AbstractThis paper aims at experimental and numerical

investigation of springback behavior of sheet metals during L-

bending process with emphasis on Stribeck-type friction modeling.
The coefficient of friction in Stribeck curve depends on sliding
velocity and contact pressure. The springback behavior of mild steel
and aluminum alloy 6022-T4 sheets was studied experimentally and
using numerical simulations with ABAQUS software with two types
of friction model: Coulomb friction and Stribeck friction. The
influence of forming speed on springback behavior was studied
experimentally and numerically. The results showed that Stribeck-
type friction model has better results in predicting springback in sheet
metal forming. The FE prediction error for mild steel and 6022-T4
AA is 23.8%, 25.5% respectively, using Coulomb friction model and
11%, 13% respectively, using Stribeck friction model. These results
show that Stribeck model is suitable for simulation of sheet metal
forming especially at higher forming speed.

KeywordsFriction, L-bending, Springback, Stribeck curves.
HE most prominent feature of sheet metal forming process
is an elastic recovery phenomenon during unloading
which leads to springback [1]. The following statement best
describes springback: A purely elastically bent sheet will
return to its original configuration upon removal of the
bending moment. After partially plastic bending; permanent
deformation and residual stress remain after unloading [2].
Evidently, bending causes the metal on the outside of the
neutral axis to be under a state of tension, whereas the inside is
subjected to compression. During the bending process, internal
stresses are developed in the sheet and upon unloading; the
internal stresses do not vanish. After bending, the extrados is
subjected to residual tensile stress and the intrados is subjected
to residual compressive stress. These residual stresses produce
a net internal bending moment which causes springback [3].
The sheet continues to springback until the internal bending
moment drops to zero.
Despite the broad application of sheet metal forming, the
design of the tools and the selection of the sheet materials are
still usually based on trial and error efforts, a very expensive
and time consuming procedure. Reliable finite element models

M. Ramezani is a Research Fellow and Lecturer at Centre for Advanced
Manufacturing Technology (CAMTECH) of Auckland University of
Technology, New Zealand (phone: +64 223 228 807; e-mail:
[email protected]).
T. Neitzert is a Professor of Mechanical Engineering and Director of
Engineering Research Institute (ERI) at Auckland University of Technology,
New Zealand.
T. Pasang is an Associate Professor and Head of Department of
Mechanical Engineeringat Auckland University of Technology, New
for describing the process would be of great value in reducing
much of the tool tryout work [4]. For these finite element
simulations, accurate models of the material behavior under
deformation as well as the friction between tool and metal are
needed. For metal forming simulations, a good friction model
is very important, as shown by [3][5]. For most forming
simulations the value of coefficient of friction is chosen as a
constant, neglecting the fact that the parameters on which this
value depends might change during the process. Coulomb
friction model is a simple model frequently used in simulation.
In this model, the ratio between friction force and normal
force, defined as the coefficient of friction, considered to be
constant, as used by [6], [7]. But, friction depends on a large
number of parameters, e.g. the micro-geometry, the macro-
geometry, the lubricant and the operational parameters:
velocity, temperature and normal load. If one of the
parameters changes; the coefficient of friction will also change
[8].
In the present paper, the L-bending of mild steel and
aluminum alloy 6022-T4 sheets is simulated with emphasis on
Stribeck-type friction modelling. From the authors
knowledge, there has been no previously published analysis of
Stribeck-type frictional behavior in L-bending. The springback
behavior of sheets was studied using numerical simulations by
ABAQUS software with two types of friction model:
Coulomb friction and Stribeck friction. The results have been
compared with experimental results. The influence of forming
speed on springback behavior of sheets is also studied
experimentally and numerically.
II. STRIBECK FRICTION MODEL
When a metal forming process is observed, it is clear that
the conditions in all the different contacts are very different.
For most forming simulations the value of coefficient of
friction is chosen as a constant, neglecting the fact that the
parameters on which this value depends might change during
the process. Often, several metal forming simulations with
different values for the coefficient of friction have to be
performed before the simulation provides acceptable results. It
is clear that these simulations have no predicting power at all
[4]. From this, it is obvious that a model which describes
coefficient of friction as a function of local contact conditions
is needed. The limiting strain of a material is not directly
changed by friction, but friction changes the stress and strain
distribution. The redistribution of stress and strain, can affect
defects on metal forming, such as springback.
Stribeck is credited for carrying out the first systematic
experiments unfolding a clear view of the characteristic curve
Simulation of Kinetic Friction in L-Bending of Sheet
Metals
Maziar Ramezani, Thomas Neitzert, Timotius Pasang
T
I. INTRODUCTION
Zealand.
World Academy of Science, Engineering and Technology
International Journal of Chemical, Nuclear, Metallurgical and Materials Engineering Vol:8 No:6, 2014
471
I
n
t
e
r
n
a
t
i
o
n
a
l

S
c
i
e
n
c
e

I
n
d
e
x

V
o
l
:
8
,

N
o
:
6
,

2
0
1
4

w
a
s
e
t
.
o
r
g
/
P
u
b
l
i
c
a
t
i
o
n
/
9
9
9
8
5
8
0


of the coefficient of friction versus speed [9]. In recognition of
his contribution, this curve is called the Stribeck curve [10].
The Stribeck curve has also been proven to be useful for
identifying boundary, mixed, elasto-hydrodynamic and
hydrodynamic lubrication regimes [11]. In bending process,
contact regions operate in boundary lubrication regimes. Only
very extraordinary conditions (such as high forming speed),
make it possible for the contact functions to be in mixed
lubrication regime.
In lubricated contact surface between punch and sheet, the
total normal load
T
F is shared by the hydrodynamic lifting
force,
H
F and the asperity interacting force,
C
F .

F F F C H T
+ =
(1)

Similarly, the total friction force
f
F is the sum of two
components
F H f ,
(the hydrodynamic friction force) and
F C f ,

(the asperity interacting friction force). The hydrodynamic
friction force
F H f ,
is represented as

aB
e F
L hc
u
L H f
2 ). 1 (
) (
,

= (2)

where B is the contact length; u is the relative velocity;
L
is
the limiting shear stress; h is the film thickness; and a is the
half width of Hertzian contact. Assuming that the coefficient
of friction f
ci
is constant for all asperities, we arrive at the
following relationship for the friction force

F
f
dA
p f
dA
p f
F c
c
N
i
c
N
i
C f
c
c
c
c c i
ci
i
i
i
ci
i
A A
= = =

= = 1 1
,
(3)

where f
c
is determined from experiments.
Hence, the coefficient of friction can be obtained from

F
F
f
F
F
F
T
c
c
H f
T
f
f
+
= =
,

(4)

A full description of the Stribeck theoretical model can be
found in [12].
III. FINITE ELEMENT SIMULATION
The finite element method has been used to simulate
springback in L-bending of sheet metals. To study the effect of
friction model on springback prediction, the plane-strain
bending process was simulated in the commercial FEM
software ABAQUS/Standard. Springback is often an
important part of a forming analysis because the springback
analysis determines the shape of the final, unloaded part.
ABAQUS provides the capability to import a deformed mesh
and its associated material state from ABAQUS/Standard into
Abaqus/Explicit and vice versa. This capability is particularly
useful in manufacturing problems; for example, the entire
sheet metal forming process (which requires an initial
preloading, forming, and subsequent springback) can be
analyzed. In this case the initial preloading can be simulated
with ABAQUS/Standard using a static procedure and the
subsequent forming process can be simulated with ABAQUS
/Explicit. Finally, the springback analysis can be performed
with ABAQUS/Standard. Since springback involves no
contact and usually includes only mild nonlinearities,
ABAQUS/Standard can solve springback problems much
faster than ABAQUS/Explicit can. Therefore, the preferred
approach to springback analyses is to import the completed
forming model from ABAQUS/Explicit into ABAQUS
/Standard.
The rigid punch and die are modeled in ABAQUS/Standard
as analytical rigid surfaces with the *SURFACE option in
conjunction with the *RIGID BODY option. The top and
bottom surfaces of the sheet are defined with the *SURFACE
option. Two friction model: Coulomb friction and Stribeck
friction, implemented to ABAQUS/Standard and the results
have been compared with the experimental results.
Models for simulating sheet metal forming processes
consist of different sub-models. Each of these sub-models
takes a specific aspect of the total process into account. The
main two sub-models for sheet metal bending are material
model and friction model. The entire analysis is carried out in
five steps. In the first step the blank holder is pushed onto the
blank with a prescribed displacement to establish contact. In
the second step the boundary condition is removed and
replaced by the applied force of 100 kN on the blank holder.
This force is kept constant during Steps 2 and 3. This
technique of simulating the clamping process is used to avoid
potential problems with rigid body modes of the blank holder,
since there is no firm contact between the blank holder, the
blank, and the die at the start of the process. The two-step
procedure creates contact before the blank holder is allowed to
move freely.
In the third step the punch is moved toward the blank by
specifying a total displacement. This step models the actual
bending process. During this step the DISCONTINUOUS
analysis option is included since contact with friction tends to
create a severely discontinuous nonlinearity and we wish to
avoid premature cutbacks of the automatic time
incrementation scheme.
The last two steps were used to simulate springback. In the
fourth step all the nodes in the model are fixed in their current
positions and the contact pairs are removed from the model.
This is the most reliable method for releasing contact
conditions. In the fifth, and final, step the regular set of
boundary conditions is reinstated and the springback is
allowed to take place.
Mild steel and aluminum alloy 6022-T4 were used for
modeling. Simulations were carried out using Hills plasticity
model. The dimensions used in simulation of L-bending
according to experimental work are shown in Fig. 1. The
thickness of the sheet is 1 mm.
World Academy of Science, Engineering and Technology
International Journal of Chemical, Nuclear, Metallurgical and Materials Engineering Vol:8 No:6, 2014
472
I
n
t
e
r
n
a
t
i
o
n
a
l

S
c
i
e
n
c
e

I
n
d
e
x

V
o
l
:
8
,

N
o
:
6
,

2
0
1
4

w
a
s
e
t
.
o
r
g
/
P
u
b
l
i
c
a
t
i
o
n
/
9
9
9
8
5
8
0


The punch and die material is H13 tool steel. In this
simulation two frictional models were used in the analysis:
Coulomb friction model with a constant coefficient of sliding
friction 0.13 and Stribeck friction model with initial
coefficient of sliding friction of 0.13 for lubricated surface
between mild steel sheet and punch and 0.14 between 6022-T4
AA sheet and punch. The value of coefficient of friction is
based on historical data for similar cases [13][15]. The
Stribeck curves used in this analysis are a function of velocity
and contact pressure. These curves were implemented to
model through contact property options.


Fig. 1 Dimensions used in L-bending




Fig. 2 Variation of (a) sliding velocity and (b) Hertzian contact
pressure with time

IV. RESULTS AND DISCUSSIONS
As mentioned before, coefficient of friction changes during
the process and has different values in different nodes and
different time during the process. The change of sliding
velocity and Hertzian contact pressure with time during plane
strain bending process based on the Stibeck curve is illustrated
in Fig. 2 for a sample point between punch and sheet for a
period of 0.1s during the punching strive. The variation of
coefficient of friction with time during plane strain bending
process based on the Stibeck curve is shown in Fig. 3 for a
sample point between punch and sheet for a period of 0.1s
during the punching strive. As can be seen in Fig. 3, the
change in coefficient of friction varies between initial value of
0.13 reducing to 0.11 at 0.07s and increases afterward which is
significant and this affects the stress distribution in sheet
which in turn affects the springback. As illustrated in Fig. 3,
the coefficient of friction in different points between sheet and
die changes significantly during process. So it is clear that
using a constant coefficient of friction as uses by Coulomb
model has no predicting power and may lead to poor results
for precise sheet metal forming simulations.
The amount of springback for mild steel and 6022-T4 AA
in L-bending was calculated numerically using finite element
program ABAQUS/Standard. The relationship between
springback and the ratio of the tensile strength to the elastic
modulus for s mm u / 20 = , using different friction models is
given in Fig. 4. As can be seen from Fig. 4, as the ratio of the
tensile strength increases, the springback increases. The FE
prediction error for mild steel and 6022-T4 AA is 22.5% and
25% respectively, using Coulomb friction model; and 13%
and 16% respectively, using Stribeck friction model. So,
Stribeck-type friction model predict springback better than
Coulomb model.
The amount of springback at different forming speeds was
measured experimentally and compared with finite element
results and illustrated in Fig. 5. According to Fig. 5, the
prediction of springback is improved by using Stribeck
frictional model. Especially at higher forming speeds, the
FEM results are very close to the experimental results. For
mild steel at s mm u / 25 = , the FE springback prediction
error using Coulomb model is 23.8%. Using Stribeck friction
model, this error decreases to 11%. For 6022-T4 AA at
s mm u / 25 = , the FE prediction error using Coulomb model
is 25.5%. Using Stribeck model, this error decreases to 13%.
This is due to more changes in coefficient of friction at higher
velocities, according to Stribeck curves.
The coefficient of friction contributes to the amount of
springback. During bending, the metal is forced between the
lower die section and the forming punch. If the clearance
between these two sections is less than the metal thickness (as
it usually is), intense friction is created. The amount of friction
determines how much the bend will be stretched [16]. If the
inside bend radius is large enough and the metal can be
stretched over it, the amount of compression is reduced or
eliminated. If compression is eliminated, both sides of the
radius are in tension, thus reducing the amount of springback
World Academy of Science, Engineering and Technology
International Journal of Chemical, Nuclear, Metallurgical and Materials Engineering Vol:8 No:6, 2014
473
I
n
t
e
r
n
a
t
i
o
n
a
l

S
c
i
e
n
c
e

I
n
d
e
x

V
o
l
:
8
,

N
o
:
6
,

2
0
1
4

w
a
s
e
t
.
o
r
g
/
P
u
b
l
i
c
a
t
i
o
n
/
9
9
9
8
5
8
0


[17]. Although an increase in friction may reduce the amount
of springback, intense friction may results in severe wear to
the die sections. For this reason, lubricants should be used.


Fig. 3 Variation of coefficient of friction with time


Fig. 4 The relationship between springback and the ratio of tensile
strength to elastic modulus for u=20mm/s


Fig. 5 Springback angle obtained from experimental and numerical
analysis at different forming speeds
V. CONCLUSION
The prediction of springback in bending processes has been
a challenging topic. For sheet metal forming processes, the
frictional behavior depends on several parameters such as the
contact pressure, sliding speed, sheet and tool material, surface
roughness, lubricant and concurrent deformation. Because all
of these variables influence friction, the question arises as to
whether the Coulomb simple friction model is capable of
describing the real frictional properties of sheet metal forming
processes. In the present paper, the Stribeck friction model
was used to investigate springback behavior of sheet metals
under L-bending. The Stribeck curves can be calculated by
varying the velocity while keeping the other operational
conditions constant. The following conclusions were reached
regarding the choice of stribeck friction model.
Coefficient of friction changes during the process and has
different values in different nodes and different time
during the process. An accurate friction model which
describes coefficient of friction as a function of local
contact conditions is needed.
Stribeck-type friction model has better results in
predicting springback in sheet metal forming. The FE
prediction error at s mm u / 20 = for mild steel and
6022-T4 AA is 22.5%, 25% respectively, using Coulomb
friction model and 13%, 16% respectively, using Stribeck
friction model. These results show that Stribeck model is
suitable for simulation of sheet metal forming.
Stribeck-type friction model has better results at higher
forming speeds. For 6022-T4 AA at s mm u / 25 = , the
FE analysis shows 25.5% error using Coulomb model and
13% error using Stribeck model.
REFERENCES
[1] J .R. Cho, S.J . Moon, Y.H. Moon, S.S. Kang, Finite element
investigation on spring-back characteristics in sheet metal U-bending
process, J ournal of Materials Processing Technology 141 (2003) 109
116.
[2] Mullan HB (2004) Improved prediction of springback on final formed
components. J . of Materials Processing Technology 153154, pp. 464
471.
[3] Li KP, Carden WP, Wagoner RH (2002) Simulation of springback. Int.
J . of Mechanical Sciences 44, pp. 103122.
[4] M. Math, B. Grizelj, Finite element approach in the plate bending
process, J . of Materials Processing Technology 125126 (2002) 778
784.
[5] Lee BH, KeumYT, Wagoner RH (2002) Modeling of the friction caused
by lubrication and surface roughness in sheet metal forming. J . of
Materials Processing Technology 130131, pp. 6063.
[6] Y.E. Ling, H.P. Lee, B.T. Cheok, Finite element analysis of springback
in L-bending of sheet metal, Journal of Materials Processing
Technology 168 (2005) 296302.
[7] F.K. Chen, S.F. Ko, Deformation analysis of springback in L-bending
of sheet metal, J ournal of Achievements in Materials and
Manufacturing Engineering, Volume 18 Issue 1-2 SeptemberOctober
2006.
[8] Matuszak A (2000) Factors influencing friction in steel sheet forming. J .
of Materials Processing Technology 106, pp. 250253.
[9] J acobson B (2003) The Stribeckmemoriallecture. Tribology Int. 36, pp.
781789.
[10] Lu X, Khonsari MM, Gelinck ERM (2006) The Stribeck curve:
experimental results and theoretical prediction. ASME J. of Tribology,
Vol. 128, October 2006.
World Academy of Science, Engineering and Technology
International Journal of Chemical, Nuclear, Metallurgical and Materials Engineering Vol:8 No:6, 2014
474
I
n
t
e
r
n
a
t
i
o
n
a
l

S
c
i
e
n
c
e

I
n
d
e
x

V
o
l
:
8
,

N
o
:
6
,

2
0
1
4

w
a
s
e
t
.
o
r
g
/
P
u
b
l
i
c
a
t
i
o
n
/
9
9
9
8
5
8
0


[11] Gelink ERM, Schipper DJ (2000) Calculation of Stribeck curves for line
contacts. Tribology Int. 33, pp. 175-181.
[12] Ramezani M, Ripin ZM, Ahmad R (2009) Computer aided modelling of
friction in rubber-pad forming process. J ournal of Materials Processing
Technology 209 (10) , pp. 4925-4934.
[13] W. Wang, R.H. Wagoner,X.J . Wang, Measurement of Friction under
Sheet Forming Conditions, Metallurgical and Materials Transactions A,
Volume 27A, December 19963971.
[14] Wang Z, Nakamura T, Dohda K, Obika T (2003) FEM analysis of
contact mechanismin press-forming of lubricant pre-coated steel sheet.
J . of Materials Processing Technology 140, pp. 514519.
[15] Hao S, Klamecki BE, Ramalingam S (1999) Friction measurement
apparatus for sheet metal forming. Wear 224, pp. 17.
[16] Kuzman K (2001) Problems of accuracy control in cold forming. J. of
Materials Processing Technology 113, pp. 1015.
[17] Gusel L, Anzel I, Brezocnik M (2005) Effect of lubrication on the stress
distribution in an extruded material. Int. J . of Advanced Manufacturing
Technology, pp. 25: 288-291.




World Academy of Science, Engineering and Technology
International Journal of Chemical, Nuclear, Metallurgical and Materials Engineering Vol:8 No:6, 2014
475
I
n
t
e
r
n
a
t
i
o
n
a
l

S
c
i
e
n
c
e

I
n
d
e
x

V
o
l
:
8
,

N
o
:
6
,

2
0
1
4

w
a
s
e
t
.
o
r
g
/
P
u
b
l
i
c
a
t
i
o
n
/
9
9
9
8
5
8
0