See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/296394758

SIMULATION OF THE COMBINED RECONFIGURABLE MULTIPOINT FORMING

AND RUBBER FORMING

Conference Paper  in  Steel Research International · January 2008

CITATIONS READS

3 36

5 authors, including:

Paunoiu Viorel P. Cekan

Universitatea Dunarea de Jos Galati TPM Engineering & Construction Ltd
51 PUBLICATIONS   77 CITATIONS    2 PUBLICATIONS   20 CITATIONS   

SEE PROFILE SEE PROFILE

Mihaela Banu Alexandru Epureanu

Universitatea Dunarea de Jos Galati Universitatea Dunarea de Jos Galati
70 PUBLICATIONS   311 CITATIONS    122 PUBLICATIONS   135 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

digital forming View project

RU-TE-0031 View project

All content following this page was uploaded by P. Cekan on 03 May 2018.

The user has requested enhancement of the downloaded file.

 SIMULATION OF THE COMBINED RECONFIGURABLE
MULTIPOINT FORMING AND RUBBER FORMING
Viorel Paunoiu1, Peter Cekan2, Mihaela Banu1, Alexandru Epureanu1, Dumitru Nicoara1
1
Dunarea de Jos University of Galati, Romania
2
PC Engineering Solutions, Slovakia
ABSTRACT
One of the methods for increasing the qual-
ity of the reconfigurable multipoint formed
parts is to use an elastic interpolator be-
tween the pins active elements and the
blank. Therefore it is important to know the
behaviour of the technological system
formed by the three media rigid, elastic and
elastic-plastic. In the paper are presented
the simulations results, using finite element
approach, regarding the influence of the
interpolator behaviour toward the parts
properties. The parts quality is evaluated in
terms of springback and stresses state varia-
tion. The results could lead to the develop-
ment of new concepts to be used in the die
design procedure.
Keywords: multipoint forming, rubber
forming, numerical simulation
1. INTRODUCTION
Forming with multipoint reconfigurable
dies is a flexible manufacturing stamping
technology based on the active elements
conjugated continuous 3-D surfaces parti-
tion with a help of an adequate discrete
punches (figure 1). The desired surface
shapes of active elements, which finally
will determine the part shape could be ob-
tained by adjusting the heights of these
punches. One of the methods utilized for
increasing the reconfigurable multipoint
formed parts quality is to interpose an elas-
tic plate (interpolator) between the pins ac-
tive elements and the blank to avoid the
dimpling phenomenon and to uniform the
pressure distribution upon the blank. The
properties of the interpolator material used
are important. The disadvantage of using
soft materials is that the pins can push
through the pad and dimple the sheet metal.
If the interpolator is hard, it will not con-
form to the blank geometry. The perfect
interpolator would be very soft initially un-
til all the gaps have been filled, then it
would become perfectly rigid so it would
not fail. According to [1], the gaps are filled
at different points for each part, so a com-
promise of this hardening point must be
discovered. The perfect interpolator would
also be very cheap and would retain its
qualities after many stampings [1].
Punch
Die
Figure 1. Technological equipment for re-
configurable multipoint forming.
A small number of papers treat the problem
of the interpolator presence in the system
[1, 2].
As the result the paper focuses in the study
of the interpolator thickness and pins pene-
tration influence toward the part quality in
terms of springback, stresses state varia-
tions. The study was made using finite ele-
ment method implemented in Dynaform
program.
2. FEM SIMULATION MODEL
The model was developed using Dynaform
finite element program, a nonlinear dy-
namic software which can simulate differ-
ent types of sheet metal forming process to
predict the stresses, strain, thickness distri-
bution, punch load and effect of various de-
sign parameters of tooling on process effi-
ciency and final product.
In order to simulate the process, it was con-
structed a tool geometry without interpola-
tor, considering the obtaining of a single
curvature plate with an interior radius of 95
mm, a width of 120 mm (maximum depth is
21.345 mm) and a length of 130 mm. Fig-
ure 2 presents the simulation model. The
model includes the fixed configuration pre-
sented above and the two interpolators (up-
per and down rubber) between the active
elements and the blank. No blankholder
was used so the ends of the rubbers are free
to expand.
Punch
Upper rubber
Blank
Die
Down rubber
Figure 2. Tooling for reconfigurable multi-
point forming.
The blank material used in experiments was
mild steel, with 1 mm thickness. The yield-
ing of the material was modelled using a
power law:
σ = K εn (1)
where: K is the material characteristic;
n – hardening exponent. In simulation the
n – value = 0.22 and K = 648 MPa. The
R values were set to: R00 – 1.87; R45 – 1.27;
R90 – 2.17.
The blank is a rectangular plate with the
dimensions of 120x130x1 mm and the
mesh consists of 900 finite elements.
The FE blank mesh consists of 4-node Be-
lytschko-Tsay shell elements, with five in-
tegration points through the thickness of the
sheet [4]. The Belytschko-Lin-Tsay shell
element are based on a combined co-
rotational and velocity-strain formulation.
The Coulomb friction law was used with a
friction coefficient of 0.125. The punch
speed was 100 mm/second.
The tooling was modelled as rigid surfaces.
The geometry of the tool is characterized in
terms of an array of pin positions. The
geometrical model of die-punch tool was
composed from two working arrays with
100 pins for each array, 10 rows on x-
direction and 10 rows on y-direction. The
pins are disposed face to face, both on x-
direction and y-direction. The mesh consists
of 488744 numbers of finite elements.
For rubber interpolator was chosen a
material type Elvax 460. The properties of
the material were: density, ρ – 0.946 g/cm3;
hardness Shore ASTM D2240 scale B – 40
and scale A – 80; tensile strength, Rm – 18
MPa; elongation – 750%; flexural modulus
– 44 MPa; stiffness, k – 43 MPa; Poisson
ratio, ν – 0.499. Solid elements were used
for discretization the rubber interpolator.
The interpolator was modelled as Mooney
Rivlin Rubber material. The above density,
flexural modulus and Poisson ration were
used. The thicknesses of the rubbers varied
between 2 and 10 mm. The simulations
were done also with different strokes for the
punch.
3. SIMULATION RESULTS
During these simulations, the penetration of
the pins into the interpolator was observed
in order to determine its influence. Also the
effect of thickness interpolator toward the
part quality was determined.
3.1. Rubber deformation
Some important observation could be made
regarding the rubber deformation.
Figure 3 presents the image of the upper
rubber interpolator form after deformation,
thickness 10 mm, 23 mm pins stroke.
Front end Waves
Lateral end
Figure 3. A rubber interpolator after defor-
mation
As the part is formed on the die, the inter-
polator is compressed in the thickness di-
rection by the pins, and thus tends to ex-
pand in x direction and bends in y direction
(figure 4). This depends from the rubber
thickness and the pins strokes for the same
rubber material.
a.
b.
Figure 4. Upper view of the 8 mm thickness
rubber interpolator after deformation, pins
stroke: a. 19 mm; b. 27 mm
The rubber deformation is characterized by
the wave appearance. Their form depends
from the gap between the pins, the pins
stroke and the rubber thickness.
Lateral end
Waves a.
b.
Figure 5. Front view of the rubber interpo-
lators and blank assembly after deforma-
tion: a. 2 mm thickness; b. 10 mm thickness
For smaller interpolator thickness the picks
of the waves are rounded (figure 5, a.) and
become sharp when the interpolator thick-
ness is greater and the gaps between the
pins are filled (figure 5, b.). An interpolator
small thickness or a greater pins stroke lead
to a non-uniform pressure toward the blank
having as a result an uneven supplementary
deformation of the blank (figure 6). The
maximum pressure is situated in the middle
of the rubber.
The presence of an interpolator with a
greater thickness in the system could assure
a uniform pressure toward the blank.
Figure 7 presents the means stresses varia-
tion in the upper and down rubber interpo-
lators for different rubber thickness and the
same pins strokes. As it can see, the in-
creasing of the rubber thickness leads to the
reduction of the means stress in the rubber
material which will affect the blank defor-
mation. Also it appears a difference be- defined function of the three parameters
tween the mean stress in the upper and presented in figure 9.
down rubber (figure 7).

Front end

a.

Figure 6. Uneven blank deformation, rubber

interpolator 4 mm thickness, 25 mm pins
stroke
120
3-23
Upper Rubber
4-23 Down rubber
3-23
100
5-23
4-23

b.
Mean Stress [MPa]

80 6-23
5-23

60
6-23
7-23

8-23
Figure 8. Lateral view of the rubber interpo-
7-23

8-23
9-23
10-23
lators after deformation, for 10 mm thick-
40
9-23 ness, pins stroke: a. 17 mm; b. 20 mm
10-23

20
B
0
Rubber thickness-Pins penetration [mm-mm]

R
Figure 7. Mean stresses variation in the
rubber interpolator H

In the considered fixed configuration of the

pins, for the established geometry, for a
greater rubber thickness appears a phe-
nomenon of supplementary bending of the
lateral ends of the rubber in y direction Figure 9. Parameters for springback defini-
which will affect the part geometry (figure tion
5). Also the front ends expand in x direc-
tion, with increasing the rubber thickness The springback was calculated with the re-
and the pins stroke (figures 4 and 8). lation:

3.2. Blank deformation Vi − V f

The geometry considered is affected by the ΔS = (2)
rubber presence in terms of profile radius
Vi
and depth.
First, the blank deformation was evaluated where: ΔS is the value of the springback, S
in terms of springback. The springback was is one of the three parametrs; Vi – intial
value of the one of the three parametrs;
Vf – final value of the one of the three stroke, and different rubber thickness, the
parametrs. values of springback are presented in figure
The amount of springback experienced by 10. When the rubber thickness are increas-
the workpiece when it is unloaded from the ing the values of the springback reported to
die is affected by the depth of penetration the width and to the radius are increasing
and rubber thickness. The depth of penetra- and the springback values reported to the
tion in the rubber transfers load to the sheet, profile radius are decreasing.
thus affecting the stress state in the sheet at This is due the fact that using a fixed rigid
the end of the deformation stage, and the configuration of the tool, different rubber
amount of springback. thickness and the same stroke, the obtained
The obtained values of the springback for radius of the parts are different, increasing
different process parameter are presented in with the increasing of the rubber thickness.
table 1. This lead to the decreasing of the Von
Mises stresses (figure 11), and to an in-
Table 1. Values of springback parameters creasing of the elastic stresses in the metal-
Rubber Values of springback lic part, respectively to an increasing of the
thickness- parameters amount of springback.
Pins
ΔB ΔR ΔH
stroke 0,035

2-21 0,0291 -0,2231 0,1969

0,03 10-23
ΔB variation [mm]

9-23
8-23
0,025

2-23 0,0150 -0,1021 0,0990 0,02

7-23

6-23
3-22 0,0225 -0,1550 0,1499 0,015
5-23

3-23 0,0094 -0,0466 0,0517 0,01

3-23
4-23

4-23 0,0117 -0,0532 0,0601

0,005

5-22 0,0290 -0,2093 0,1886 Rubber thickness-Pins stroke [mm-mm]

5-23 0,0150 -0,0725 0,0789 a.

5-24 0,0141 -0,0139 0,0352 0

6-21 0,0337 -0,2653 0,2260 3-23 4-23

ΔH variation [mm]

-0,05
5-23

6-23 0,0193 -0,1066 0,1079 -0,1 6-23

6-25 0,0224 -0,0047 0,0423 -0,15

7-23

7-20 0,0334 -0,2610 0,2221 -0,2

8-23
9-23
10-23

7-23 0,0232 -0,1411 0,1375 -0,25

8-19 0,0353 -0,2824 0,2370 Rubber thickness-Pins stroke [mm-mm]

8-23 0,0272 -0,1782 0,1669 b.
8-27 0,0155 -0,0398 0,0563 0,2

9-18 0,0359 -0,2741 0,2345

10-23
0,18
9-23
ΔR variation [mm]

9-23 0,0298 -0,2036 0,1856

0,16 8-23
0,14
7-23

9-28 0,0162 -0,0591 0,0720 0,12

0,1 6-23

10-17 0,0401 -0,3221 0,2641 0,08

0,06
5-23

10-20 0,0365 -0,2863 0,2409 0,04 3-23

4-23

10-23 0,0307 -0,2080 0,1905 0,02

0
Rubber thickness - Pins stroke [mm-mm]
10-29 0,0197 -0,0898 0,0986
10-30 0,0197 -0,0901 0,0991 c.

The amount of springback experienced by
the workpiece when it is unloaded from the
die is affected by the rubber thickness and
depth of penetration. For the same depth of
penetration, this means the same pins
Figure 10. Springback variation for the
same pins stroke: a. width variation; b.
height variation; c. radius variations.
 The decreasing of the Von Mises stresses It is obviously when the interpolator thick-
are in accordance with the mean stresses ness is small appears form errors in the part
variation in the rubber (see figure 7). sections when the pins stroke is big. When
the rubber interpolator is big the process of
400 deformation could be affected because of
3-23
the uncontrolled flow of the elastic me-
Von Mises Stress [MPa]

350

dium. The geometry of the part will affect

300
4-23 5-23 6-23 7-23 8-23 9-23
250
10-23
200 the rubber deformation with a negative in-
150 fluence toward the metallic part quality.
100

50
The obtained results could lead to the de-
0
velopment of new concepts to be used in the
Rubber thickness-Pins stroke [mm-mm] die design procedure. The very complex
behaviour of the system make necessary to
Figure 11. Von Mises stresses variation for develop new studies in this field.
the same pins stroke and different rubber
AKNOWLEDGEMENTS
thickness
The work reported here was an activity in
the frame of Romanian Grants: CEEX-P-
The depth of penetration in the rubber
CD no. 130/2006 – Materials and technolo-
transfers load to the sheet, thus affecting the
gies for plane and spatial profiling and
amount of the springback. This thing is pre-
CNCSIS Grant A686/2007 – A new theory
sented in figure 12.
of nanostructuring as a recursive chaotic
0
process and designing of a new method of
nanostructuring by controlled multidirec-
6-25 5-24
8-27
-0,05
5-23 9-28
tional deformation.
ΔH variation [mm]

10-29
-0,1
6-23
-0,15
8-23
-0,2
5-22
9-23 10-23 LIST OF REFERENCES
-0,25 6-21
9-18
1. Eigen G., “Smoothing method for dis-
-0,3 8-19
10-17
10-20
crete die forming”, PhD Thesis, MIT
-0,35

Rubber thickness-Pins stroke [mm-mm]

University, Massachusetts, 1992.
2. Socrate S., Boyce M., “A finite Element
based die design algorithm for sheet-
Figure 12. Springback variation in terms of metal forming on reconfigurable tools”,
height variation, for the different pins JEMT, vol. 123/489, 2001, pp. 489-495.
strokes and rubber thicknesses 3. Paunoiu V., Oancea N., Nicoara D.,
“Simulation of Plate’s Deformation Us-
With increasing the pins stroke and the rub- ing Discrete Surfaces, Materials Proc-
ber thickness the amount of springback will essing and Design: Simulation and Ap-
increase also. plication”, Proc. Numiform’2004, Ohio
State University, American Institute of
4. CONCLUSIONS Physics, Melville, pp. 1007-1010.
In the present numerical study, considering 4. Belytschko T., Liu W.K., Moran B.,
a fixed rigid configuration of the tool, the “Finite Elements for Nonlinear Con-
same friction coefficient between the sys- tinua and Structures”, Wiley, New
tem components, different rubber thickness, York, 1996.
different pins penetrations, free rubber ex- 5. Cai Z.Y., Li M.Z., “Multi-point forming
pansion, and the same rubber material, it of three-dimensional sheet metal and
results some conclusions: the control of the forming process”, Int.
The presence of the interpolator changes the J. Pressure Vessels Piping, vol. 79/4,
behaviour of the technological system 2002, pp. 289–296.
which affects the part quality.

View publication stats