J Fail. Anal. and Preven.

(2020) 20:1212–1228
https://doi.org/10.1007/s11668-020-00922-w

TECHNICAL ARTICLE—PEER-REVIEWED

Investigation on the Crack and Thinning Behavior of Aluminum

Alloy 5052 Sheet in Stretch Flanging Process
Surendra Kumar . M. Ahmed . S. K. Panthi

Submitted: 5 October 2019 / in revised form: 28 February 2020 / Published online: 28 July 2020
Ó ASM International 2020

Abstract Stretch flanging is the important sheet metal- Keywords Stretch flanging process 
forming process which is widely used in automobile and Finite element simulation  Edge crack  Crack length 
aerospace sectors. The formability of sheet metal depends Forming load  Strain and thickness distribution 
on various parameters such as material properties, geom- Fractography
etry of the tool setup and process parameters. In the present
work, effects of different die radius and sheet width on
deformation behavior of sheet are studied by FEM simu- Introduction
lation and experiments. The predicted FEM results are
presented in the form of edge crack location and its prop- Flanging is one type of bending process which is used in
agation, crack length, forming load and strain distribution automotive, aerospace and household appliance industries
in sheet along the die profile radius. This study indicates to manufacture the various components. In the stretch
that crack length increases with increase in the sheet width, flanging process, a definite part of the sheet is fixed
while the crack length decreases with increase in the die between the die and the blank-holder and the free part of
radius. It is found that the crack propagation in stretch sheet is bend to 90° to form the concave shaped flange. In
flanging process is affected by the strain distribution in this process, sheet material is stretched circumferentially
sheet and this distribution of strain depends on many along a bend line [1, 2]. The flanged portion is generally
parameters. The crack initiates during deformation of the used to provide the stiffness to a component, smoothens the
sheet at the die corner edge and propagates toward the sharp edges of the part and to assemble the different parts
center of the sheet along the die profile radius. Simulation [3, 4]. In this process, optimization of tensile or hoop stress
results are compared with the experimental one in terms of is a major challenge that is responsible for the satisfactorily
crack length and variation in sheet thickness. FE simulation deformation of the sheet during flange formation. If this
results are found in very good agreement with experimental tensile stress exceeds beyond a certain limit, generation of
results. Fractography study is also presented in terms of cracks takes place in the sheet that is undesirable. Necking,
size, shape of the dimples along with their distribution on localized thinning and localized fracture are the major
the fractured surface. defects in the sheet during flanging operation. These
defects can be minimized with the help of FEM simulation
by designing the process in advance. FEM simulation is a
valuable tool for the prediction of deformation behavior of
S. Kumar
Academy of Scientific and Innovative Research (AcSIR), Bhopal
sheet and load requirements in flanging process within
462026, India efficient time. It has also less equipment cost as compared
to experimental one, which is a cumbersome process to
S. Kumar  M. Ahmed  S. K. Panthi (&) optimize any process [5].
Council of Scientific and Industrial Research (CSIR)-Advanced
Materials and Processes Research Institute (AMPRI), Bhopal
In the past, various researchers have been attempted to
462026, India predict the failure phenomenon in the sheet metal-forming
e-mail: sanjay_panthi@yahoo.co.in

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1213

operation by using FEM simulation. Wang and Wenner [6] sheet in different sheet metal-forming processes such as
used numerical procedure for the study of strain and stress stretching, stretch bending and single-point incremental
distribution in stretch flanging process. This theory is based forming. Dewang et al. [21] studied the effect of punch-die
on the total membrane strain of rigid-plasticity and clearance, coefficient of friction, punch and die profile
approximate theory. Li et al. [7] analyzed the stretch radius on strain distribution for the prediction of crack
flanging of V-shaped sheet metal forming and study the location and propagation of aluminum alloy AA-5052
effect of material parameters and geometrical parameters sheet metal blank. Feng et al. [22] studied the effect of
on flanging. An analytical model was developed for geometrical parameters on the formability of stretch curved
axisymmetric case on total strain theory and membrane flanging by using large-step static implicit FE code and
assumption. Hu et al. [8] found that the smaller die radius proposed that punch curvature radius should be larger than
than the punch radius is detrimental for the deformation. the die radius for the better results. Yohei Abe et al. [23]
Further, this study also presented the effect of geometrical used the gradually contacting punch approach for
parameters and anisotropic characteristics on stretch- improving the stretch flangeability of ultra-high strength
curved flanging by using different sheet metal-forming steel (UHSS) sheet. Inclined bottom punch was used to
processes. Hu et al. [9] developed two analytical model for reduce the tensile stress around the corner edge of the
analysis of introflexion/stretch and outcurve/shrink flanges sheet. Sartkulvanich et al. [24] developed the FEM model
and used them to predict the trim-line of the blank and the for the characterization of blank edge quality for different
shape of the rolling-stock and it is based on the total punch/die clearance of advanced high-strength steel
plasticity theory and the membrane strain assumption. (AHSS) in stretch flanging process. Vafaeesefat Abbas and
Asnafi [10] studied the fracture limit in vertical stretch and Khanahmadlu Morteza [25] compared the FEM simulation
shrink flanging by fluid-forming process and compared the with experimental results in terms of shell-element in
results, experimental results with theoretical predictions. stretch z-flange-forming process. Golovashchenko [26]
Yoshida et al. [11] prevented the crack initiation in high- improved the quality of trimming process in stretch
strength steel sheets by using burring process and devel- flanging of aluminum alloys AA6111-T4 sheet. Wen et al.
oped the suitable die shape and piercing method. The [27] used tapered shoulder tool like a two-step procedure in
method of stretch flangeability was improved by using the flanging process of aluminum alloys AA 6061 and it is
side bending test method and limit diagram. Paul et al. [12] based on single-point incremental sheet-forming (ISF)
analyzed the forming limit diagrams (FLD) on the basis of technique. McDougall et al. [28] studied on the fracture in
strain and stress. Voswinckel et al. [13] used the method the deep drawing of steel sheet. It was revealed from
for improving the geometrical accuracy in flanging process metallographic evaluation that a severe variation with
by incremental sheet metal-forming method and new respect to grain size through the thickness of the steel
adaptive blank-holder technique. Lu et al. [14] developed sheet, as well as a slight segregation of pearlite. Gupta et al.
two models, forward and inverse model by using adaptive [29] carried out a detailed microstructural analysis,
network-based fuzzy inference system (ANFIS) for the mechanical properties evaluation on the samples drawn
prediction of initial circle hole diameter and deformed from the failed deep drawing component. Pantazopoulos
circle hole diameter in the sheet bore-expanding process. and Sampani [30] carried out the Failure Analysis of
Zhang et al. [15] proposed the analytical model for the Fractured Deep-Drawn Aluminum Circles and revealed
prediction of circumferential strain. This model is based on that the tensile strength was leading to fracture during the
uniaxial stress in shrink flanging assumption. Chen et al. process. Wu and Zou [31] studied on deep drawing of a
[16] studied the failure due to wrinkling in shrink flanging coated metal sheet by finite element simulation and
process by using the rubber forming process. It is based on dimensional analysis. A failure map was established based
the effect of four points such as-die radius, flange length, on a few dimensionless process parameters, which can be
die fillet radius and forming pressure in aluminum alloy divided into three regions, i.e., fracture, wrinkling and
AA2024-O, AA7075-O and 2024-T3 sheet metal. Cao success. Yoganjaneyulu and Narayanan [32] carried out
et al. [17] analyzed the onset of the wrinkling condition investigation to study and compare the forming limit dia-
based on the energy method in shrink flanging process. grams (FLDs) and fracture limit curves (FLCs) of titanium
Kasaei et al. [18] developed the mathematical model and grade 2 and titanium grade 4 sheets. Kumbhar [33] dis-
studied the flange wrinkling on the basis of FEM simula- cussed on different failures like wrinkling, fracture, tearing
tion in flexible roll forming process. Wang et al. [19] and earring of the sheet metal in deep drawing process that
proposed the analytical approach (model) for the prediction are generally experienced in industries.
of wrinkling in sheet metal flange-forming operation under In the past few years, a lot of research had been done in
a constant binding force and pressure. Centeno et al. [20] the field of different sheet metal forming processes but less
studied the formability of aluminum alloy AA 2024-T3 focus is given on the curved flanging process which is

123
1214 J Fail. Anal. and Preven. (2020) 20:1212–1228

widely used in automobile industries, aerospace industries

and in the manufacturing of household components. FE
simulation tool like ABAQUS/Explicit software is capable
in designing the tool setup and various process parameters
before the experimental setup for such manufacturing
process, which helps in reducing the manufacturing cost.
Failure likes crack at the edge corner of the flange occurs in
stretch flange-forming process depends on the strain dis-
tribution in sheet during deformation, and this distribution
of strain depends on many geometrical and process
parameters like die radius, flange length, friction, punch
profile, width of sheet and sheet profile. In this study, effect
of the sheet width and the die radius are investigated on the
aluminum alloy AA-5052 sheet metal blank. Effect of Fig. 1 True stress-true strain curve of AA 5052 [34]
others parameters are also studied and has been published
in the other manuscript [34]. The work was carried out by Table 1 Mechanical material properties of AA-5052
ABAQUS/Explicit software to predict the edge crack
Material properties (AA-5052)
location and propagation, thickness distribution in the
flange, forming load and strain distribution. Experiments Mass density (q) 2680 kg/m3
are conducted on double acting hydraulic press to validate Young’s modulus (E) 70.3 GPa
the results of simulation for the same parameters and Poisson’s ratio (t) 0.33
results show a very good relationship. In the passed a very
less attention had been paid by the researchers in the area
of fractography, i.e., the study of fractured surface. Second Table 2 Chemical composition of aluminum alloy AA5052 (wt.%)
part of this investigation is fractography that is the study of [36]
fractured surface of sheet through field emission scanning Others
electron microscopy (FESEM) and results are discussed in Composition Mg Cr Si Fe Mn Cu Zn total Al
terms of size, shape and distribution of dimples on the Actual 2.45 0.17 0.12 0.2 0.08 0.04 0.04 0.06 Rest
cracked surface.

Mechanical Properties of Material

Tensile strength, elastic modulus and Poisson’s ratio of

aluminum alloy AA5052 sheet were obtained by tensile
test on a computerized universal testing machine (UTM) by
following the standard method, i.e., E8 ASTM (Tensile
testing, 2nd edited by J.R. Davis, Davis & Associates) [35].
This test was performed in the rolling direction of the sheet
at a constant strain rate and at the room temperature under
the normal atmospheric conditions. Data collected through
the tensile test were used to prepare the true stress-true
strain curve that is shown in Fig. 1. Tables 1 and 2 rep-
resent the mechanical properties and chemical composition Fig. 2 A CAD model of stretch flange [34]
of aluminum alloy AA 5052 sheet.
perform the stretch flanging simulation and to predict the
crack initiation and propagation in the flange, strain dis-
Finite Element Modeling and Simulation tribution in the sheet along the die profile radius and to
obtain the maximum forming load. In this, die was mod-
Finite element simulation of stretch flanging process was eled with radius of 20, 25, 30 and 35 mm with the height of
carried out by using commercially available ABAQUS/ 60 mm. The shape of the blank-holder was considered
Explicit software. CAD model of a stretch flange is shown similar to the top surface of the die for better distribution of
in Fig. 2. FE model, as shown in Fig. 3, was developed to load and contact among the blank-holder, die and blank. In

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1215

the simulation, die, blank-holder and punch were consid- formed the flange. The blank material was assumed to
ered as rigid bodies and modeled with R3D4 element. behave as an elasto-plastic material with isotropic hard-
R3D4 are rigid 3D four-nodded elements and formed the ening. The constant blank-holding force of 20 kN was
master surfaces during the contact definition conditions. applied on the blank-holder. The coefficient of friction
In the present study, a sheet metal blank of 120 mm in between the contact surfaces was considered as 0.1 and the
length, 30 mm in width and 0.5 mm in thickness was used clearance of 1 mm was provided between the die and the
for the flanging purpose. Width and thickness of the sheet punch [1, 2, 38] in all cases. The parameters were chosen
were kept constant to investigate the effect of different die based on the published literature as mentioned in Table 3.
radius (20, 25, 30 and 35 mm) on the stretch flanging In simulation, the blank was fixed between the die and
process. Die radius and sheet thickness were kept constant the blank-holder in such a way that 90 mm of the blank
to analyze the effect of blank width. Width of the blank portion lies between the die and the blank-holder and
was taken 20, 25, 30, 35 and 40 mm at a constant die radius remaining portion, i.e., 30 mm was kept free to deform and
of 35 mm. Blank was prepared using three-dimensional to get the final shape similar to the die profile. The free
C3D8R-elements, while S4R shell-type elements were portion of the blank that deforms is called the flange length.
used to analyze the thickness distribution in sheet for the The simulation process was carried out for the different
parameters mentioned in Table 3. The C3D8R-element is parameters as mentioned in Table 3, and data were col-
one degree, continuum 3D eight nodded reduction inte- lected in terms of crack location, strain distribution etc.
gration, hexahedra solid element. The blank was The simulated results were compared with the experi-
considered as a deformable body and form the slave sur- mental results for the validation purpose.
faces during contact definition condition. Clearance Crack initiation and propagation were predicted by
between the top surface of the die and the bottom surface of defining the ductile and the shear damage failure criteria
the sheet was kept at 0.1 mm to mitigate the penetration of based on the plastic strain conditions [37, 39–43]. Elements
sheet [37] in the die. In the rigid bodies, all boundary deletion technique was used for the damage evolution
conditions were applied at a reference node that is defined during stretch flanging process. Crack initiation takes place
on them. The die was fixed in all directions to define the when the damage parameter (D) reached to 1 [37].
stationary condition, whereas the movement of the punch
takes place only in the z-direction to deform the blank and
Experimentation of Stretch Flanging Process

The 50 kN (30 kN ? 20 kN) double acting hydraulic press

equipment with a load cell was used to carry out the
experiment of stretch flanging process and results were
used to validate the simulation results. Figure 4 shows the
tool setup of stretch flanging process with the punch, the
die, the blank-holder and the sheet. Different parameters in
experiment were kept similar to FE simulation as men-
tioned in Table 3. Die and punch were made of mild steel
with well-polished surfaces. Blank was placed between the
die and the blank-holder and tightened firmly by bolts. Free
portion and the clamped portion of the sheet were kept
similar to the FE simulation. 1 mm of clearance was
maintained between the die and the punch, and filler gauge
was used to measure the clearance. Experiments were
performed at the room temperature, and the blanks were
Fig. 3 FEM Model for stretch flanging process used without annealing. Lubricant was not used among the

Table 3 Parameters used in FE simulation and experimental work

S. No. Parameter Value considered in FE simulation and experimental

1. Die radius (mm) Die radius of 20, 25, 30 and 35 mm at constant sheet width of 30 mm
2. Sheet width (mm) Sheet width of 20, 25, 30, 35 and 40 mm at constant die radius of
35 mm

123
1216 J Fail. Anal. and Preven. (2020) 20:1212–1228

Fig. 4 Double acting hydraulic

press and tool setup

die, blank, blank-holder and punch. Downward movement

speeds of the punch were fixed at 25 mm/s until the punch
reaches for a certain distance or just makes a contact with
the blank. Punch speed of 1 mm/s was controlled with the
help of a computerized control panel. After completion of
deforming process, forming load was released by defining
the punch movement in upward direction and data were
recorded in the system. The schematic diagram of the
process is shown in Fig. 4. The flange, with the final shape,
was taken out from the tool setup to measure the crack
length and thickness of the sheet.
Fractography of the final product were carried out by
field emission scanning electron microscopy (FESEM) to
depict the fracture modes for different die radius and sheet
width. Different forces like tensile, compressive, bending
and shearing generate during stretch flanging process. Any Fig. 5 Schematic diagram of tool setup for the assessment of bending
force
failure occurs during the forming depends on the type of
forces or combination of these forces and it affects the size  
and the shape distribution of the dimples on fractured Kbf  ðTSÞ  W  t2
Maximum bending force ðF Þ ¼
surface. D
ðEq 1Þ
where Kbf = constant factor which depends on the types of
Analysis of the Bending Force
bending process (Kbf = 0.33 for edge bending and Kbf =
0.7 for U type bending), t = sheet thickness, W = sheet
The maximum bending force can be determined by using
width, TS = Tensile strength of the materials, D = die
the following equation [1, 2]
opening dimension (span length). The schematic diagram
for assessment of bending force of stretch flanging process
is shown in Fig. 5.

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1217

Results and Discussion Effect of Sheet Width

Validation of the Simulation Results Sheet width is an important parameter which affects the
deformation behavior of sheet in stretch flanging process.
Simulation results were compared with the experimental Deformation pattern of the Al material sheet is shown in
results in terms of crack length, edge crack location and its Figs. 6 and 7. Al alloy 5052 sheets of different width were
propagation with respect to punch displacement. used to minimize the edge crack length along the die

Fig. 6 Comparison of crack length for different sheet width (20, 25, 30, 35 and 40 mm) at 35 mm die radius

123
1218 J Fail. Anal. and Preven. (2020) 20:1212–1228

Fig. 7 Comparison of crack

length for different sheet width
(25, 30, 35 and 40 mm) at
35 mm die radius

Fig. 8 Comparison of
reduction in thickness of sheet
for different sheet width (25, 30,
35 and 40 mm) at constant die
radius of 35 mm

profile radius. The final shape of the deformed sheet by and the blank. It also results in increase in bending force
FEM simulation is shown in columns 2 and 3, while the which is directly proportional to the sheet width as given in
experimental results are shown in columns 4 and 5 of Eq 1. It is clear from Eq 1 that large forming load is
Fig. 6. Initiation of crack takes place from both corners required to bend the sheet of high width [1, 2]. Crack
along the die profile radius and propagates toward the length was found minimum for 25 mm sheet width.
center of the sheet in all cases. The crack contours Therefore, crack length decreases in simulation as well as
observed in FE simulation are found similar to the exper- in experiments. However, the flange of proper concave
imental one. The crack length in simulated results was shape without fracture could not be achieved at any width
measured with the help of the query information tools of sheet. For 20 mm sheet width, flange is formed without
available in ABAQUS/Explicit software while in experi- concave shape. In this condition, it behaves like a straight
ments it was measured by using digital Vernier calipers flanging process and considerable spring back is found.
with a least count of 0.01 mm. According to the figures, Hence, formability of flange of concave shape reduces for
crack length in sheet increases with increase in sheet width. the sheet width less than 20 mm.
It indicates that the deformability of the sheet reduces with Figure 8 shows the variation in thickness of sheet for
increase in sheet width. Edge crack length is found maxi- different sheet width at constant die radius of 35 mm.
mum when width of sheet is 40 mm, i.e., maximum. The Reduction in thickness is measured by defining the
reason of increase in crack length with increase in sheet meridian path-2, as shown in Fig. 9, which is considered
width may be due to the higher friction between the tools near to edge crack position along the die profile radius. A

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1219

Fig. 9 Path-1, considered for

circumferential and radial strain
near the crack and Path-2,
considered for measuring the
thickness of sheet

Fig. 10 Contour plots for sheet

thickness distribution for
different sheet width (25, 30, 35
and 40 mm) at constant die
radius of 35 mm by FEM
simulation

graph is plotted between the flange thicknesses with respect Effect of Die Radius
to distance between free edge of the flange to the center of
the deformed blank. The maximum reduction in thickness Die radius is also an important parameter which affects the
is obtained when larger sheet width is used whereas it is edge crack length and edge crack propagation in stretch
found minimum for smaller sheet width. Comparison of flanging process. In the present study, different die radii
simulation and experimental results is shown in Figs. 8 and were used at constant sheet width for studying the defor-
10. mation behavior of sheet and failure of flange in the form

123
1220 J Fail. Anal. and Preven. (2020) 20:1212–1228

Fig. 11 Comparison of crack length for different die radius (20, 25, 30 and 35 mm) at constant sheet width of 30 mm by FEM simulation and
experimental results

Fig. 12 Comparison of crack

length for different die radius
(20, 25, 30 and 35 mm) at
constant sheet width of 30 mm
by FEM simulation and
experimental results

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1221

of crack length and its propagation. The final shapes of the magnification. The procedure of measuring the crack
stretch flange with crack after forming are shown in length in FE simulation and in experiments is identical to
Fig. 11. The crack initiation takes place along the die the procedure adopt in measuring the effect of sheet width
profile radius and propagates toward the sheet center up to section. Figure 12 shows the comparison of maximum
some extent. Columns 2 and 3 in Fig. 11 show the FEM crack length for various die radiuses. It can be easily seen
simulation results, while the columns 4 and 5 represent the from Fig. 12, that maximum edge crack length is found in
experimental results at lower as well as at higher 25 mm die radius, whereas minimum in 35 mm die radius
by simulation as well as by experimentation. Hence, it can
be concluded that crack length increases with decrease in
die radius. It may be due to a smaller amount of stress
concentration in the sheet at higher die radius due to suf-
ficient space is available for deformation of sheet. It might
be due to the reduction in friction force also with increase
in curvature of die radius. Hence, forming load also redu-
ces in these conditions.
Variation of thickness in sheet is also measured for
different die radius at constant sheet width. A graph is
plotted between the flange thickness and the distance
between free edge of the flange to the center of the
deformed blank and it is shown in Fig. 13. It can be seen
from figure that maximum thinning is occurs when smaller
die radius is used as shown in Figs. 13 and 14. These
Fig. 13 Comparison of reduction in thickness of sheet for different results are validated with experimental one and found in
die radius (20, 25, 30 and 35 mm) at constant sheet width (30 mm)
very good agreement.

Fig. 14 Contour plots for sheet

thickness variation for different
die radius (a) 20 mm (b) 25 mm
(c) 30 mm (d) 35 mm at
constant sheet width of 30 mm
by FE simulation

123
1222 J Fail. Anal. and Preven. (2020) 20:1212–1228

Fig. 15 Deformed shape of sheet at different percentage of punch displacement by FEM for different sheet width at constant die radius of
35 mm

Fig. 16 Influence of sheet width on punch load at different sheet (20, Fig. 17 Influence of sheet width on circumferential strain at different
25, 30, 35 and 40 mm) at constant die radius of 35 mm by FE sheet width (20, 25, 30, 35 and 40 mm) at constant die radius of
simulation 35 mm by FE simulation

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1223

FEM Simulation Results Figure 15 shows the deformed shape of the sheet for dif-
ferent width at 15, 30, 45, 60 and 100% of the total punch
Effect of Sheet Width displacement. From Fig. 15, it is observed that the crack
starts probably after 30% of the total punch displacement in
In this study, effect of the sheet width and the die radius on all cases. No crack is observed at any % of punch dis-
deformation behavior of sheet was investigated. The sheet placement in the case of 20 mm sheet width with 35 mm of
width is one of the significant parameter in stretch flanging die radius because this combination behaves like a straight
process that affect the deformation behavior of the sheet. flanging process. Therefore, it is found that sheet width of
20 mm with die radius of 35 mm should not be use in
stretch flange forming. Crack length and deformations are
also observed different for different sheet width at constant
die radius. It can be seen from Fig. 15 that the crack length
increases with increase in sheet width and better stretching
and minimum crack length is observed at 30 to 35 mm
sheet width with 35 mm die radius.
A graph is plotted between the punch load and the punch
displacement for different sheet width is shown in Fig. 16.
Forming load increases with advancement in punch dis-
placement up to a certain distance (5 mm) and reached at
its maximum value. The reason of increased forming load
is due to the resistive capability of the sheet in unbend
Fig. 18 Influence of sheet width on radial strain at different sheet position (elastic deformation limit). When the sheet deform
width (20, 25, 30, 35 and 40 mm) at constant die radius of 35 mm by up to a certain angle then forming load decreases
FE simulation

Fig. 19 Deformed shape of sheet at different percentage of punch displacement by FEM for different die radius at constant sheet width of
30 mm

123
1224 J Fail. Anal. and Preven. (2020) 20:1212–1228

prominently because of the crack propagation that reduces

the resistive capacity of the sheet. After a definite bend
angle, punch slides over the sheet that result in decrease in
the forming load. The sliding of the punch starts after the
5 mm punch displacement, approximately, which result in
decrease in punch load sharply and then it remains constant
for the entire punch displacement. It is also observed that
punch load decreases with decrease in sheet width and it
may be due to the friction between the blank and the tool
interface. Therefore, the higher forming load is required to
bend a blank of larger sheet width (40 mm sheet width).
Results are also found in the line of bending Eq 1 [1, 2],
whereas less forming load is required for 20 mm sheet
Fig. 20 Influence of die radius on punch load at different die radius width. It may be due to the less friction between the tool
(20, 25, 30 and 35 mm) at constant sheet width 30 mm by FE and sheet [1, 2]. The maximum forming load increases
simulation
approximately 76% with increase in sheet width from
20 mm to 40.
Sheet width also influences the circumferential and the
radial strain distribution in stretch flanging operation. The
strain distribution was predicted in the deformed blank by
drawing the meridian path that was selected very near to
edge crack point along the die profile radius as shown in
Fig. 8. A graph is plotted between the circumferential
strain with respect to distance along die profile radius up to
the deformed sheet center for all cases as shown in Fig. 17.
Figure 18 shows the graph of the radial strain with respect
to distance along die profile radius up to the deformed sheet
center for all sheet width. The maximum circumferential
and the radial strain are found in larger sheet width, i.e.,
40 mm because maximum friction between the blank and
Fig. 21 Influence of die radius on circumferential strain at different tool interface takes place that results in more stretching of
die radius (20, 25, 30 and 35 mm) at constant sheet width 30 mm by sheet circumferentially. Whereas, a reverse condition was
FE simulation
observed for the smaller sheet width because it behaves
like a straight flanging deformation with large die radius. It
also reduces the stress concentration in unit area of the
material [1, 2]. The maximum circumferential and the
radial strain occur at edge portion of the flange and
decreases toward the center of the sheet along the die
profile radius as shown in Figs. 17 and 18. These are found
minimum at mid-section of deformed sheet. The circum-
ferential strain, radial strain and edge crack propagation
increases with the increase in sheet width. Increment in
circumferential and radial strain is observed 23 and 48%,
respectively when the sheet width increases from 20 to
40 mm.

Influence of Die Radius

Fig. 22 Influence of die radius on radial strain at different die radius Figure 19 shows the deformed shape of the sheet at dif-
(20, 25, 30 and 35 mm) at constant sheet width 30 mm by FE ferent percentage of total punch displacement, i.e., 15, 30,
simulation 45, 60 and 100% for different die radii at the constant sheet
width. Crack initiated nearly at 30% of the total punch

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1225

Fig. 23 Fractographs of
following die radius (a) die
radius 20 mm (b) die radius
25 mm (c) die radius 30 mm
and (d) die radius 35 mm, at
constant sheet width of 30 mm

displacement in all cases. Deformation of sheet in stretch Figs. 21 and 22 that the maximum radial as well as the
flanging process is also affected by the die radius for the circumferential strain occurs near to the edge of the flange
same size and shape of the blank that can be seen from along to the die profile radius, while it decreases toward the
Fig. 19. The crack length decreases with increase in the die center. Circumferential and radial strain decreases with
radius and deformation of sheet also reduces due to the increase in die radius because of reduction in the stress
large radius of curvature of the die. The maximum crack concentration due to the less curvature of die radius and
length is found for 20 mm die radius, while the minimum sufficient space available for deformation of sheet.
crack length is observed at 35 mm die radius. It can be Decrements in circumferential and radial strain are found
concluded from this that if the sheet width is used nearly to 29.63 and 38.46%, respectively, when increase in the die
30 mm with 25 to 30 mm die radius, it provides better radius from 20 to 35 mm.
deformability of the sheet.
Figure 20 shows punch load v/s punch displacement graph Fractography of AA5052 Alloy Sheets After Stretch
for different die radius. The variation in punch load is found Flanging Process
similar to the sheet width variation. The forming load
decreases with increase in die radius because of less curvature Fractographs of samples has been analyzed from the
of die radius and availability of more suitable space for FESEM fracture micrographs at 4000 magnifications for
deformation of the sheet as shown in Fig. 20. It can be seen each test. Figure 23 shows the fractured surfaces of
from figure that maximum forming load is required to deform AA5052 aluminum alloy sheet after the stretch flange
the sheet when die radius is small or due to large curvature of forming at different die radius 20, 25, 30 and 35 mm
die. On the other hand, forming load requirement is minimum having constant sheet width of 30 mm. As the die radius
for the large die radius (35 mm die radius). The maximum increases from 20 to 35 mm, elongation in dimple size
forming load decreases approximately 25% with increase in decreases due to reduction in stress concentration and
the die radius from 25 to 35 mm. forming load distribution. It is because there is a proper
Figures 21 and 22 represent the graph of circumferential space available for stretching the sheet with the large die
and radial strains for all die radii. It can be seen from radius. It can also be seen in Fig. 23 that the large number

123
1226 J Fail. Anal. and Preven. (2020) 20:1212–1228

Fig. 24 Fractographs of
following sheet width (a) sheet
width 25 mm (b) sheet width
30 mm (c) sheet width 35 mm
and (d) sheet width 40 mm, at
constant die radius of 35 mm

of small dimples are produced due to the fracture of the properties of the material, magnitude of forming load,
samples when large die radius is employed. It might be a strain rate etc. Finally, it can be conclude that the charac-
reason of generation of minimum crack length at minimum teristic of the failure modes are ductile and shear types
forming load. fracture because variation in dimples size, shape and its
Fractographs of different samples have been also studied distribution are found different for different die radius and
for different sheet width such as 25, 30, 35 and 40 mm with sheet width. The micrographs of fractured surface depict
a constant die radius of 35 mm and it is shown in Fig. 24. that there is a ductile plastic deformation and it is clear
In the figure, small and large sized dimples are detected at from the elongated dimples and voids.
minimum sheet width (25 mm). However, small sized
dimples are comparatively more in number than large size
dimples. This happens because minimum friction takes Conclusions
place between the tool and the sheet interface. Hence small
bending force is required to bend the sheet in the case of In this paper, effect of sheet width and die radius in stretch
less sheet width [1, 2]. In this condition, minimum crack flanging process of AA5052 aluminum alloy sheet were
length and forming load were obtained during the defor- successfully studied by FE simulation and validated with
mation of sheet, whereas a reverse condition was observed experimental results. According to the observation, fol-
in the case of higher sheet width. More number of dimples lowing outcome can be summarized:
was found of large size when sheet is of 40 mm width. It
1. Maximum crack length in the flange is obtained by FE
may be due to high friction between the sheet and the tool
simulation in a case when the smaller die radius is
that causes the requirement of high forming load to bend
used. These results are found in very good agreement
the sheet [1, 2].
with the experimental results. Forming load is also
Dimple size and shape, number of dimple density and
found higher in the case when large size die radius is
dimple distribution in fractured surface depend on the
used.
different variables such as tooling conditions, mechanical

123
J Fail. Anal. and Preven. (2020) 20:1212–1228 1227

2. Minimum thinning in sheet is obtained in the following Operations in Sheet-Metal Forming. J. Mater. Proc. Technol.
cases: (1) large radius of die with constant sheet width 53(3), 759–780 (1995). https://doi.org/10.1016/0924-
0136(94)01766-T
(2) smaller sheet width with constant die radius. These 4. F. Stachowicz, Estimation of Hole-Flange Ability for Deep
results are compared with the experimental one and Drawing Steel Sheets. Arch. Civ. Mech. Eng. 8(2), 167–172
found in very good agreement. Thinning in sheet is (2008). https://doi.org/10.1016/S1644-9665(12)60203-9
more affected by the die radius as compared to the 5. D.L. Logan, A First Course in the Finite Element Method
(Cengage Learning, Boston, 2011)
sheet width. Hence, die radius should be used nearly to 6. N.-M. Wang, M. Wenner, An Analytical and Experimental Study
the width of the sheet for better formability in stretch of Stretch Flanging. Int. J. Mech. Sci. (1974). https://doi.org/10.
flanging process. 1016/0020-7403(74)90082-4
3. The minimum circumferential and the radial strain are 7. D. Li, Y. Luo, Y. Peng, P. Hu, The Numerical and Analytical
Study on Stretch Flanging of V-Shaped Sheet Metal. J. Mater.
found in large die radius. Forming load is also required Proc. Technol. 189(1), 262–267 (2007). https://doi.org/10.1016/j.
minimum to form the flange in higher die radius with jmatprotec.2007.01.035
fixed width of sheet. Crack length is also found 8. W. Hu, Z. Wang, Anisotropic Characteristics of Materials and
minimum in large die radius case as compared to lower Basic Selecting Rules with Different Sheet Metal Forming Pro-
cesses. J. Mater. Proc. Technol. 127(3), 374–381 (2007). https://
die radius. Therefore, it can be concluded that large die doi.org/10.1016/S0924-0136(02)00410-7
radius is more preferable than the smaller one for the 9. P. Hu, D. Li, Y. Li, Analytical Models of Stretch and Shrink
fixed width of the sheet. Flanging. Int. J. Mach. Tools Manuf 43(13), 1367–1373 (2003).
4. The circumferential and the radial strain increases with https://doi.org/10.1016/S0890-6955(03)00150-0
10. N. Asnafi, On Stretch and Shrink Flanging of Sheet Aluminium
increase in the sheet width, approximately 23 and 48%, by Fluid Forming. J. Mater. Proc. Technol. 96(1), 198–214
respectively, when the sheet width increases from 20 to (1999). https://doi.org/10.1016/S0924-0136(99)00352-0
40 mm. Maximum forming load requirement is also 11. H. Yoshida, T. Yoshida, K. Sato, Y. Takahashi, T. Matsuno, J.
increased approximately by 76% for the same case. Nitta, Evaluation and Improving Methods of Stretch Flangeabil-
ity. Nippon Steel Tech. Rep. 1035, 18–24 (2013)
5. The circumferential and the radial strain decrease 12. S.K. Paul, Theoretical Analysis of Strain-and Stress-Based
approximately 29.63 and 38.46%, respectively, when Forming Limit Diagrams. J. Strain Anal. Eng. Des. 48(3), 177–
the die radius is increased from 20 to 35 mm and 188 (2013). https://doi.org/10.1177/0309324712468524
maximum forming load also decreased by 25% for the 13. H. Voswinckel, M. Bambach, G. Hirt, Improving Geometrical
Accuracy for Flanging by Incremental Sheet Metal Forming. Int.
same processing condition. J. Mater. Form. 8(3), 391–399 (2015). https://doi.org/10.1007/
6. The crack initiates in sheet at the corner of the die and s12289-014-1182-y
propagates toward the center of the sheet along the die 14. Y.-H. Lu, F.-H. Yeh, C.-L. Li, M.-T. Wu, Study of Using ANFIS
profile radius in all cases. to the Prediction in the Bore-Expanding Process. Int. J. Adv.
Manuf. Technol. 26(5–6), 544–551 (2005). https://doi.org/10.
7. Fracture phenomenon of the samples in the stretch 1007/s00170-003-2024-0
flanging process is different for different die radius and 15. G.E. Zhang, J. Yao, S.J. Hu, X. Wu, Shrink Flanging with Sur-
sheet width. The large number of small dimples and face Contours. J. Manuf. Process. 5(2), 143–153 (2003). https://
more uniform distribution are found in larger die radius doi.org/10.1016/S1526-6125(03)70049-8
16. L. Chen, H. Chen, Q. Wang, Z. Li, Studies on Wrinkling and
and smaller sheet width. Therefore, in stretch flanging Control Method in Rubber Forming Using Aluminium Sheet
process die radius should be approximately equal to the Shrink Flanging Process. Mater. Des. 1980–2015(65), 505–510
width of the sheet for better formability. (2015). https://doi.org/10.1016/j.matdes.2014.09.057
17. J. Cao, M. Li, Wrinkling Analysis in Shrink Flanging. J. Manuf.
Sci. Eng. (2011). https://doi.org/10.1115/1.1381397
Acknowledgments The authors would like to express thanks to the 18. M.M. Kasaei, H.M. Naeini, B. Abbaszadeh, M. Mohammadi, M.
Director, CSIR-AMPRI, Bhopal for the support and providing the Ghodsi, M. Kiuchi, R. Zolghadr, G. Liaghat, R.A. Tafti, M.S.
facilities to execute this work. Tehrani, Flange Wrinkling in Flexible Roll Forming Process.
Proc. Eng. 81, 245–250 (2014). https://doi.org/10.1016/j.proeng.
2014.09.158
Conflict of interest The authors declare that they have no conflict 19. X. Wang, J. Cao, An Analytical Prediction of Flange Wrinkling
of interest in this work. in Sheet Metal Forming. J. Manuf. Process. 2(2), 100–107 (2000).
https://doi.org/10.1016/S1526-6125(00)70017-X
20. G. Centeno, A. Martı́nez-Donaire, C. Vallellano, L. Martı́nez-
Palmeth, D. Morales, C. Suntaxi, F. Garcı́a-Lomas, Experimental
References Study on the Evaluation of Necking And Fracture Strains in Sheet
Metal Forming Processes. Proc. Eng. 63, 650–658 (2013). https://
1. S. Kalpakjian, K. Vijai Sekar, S.R. Schmid, Manufacturing doi.org/10.1016/j.proeng.2013.08.204
Engineering and Technology (Pearson, London, 2014) 21. Y. Dewang, M. Hora, S. Panthi, Prediction of Crack Location and
2. M.P. Groover, Fundamentals of Modern Manufacturing: Mate- Propagation in Stretch Flanging Process of Aluminum Alloy AA-
rials Processes, and Systems (Wiley, New York, 2007) 5052 Sheet Using FEM Simulation. Trans. Nonferrous Met. Soc.
3. W. Chuan-Tao, G. Kinzel, T. Altan, Failure and Wrinkling Cri- China 25(7), 2308–2320 (2015). https://doi.org/10.1016/S1003-
teria and Mathematical Modeling of Shrink and Stretch Flanging 6326(15)63846-8

123
1228 J Fail. Anal. and Preven. (2020) 20:1212–1228

22. X. Feng, L. Zhongqin, L. Shuhui, X. Weili, Study on the Influ- During Single Point Incremental Forming (Anal. Prev., J Fail,
ences of Geometrical Parameters on the Formability of Stretch 2019). https://doi.org/10.1007/s11668-019-00721-y
Curved Flanging by Numerical Simulation. J. Mater. Proc. 33. S.V. Kumbhar, Pressure Optimization and Failure Prediction for
Technol. 145(1), 93–98 (2004). https://doi.org/10.1016/S0924- Deep Drawing Process of Sheet Metal Products: A Case Study. J.
0136(03)00866-5 Fail. Anal. Prev. 18(4), 948–956 (2018). https://doi.org/10.1007/
23. Y. Abe, K.-I. Mori, K. Norita, Gradually Contacting Punch for s11668-018-0485-6
Improving Stretch Flangeability of Ultra-High Strength Steel 34. S. Kumar, M. Ahmed, S.K. Panthi, Effect of Punch Profile on
Sheets. CIRP Ann. Manuf. Technol. 62(1), 263–266 (2013). Deformation Behaviour of AA5052 Sheet in Stretch Flanging
https://doi.org/10.1016/j.cirp.2013.03.059 Process. Arch. Civ. Mech. Eng. 20, 18 (2020). https://doi.org/10.
24. P. Sartkulvanich, B. Kroenauer, R. Golle, A. Konieczny, T. 1007/s43452-020-00016-2
Altan, Finite Element Analysis of the Effect of Blanked Edge 35. J.R. Davis, Tensile Testing (ASM International, Ohio, 2004)
Quality Upon Stretch Flanging of AHSS. CIRP Ann. 59(1), 279– 36. M. Ahmed, D.R. Kumar, M. Nabi, Enhancement of Formability
282 (2010). https://doi.org/10.1016/j.cirp.2010.03.108 of AA5052 Alloy Sheets by Electrohydraulic Forming Process. J.
25. A. Vafaeesefat, M. Khanahmadlu, Comparison of the Numerical Mater. Eng. Perform. 26(1), 439–452 (2017). https://doi.org/10.
and Experimental Results of the Sheet Metal Flange Forming 1007/s11665-016-2446-0
Based on Shell-Elements Types. Int. J. Precis. Eng. Manuf. 12(5), 37. A. Documentation, Getting Started with Abaqus Interactive
857 (2011). https://doi.org/10.1007/s12541-011-0114-8 Edition, Version (2013)
26. S.F. Golovashchenko, Quality of Trimming and its Effect on 38. G.E. Dieter, D.J. Bacon, Mechanical Metallurgy (McGraw-Hill,
Stretch Flanging of Automotive Panels. J. Mater. Eng. Perform. New York, 1986)
17(3), 316–325 (2008). https://doi.org/10.1007/s11665-008-9220- 39. H. Hooputra, H. Gese, H. Dell, H. Werner, A Comprehensive
x Failure Model for Crashworthiness Simulation of Aluminium
27. T. Wen, S. Zhang, J. Zheng, Q. Huang, Q. Liu, Bi-Directional Extrusions. Int. J. Crashworthiness 9(5), 449–464 (2004). https://
Dieless Incremental Flanging of Sheet Metals Using a Bar Tool doi.org/10.1533/ijcr.2004.0289
with Tapered Shoulders. J. Mater. Proc. Technol. 229, 795–803 40. R. Kiran, K. Khandelwal, Gurson Model Parameters for Ductile
(2016). https://doi.org/10.1016/j.jmatprotec.2015.11.005 Fracture Simulation in ASTM A992 Steels. Fatigue Fract. Eng.
28. J. McDougall, M. Stevenson, K. McKeever, Analysis of Sheet Mater. Struct. 37(2), 171–183 (2014). https://doi.org/10.1111/ffe.
Steel Fracture During Deep Drawing. J. Fail. Anal. Prev. 5(5), 12097
20–25 (2005). https://doi.org/10.1016/j.jmapro.2017.09.033 41. Y. Bao, Prediction of Ductile Crack Formation in Uncracked
29. R. Gupta, V.A. Kumar, M. Karthikeyan, P. Ramkumar, P.R. Bodies. Ph. D. thesis, Impact and Crashworthiness Lab, Mas-
Narayanan, P. Sinha, Investigation of Cracks Generated in sachusetts Institute of Technology, Cambridge, MA (2003)
Columbium Alloy (C-103) Sheets During Deep Drawing Oper- 42. F.A. McClintock, A Criterion for Ductile Fracture by the Growth
ation. J. Fail. Anal. Prev. 10(3), 228–232 (2010). https://doi.org/ of Holes. J. Appl. Mech. 35(2), 363–371 (1968). https://doi.org/
10.1007/s11668-010-9341-z 10.1115/1.3601204
30. G. Pantazopoulos, A. Sampani, Failure Analysis of Fractured 43. S. Kut, A Simple Method to Determine Ductile Fracture Strain in
Deep-Drawn 1050 Aluminum Circles. J. Fail. Anal. Prev. 6(3), a Tensile Test of Plane Specimen’s. Metalurgija 49(4), 295–299
24–28 (2006). https://doi.org/10.1361/154770206X107307 (2010)
31. J. Wu, F. Zou, Deep Drawing Failure Map of a Coated Metal
Sheet Based on the Process Parameters. J. Fail. Anal. Prev. 16(3), Publisher’s Note Springer Nature remains neutral with regard to
361–368 (2016). https://doi.org/10.1007/s11668-016-0097-y jurisdictional claims in published maps and institutional affiliations.
32. G. Yoganjaneyulu, C.S. Narayanan, A Comparison of Fracture
Limit Analysis on Titanium Grade 2 and Titanium Grade 4 Sheets

123