
For spring-back measurement, one of the nodes on the blank Fig. 4a and b shows the experimental illustrations of spring-
was selected as a SET, and its position history was extracted during back and spring-go in V-die bending, respectively, which are com-
the two steps of loading and unloading. Then, the spring-back or pared with bent St-12 sheet which has nearly no spring-back.
spring-go could be obtained by calculating the difference between Fig. 4c shows the deformed sheet in U-die bending which has some

Fig. 11. Effect of punch tip radius for thickness 1 mm; V-die bending. Fig. 12. Effect of punch tip radius for thickness 1 mm; U-die bending.
 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419 2417

spring-back. Fig. 5 presents the FE simulation of bending processes 4.1. Effect of sheet thickness
after unloading. Fig. 6 illustrates the results of the simulation, in
which part 1 is the SET location before unloading, and part 2 is Figs. 7 and 8 show the influence of sheet thickness on spring-
its location after unloading. The vertical difference between these back in V-die bending at various sheet orientations for punch tip
two parts on the coordinate system gives spring-back or spring- radii 2 mm and 4 mm, respectively. In the figure (also in the figures
go angle. are shown later), the values on the vertical axis may be positive or
Tables 3 and 4 describe the spring-back or spring-go angles, ob- negative. The negative value of spring-back is called spring-go. As
tained for the different parameters examined. In Table 3, the neg- Figs. 7 and 8 show, by increasing the sheet thickness from 0.5 mm
ative values correspond to spring-go. to 1 mm, the amount of spring-back and spring-go decreases; i.e.

Fig. 13. Effect of sheet anisotropy for punch tip radius 2 mm; V-die bending. Fig. 14. Effect of sheet anisotropy for punch tip radius 4 mm; V-die bending.
2418 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419

the sheet bend angle becomes closer to the die angle. There was an to the diagrams, the experimental and simulation results are in
exception for the thickness 0.7 mm at the punch tip radius 4 mm good agreements. The results obtained in this research are gener-
(shown in Fig. 8a). Fig. 9 shows the effect of the sheet thickness ally in agreement with those reported in the literature [3,6].
on spring-back at various sheet orientations for the punch tip ra- Similar qualitative results were obtained for the other punch tip
dius 4 mm in U-die bending. As it is seen from the figure, by radii 3 mm and 5 mm in V-die bending (data are not shown).
increasing the sheet thickness, the amount of spring-back de-
creases. This is in agreement with the results obtained for V-die 4.2. Effect of punch tip radius
bending.
In sheet metal forming processes, the less the spring-back or Fig. 10 illustrates the changes in the spring-back/spring-go ver-
spring-go, the better be the process parameters selected. According sus punch tip radius for orientation 90° at various sheet thick-
nesses in V-die bending. As the results show in the illustrated
figures, for any of the sheet thicknesses, at a certain value of the
punch tip radius there is no spring-back/spring-go. The spring-go
phenomenon occurs at smaller punch tip radii. By increasing the
punch tip radius, spring-go is converted into spring-back; at punch
tip radius 2 mm there is only spring-go, and at punch tip radii
4 mm and 5 mm there is only spring-back. The experimental and
FEM simulation results showed that the spring-go characteristic
occurred at the punch tip radius less than a certain value, and
the amount of the spring-go decreased as the punch tip radius in-
creased. In addition, the amount of the spring-back increased as
the punch tip radius increased. As the data presented in Fig. 10
and also in Table 3 show, the thickness of the sheet is very vital
to select a suitable punch tip radius for the reduction of the
spring-back/spring-go phenomena. For instance, for thickness of
0.5 mm, R = 3 mm is a suitable radius. Figs. 11 and 12 show the ef-
fect of the punch tip radius on the spring-back/spring-go amount
in V-die and U-die bending at similar conditions. As it is shown
in the figures, the trends of the experimental and simulation re-
sults are similar. The obtained results are generally in agreement
with the findings and the results reported in the literatures
[6,10]. Similar qualitative results are obtained for the other thick-
nesses 0.5 mm and 0.7 mm (data are not shown).

4.3. Effect of sheet anisotropy

Fig. 13 shows the effect of sheet anisotropy on the spring-back

at various sheet thicknesses for punch tip radius 2 mm in V-die
bending. As it is shown in the figure, in general, increasing the
bending direction to the rolling direction resulted in an increase
in the spring-go. Figs. 14 and 15 show the effect of this parameter
on spring-back for both the die shapes in similar conditions.
Accordingly, it was concluded that the bending of the sheet at
orientation 0° was a suitable condition for spring-back or spring-
go reduction in V- and U-die bending processes. These findings
are in agreement with the study conducted by other researchers
[2,8].
Similar qualitative results were obtained for the other punch tip
radii 3 mm and 5 mm in V-die bending (data are not shown).

5. Conclusion

The following results were obtained from the present research:

(1) In V-die bending, spring-go appeared in the sheet in some

cases, but in U-die bending no spring-go appeared.
(2) In V-die and U-die bending processes, increasing the sheet
thickness resulted in a decrease in the spring-back and
spring-go angle.
(3) In V-die bending, in certain process conditions, there was a
significant punch tip radius at which there is no spring-back.
That was an optimum punch tip radius. The amount of the
spring-go decreased as the punch tip radius increased, and
Fig. 15. Effect of sheet anisotropy for punch tip radius 4 mm; U-die bending. the amount of the spring-back increased as the punch tip
 M. Bakhshi-Jooybari et al. / Materials and Design 30 (2009) 2410–2419 2419

radius increased. In U-die bending there is no spring-go, and [3] Tekiner Z. An experimental study of the examination of spring-back of sheet
metals with several thicknesses and properties in bending dies. J Mater Process
the amount of spring-back increased as the punch tip radius
Technol 2004;145:109–17.
increased. [4] Moon YH, Kang SS, Cho JR, Kim TG. Effect of tool temperature on the reduction of
(4) The bending angle to the rolling direction will influence the the spring-back of aluminum sheets. J Mater Process Technol 2003;132:365–8.
spring-back and the spring-go; the greater the angle to the [5] Li X, Yang Y, Wang Y, Bao J, Li S. Effect of the material-hardening mode on the
spring-back simulation accuracy of V-free bending. J Mater Process Technol
rolling direction, the bigger the spring-back or the spring- 2002;123:209–11.
go angle would be. Therefore, the orientation 0° to the roll- [6] Cho JR, Moon SJ, Moon YH, Kang SS. Finite element investigation on spring-
ing direction provides a suitable condition in V-die and U- back characteristics in sheet metal U-die bending process. J Mater Process
Technol 2003;141:109–16.
die bending processes. [7] Xu WL, Ma CH, Li CH, Feng WJ. Sensitive factors in spring-back simulation for
sheet metal forming. J Mater Process Technol 2004;151:217–22.
[8] Gomes C, Onipede O, Lovell M. Investigation of spring-back in high strength
anisotropic steels. J Mater Process Technol 2005;159:91–8.
Acknowledgements [9] Papeleux L, Ponthot JP. Finite element simulation of spring-back in sheet metal
forming. J Mater Process Technol 2002;125–126:785–91.
The authors would like to thank Mr. J. Sadeghi for his technical [10] Thipprakmas S, Rojananan S. Investigation of spring-go phenomenon using
finite element method. J Mater Des 2008;29:1526–32.
support. [11] ASTM. Die design handbook. 3rd ed. Mc Graw-Hill; 1965.
[12] Marciniak Z, Duncan JL, Hu SJ. Mechanics of sheet metal
forming. Oxford: Butterworth Heinemann; 2002.
References [13] ABAQUS 6.4-PR11, Documentation, User manual.
[14] Hill R. The mathematical theory of plasticity. Oxford: Carlendon; 1950.
[15] Hill R. A theory of the yielding and plastic flow of anisotropic metals. Proc Roy
[1] Kalpakjian S, Schmid SR. Manufacturing engineering and technology. 4th
Soc London 1948:281–97.
ed. New Jersey: Prentice-Hall; 2001.
[2] Ragai I, Lazim D, Nemes A. Anisotropy and spring-back in draw-bending of
stainless steel 410: experimental and numerical study. J Mater Process Technol
2005;166:116–27.