THE ANNALS OF DUNREA DE JOS UNIVERSITY OF GALAI FASCICLE V, TECHOLOLOGIES IN MACHINE BUILDING, ISSN 1221-4566 2007

Designing of a Reconfigurable Set of inverse Re-Drawing Dies Aided by FEM Simulation

C. Maier1, V. Tabacaru1, M. Banu1, S. Bouvier2, V. Marinescu1
1

Universitatea Dunrea de jos Galai, Facultatea de Mecanic 2 University Paris13, LPMTM 93430 Villetaneuse, France Abstract This paper deals with the methodology for developing a laboratory inverse redrawing device. The drawing process is performed in two phases: a direct drawing of a circular blank followed by a second reverse re-drawing phase on the same device. Finite element simulations are carried out in order to i).define geometrical characteristics of the modular re-drawing device and to ii). estimate the punch force evolution for different dimensions of punch, die and blankholder and for a large class of materials. Based on such FEM simulations, springs for the developed reverse deep drawing device are dimensioned. The use of springs gives the possibility to deform the material with an imposed blank-holder force. Finally, a draw of the designed modular device is presented considering all the results of the finite element simulation.. Keywords: finite element simulations, inverse re-drawing, strain path, modular device.

1. Introduction
The analysis of the inverse re-drawing process provides several advantages as (i) the imposed strain path changes are severe and complex leading to a better analysis of the accuracy of the finite element analysis, (ii) large amount of deformation can be imposed where limitations due to the localization problems are reduced, (iii) the imposed strain path changes can determine the compensation of the bendingunbending effects (tensile stress and bending moment) due to the material flow over the tools radius during the first stage of the deep-drawing process. In this study we propose to proceed with a direct drawing of a circular blank followed by a second inverse re-drawing stage using a laboratory device. These two stages must to be progressive in order to avoid the errors due to the positioning of the piece obtained in the first stage. T ab le 1 Mat er ia l AL 5 1 8 2 DP 6 0 0 HS L A Y0 [ Mp a] 130 260 370 Rm [ Mp a] 340 840 530 E [ Mp a] 72000 210000 210000

0 .3 2 0 .3 3 0 .3 3

T yp e o f t he p iece 1 2

D [ mm] 78 74 70 60 52

T [ mm] 1 1 ,2 1 ,5 1 1 ,2

H [ mm] 70 70 70 70 77

T ab le 2 R [ mm] 8 ,5 10 11 8 ,5 10

2. Design of the inverse re-drawing dies

The laboratory device is designed considering these cases: - material AL5182, DP600, HSLA (table 1); - pieces type 1 and 2 (fig.1, table 2).

This device has changeable tools (punch, die, blankholder) in order to assure the conditions to obtain all pieces, using all materials considered.

15

FASCICLE V

THE ANNALS OF DUNREA DE JOS UNIVERSITY OF GALAI

3. Finite element simulation

Finite element simulations using a staticexplicit finite element code STAMP3D, released within Integrated V-CAD Research System Program in RIKEN Institute, Japan, and an implicit code MARC Mentat 3.2 are carried out in order to estimate the punch force evolution for different dimensions of punch, die and blankholder, and for a large class of materials. We considered the symmetry of the system and its result, in order to model only half of it (figure 3).

Figure 1. The definition of the tool-sets (punch, die, blankholder, clearence) and choice of process parameters (number of forming stages, forces to apply) must be performed for every type of the piece and material considered. Improvement of design and tryout procedures using numerical simulation may have a significant impact on the cost of the tools and on the reduction of the total time from design to manufacture, also with the possibility to provide better solution than those determined from purely experimental tryout procedures. Following the preliminary design, the tools must have the geometry presented in figure 2 in order to perform the deep-drawing in two stages on the same device.

Fi g ur e 3 . E q u iva len t p la st ic st ra in d is t rib u t io n d u rin g th e s eco n d sta g e . T he t yp e o f t h e f i ni te el e me nt u sed i s p la ne s tr e s s q uad r at ic ( 4 no d es ) e le me n t i n M AR C M e nta t 3 .2 Har d e n i n g law co n s id er ed i s S wi f t ( is o tr o p ic h ar d e n i nd ) fo r s te el s ( DP 6 0 0 a nd H S L A) a nd Vo ce ( iso tr o p i c har d e ni n g) fo r al u mi n i u m al lo y. T ab le 3 . S wi ft la w ( iso t r o p ic h a rd en in g ) p a ra m ete r s [ 1] . P ar a met er DP 6 0 0 HS L A 3 0 8 .3 3 6 7 .7 Y
0

0
n C

0 .0 0 0 8 2 0 .1 3 2 7 2 0 .2

0 .0 7 1 5 7 0 .1 3 9 5 3 0 .9

T ab le 4 . V o c e la w w ith kin e ma t ic h a rd en in g p a ra me te r s [ 1] P ar a met er Val u e 1 4 8 .5 Y

0

CR R sat

9 .7 1 9 2 .4

Figure 2

T he si mu la tio n wi t h the i n it ia l p ar a me ter s h a s t wo p r o b le ms : fir s tl y t he si mu l a tio n d o e s n t r e ac h t h e e nd a nd u nd u lat io n s ap p e ar .

THE ANNALS OF DUNREA DE JOS UNIVERSITY OF GALAI

FASCICLE V

Ge ner al l y wh e n t he si m u la tio n d o es n t r e ac h t he e nd it s d ue to t he no d e s p o si tio n s wh e n t h e y e nt er i n co nt act , o r fa s t str es s i ncr ea se s at an i nc r e me n t o f t he si mu l a tio n. T hes e p r o b le ms co uld b e so l ved b y so me g eo me tr i cal mo d i f ica tio n s, o r p u n c h sp eed mo d i f ica tio n . T he p ar a me ter s a r e mo d i fied o ne b y o ne i n o r d er to k eep r ev er sib il it y i f t h e si mu l a tio n d o es n t we ll r u n. T he f ir s t mo d i f ica tio n s we r e mad e to ma k e t h e si mu l a tio n r e ac h e nd . T he f ir st id ea i s to d e cr ea se st r e s s d u e to t he f ir s t

p as s b y mo d i f yi n g t he f ir s t p a s s g eo me tr y. B ut t he so l u tio n wa s to ch a n ge t h e s eco nd st ep c lear a nce. O nce t he si mu l a tio n r ea ch t h e e nd , we tr y to r ed uc e u nd u la ti o n b y r ed uc e t he seco nd step c lea r a nc e. T he o p t i mi zed val u e o f t he seco nd st ep cle ar a nce i s 1 ,2 2 m m ( cal c ula te b y t he Ka s z ma r e k r ela tio n s h ip b ut we ar r i ved to o b tai n r e s u lt s fo r clea r a nc e a fe w b it i n fer io r to t ha t ( 1 ,2 m m) . T he F ir st s tep c lear a nce wa s al so en lar g e to d ecr e a se str es s. C hr o no lo g y of g eo me tr ic al ad j u st me n t is p r e se n ted in t he tab le 5 .

T ab le 5 Punch Round Radius 2 Punch Round Radius 1 Die Round Radius 1 Die Round Radius 2

Simulation #

0 593 30 1,3 5,2 5,5 10 809 58,5 1,3 5,2 5,5 20 924 54,5 1,3 5,2 5,5 Seek optimized first pass parameters 21 802 58,5 1,3 5,2 5,5 6 22 809 58,5 1,3 5,2 23 419 58,5 1,3 5,2 6,5 24 413 58,5 1,3 5,2 5,5 25 417 58,5 1,3 5,2 5,5 6 26 840 58,5 1,3 5,2 31 407 58,5 2,1 5,2 6 32 289 58,5 3,7 5,2 6 Seek optimized second pass parameters 40 1000 58,5 1,3 2,2 6 41 1014 58,5 1,3 2,2 5,5 Reduction of the undulation 50 787 58,5 1,3 1,5 5,5 60 1005 57,5 1,5 1,5 5,5 70 1014 57,5 1,5 1,5 5,5 80 910 58,5 1,5 1,3 5,5 81 1014 58,5 1,5 1,3 6 90 783 58,5 1,5 1,2 6 91 1014 58,5 1,5 1,22 6 100 1014 58,5 1,5 1,2 5,5 Following the finite element analysis we obtain the punches force evolution (fig. 4) necessary to define the final dimensions of the tools and to select the optimum necessary pressmachine. We represented in figure 4 the evolution of the punchs force in the

8 8,5 5,5 8 8,5 5,5 8 8,5 5,5

8 8 8 8,5 9 8,5 8,5 8,5

8,5 8,5 8,5 8,5 8,5 8,5 8,5 8,5

5,5 5,5 5,5 5,5 5,5 5,5 5,5 5,5

8,5 8,5 5,5 8 8,5 5,5

8 8 8 8 8 8 8 8

8,5 8,5 8,5 8,5 8,5 8,5 8,5 8,5

5,5 5,5 5,5 5,5 5,5 5,5 5,5 5,5

1 1 0,075 0,075 0,075 0,075 0,075 0,075

intermediary state during the second stage, the same state like in figure 3. Maximum forces to apply are obtained for first stage (on the punch 1) of the process. This values is used to define the final dimensions of the tools.

Relative sliding velocity 1 1 1 1 1 1 1 1 1 1 1 1 1

Drawing depth

Inc stop (last inc=1014)

Clearance 1

Clearance 2

FASCICLE V

THE ANNALS OF DUNREA DE JOS UNIVERSITY OF GALAI

D P600 (1.0 mm) : Variation of Hill's coeffi cient of anisotropy

1.2

Coefficient of anisotropy r

0.8

0.6

0.4

0.2 Experimental values Theoretical values 0 0 10 20 30 40 50 60 70 80 90 Orientation of the tensile axi s with respect to RD, in degrees

Fi g ur e 6 . A n i so t ro p y o f DP 6 0 0 Fi g ur e 3 . P u n ch 1 a n d p u n ch 3 fo rc e evo lu tio n
DP600 Punch diameter 82.3 mm

4. Conclusions
For the reverse re-drawing simulation the geometry of the dies is defined. The DP600 and HSLA steel have the abilities to support the severity of reverse re-drawing process. In the same time theses steels are designed for this kind of operation. The evaluate draw ratio ( 2lim =1,75 ) on the second pass is very good (superior to the draw ratio on the first pass).

120 100 Punch force [kN] 80 60 40 20 0 0 10 20 30 40 50 Punch stroke [mm]

5. Acknowledgements
This work has been material and logistic funded by the National Excellence Research Project -CEEX contract no. 22/2005. References
1. . M. Banu, C. Maier, S. Bouvier, H. Haddadi, C. Teodosiu Data Preprocessing and Identification of the Elastoplastic Constitutive Models - WP3, Task 1, 18-Months Progress Report, Digital Die Design Systems (3DS) IMS 199 000051, (2001), 22-29. 2. R. E. Dick, J. W. Yoon, F. Barlat, Convolute Cut-Edge Design for an Earless Cup in cup Drawing, In: NUMISHEET 2005, Eds. L.M. smith, F. Pourbograth, J.W Yoon, T.B. Stoughton, (2005), 713-718. 3. A. Baptista, J. L. Alves, M. Oliveira, D.M. Rodrigues, L. F. Menezes, Application of the Incremental Volumetric Remapping Method in the Simulation of Multi-Step Deep Drawing Processes, In: NUMISHEET 2005, Eds. L.M. smith, F. Pourbograth, J.W Yoon, T.B. Stoughton, (2005), 173-178

Fi g ur e 4 . P u n ch 1 fo r ce evo lu tio n fo r DP 6 0 0 . On the figure 5 we can notice the earing effect due to the anisotropy (fig. 6) of the material and we use this one in order to check the height of the piece obtained in the first stage of the process.

Fi g ur e 5 . E a r in g e ffe ct f o r D P 6 0 0 a t 5 0 mm h eig h t o b ta in ed a fte r th e f i rs t sta g e

THE ANNALS OF DUNREA DE JOS UNIVERSITY OF GALAI

FASCICLE V

Proiectarea unui echipament tehnologic reconfigurabil pentru ambutisarea inversa, utilizand simularea cu EF
Rezumat: Aceasta lucrare are ca obiectiv dezvoltarea metodologiei de proiectare a unui echipament tehnologic de laborator pentru ambutisare inversa, avand elemente active schimbabile. Procesul de deformare se realizeaza in 2 etape: ambutisarea directa a unui semifabricat plan de forma circulara, urmata de a doua ambutisare pe acelasi echipament, in sens invers sensului de deplasare a materialului de la prima ambutisare. Simularea cu elemente finite s-a realizat cu scopul: i). definirii caracteristicilor dimensionale ale echipamentului modular si ii). estimarii evolutiei fortei de deformare pentru diferite dimensiuni ale poansonului, placii active si placii de retinere, pentru o gama larga de materiale. In final, echipamentul tehnologic de laborator este dimensionat pe baza rezultatelor simularilor numerice.

Conception dun quipement de laboratoire pour lemboutissage inverse, en utilisant la simulation EF

Rsume: Ce papier a comme objectif le dveloppement de la mthode de conception dun quipement de laboratoire pour lemboutissage inverse, ayant le couple poinon - matrice changeable. Le processus de mise en forme se droule pendant deux tapes : lemboutissage directe dun bauche plan de forme circulaire, suite par le deuxime emboutissage ralis sur le mme quipement, dans le sens inverse de dplacement du matriau que le premier emboutissage. La simulation EF a t ralise au but de : i). dfinir les characteristiques dimensionales des diffrents modules de lquipement ; ii). valuation de lvolution de la force du poinon pour diffrents dimensions des modules de lquipement et diffrents matriaux de lbauche. A la fin, la conception de lquipement de laboratoire est faite base des rsultats de la simulation numrique