![Half of ingot presentation during elongation (A) with detailed distribution of equivalent strain [-] in axis section (B). Reduction 100mm, width of dies 550 mm, overlap of dies to stroke 100 mm, number of elements approx. 17,500, total length 3,400 mm, diameter 1,800 mm, initial diameter 2,500 mm. Total reduction of the ingot is 110 mm. The forging around is accomplished by 5 reductions at 6 turns .](https://www.researchgate.net/profile/Lothar-Meyer-2/publication/268432694/figure/fig3/AS:669574848585748@1536650492703/Half-of-ingot-presentation-during-elongation-A-with-detailed-distribution-of-equivalent_Q320.jpg)
Content uploaded by Lothar W. Meyer
Author content
All content in this area was uploaded by Lothar W. Meyer on Aug 01, 2015
Content may be subject to copyright.
1/5
COMPUTER SIMULATION OF
TECHNOLOGICAL CHAIN BY FREE
FORGING
Bohuslav Mašek 1+3)
Zbyšek Nový 2+1)
Dušan Kešner 1)
Lothar - W. Meyer 3)
1) University of West Bohemia,
Pilsen, Czech Republic
2) Škoda Research Ltd.
Pilsen, Czech Republic
3) Chemnitz University of Technology,
Chemnitz, Germany
SUMMARY
Single part production and high material and energy
demands are typical characteristics of free forging of
large pieces. That is a reason why it is very difficult
to optimise technological processes by means of the
piloting plant experiments.
FEM - based numerical simulation of the forging
process is used for large forging production analysis.
The mass of forging can be up to 30 MT. By
achieving correct material modelling and complex
data conversion, it is possible to perform the simula-
tion of technological operation chains successfully.
The production of large one-piece crankshafts for big
diesel engines is one example for employing such a
technological operation chain simulation.
1. INTRODUCTION
This demonstration of 3D simulation of technological
processing of the free forging of a crankshaft on FEM
-base consists of a chain of upsetting, elongation and
forging of stroke operations. The main objective was
to analyse with the aid of numerical simulation the
production of a one-piece crankshaft and determine
the material behaviour under conditions of particular
technological steps and to rationalise production with
the assistance of identified facts. The simulation was
designed by Forge3 program and some partial
problems were solved in Forge2 and DEFORM
program for faster optimalization. The ingot of
34CrNiMo6 steel and mass 45MT was the initial
semi-product.
2. DESIGN OF BASE DATA FOR SIMULATION
The model geometry is created according to the
preforging dimensions in the HyperMesh program.
After this the geometry surfaces are discretized with
careful attention for continuity between particular
surfaces. The resulting continuing surface mesh of
particular parts (preforging and dies) is then exported
from HyperMesh into the Patran format. This file is
within the program Forge3 which is then imported
with an assistance of a subprogram Interg into own
data format. These data are the source for the
subprogram Meshv, which makes a 3D mesh of the
model. The mesh is composed from uniform
triangular elements with optimalized dimensions
dependent on model size and number of the ele-
ments required for exact precision of the solution.
The mesh is condensed in very key areas.
2.1 DESIGN OF DIE GEOMETRY AND MODEL
ASSEMBLING
The representation of dies is designed in the same
way as the model of preforging, just 3D meshing is
not present. Functional surfaces are represented by
triangular elements and their normal vectors against
normal vectors of preforging. Model assembling is
performed in the subprogram Creout. First of all, the
mesh of preforging is read, then the mesh of the dies
is read and these are adjusted into the initial position.
(Fig. 1)
2.2 KINEMATIC MOVEMENT DEFINITION
The movement definition is carried out in the
subprogram Creout. A direction of working move-
ment is assigned to the dies. Next, a stroke of moved
dies and a speed of dies movement are defined
dependent on the working machine. (Fig. 1) It is
possible to define not only a linear movement but, for
example rotation or any function of time as well.
Fig. 1: Kinematics design of adjusted model
of ingot and upsetting matrices
2/5
2.3 MATERIAL PARAMETERS
The domain of measurement and computation of
material data is briefly described as well as the
domain of used material mode [3-7] because it is
very important for accuracy of the calculation. For
description of a visco-plastic behaviour under high
temperature the Norton-Hoff law is used in this form:
σϕϕϕ
β
0
1
00
3=•+
+
() ( ) &
mnm
Ke
Τ
(1)
whereby
:
K0 = 0,370E+4 material constant term
m = 0,118 sensitivity to the strain rate
n = 0,186 sensitivity to the strain hardening
β = 0,258E-2 sensitivity to the temperature
The material data were computed from measured
data, in this case from the compression deformation
dilatometer. (Fig. 2) [9]
2.4 OTHER PARAMETERS
In general the other parameters are saved in the
program database and the nearest parameters to a
real state are always used. For instance, it acts on
model and coefficient of friction, temperature and
heat transfer properties into dies and initial
temperature of preforging. It is also possible to fix
any node of the mesh into displacement boundary
conditions. The period of remeshing, minimum
acceptable quality value of elements or frequency of
saving results are also employed.
3. SIMULATION OF UPSETTING
3.1 MODEL OF THE INGOT, DIES AND
UPSETTING MATRICES
The model of forging ingot I45, with initial length of
4023 mm with in average diameter of 1456 mm and
initial temperature 1150 °C, contained a total of 2476
nodes. Number of surface nodes is 1125 - on the
real ingot surface of 1,781.107 mm2. This number of
nodes was chosen with care of a good precision of
results and the computing time. To save material,
optimised-upsetting matrices instead of flat dies were
used. [8] (Fig. 1)
3.2 UPSETTING PARAMETERS
A real upsetting is realised on a hydraulic press,
which is characterised by a force of 100 MN. By
upsetting upper die moves by a velocity of 10 mm/s
in the direction of 0,0,-z, this velocity corresponds to
real state of upsetting through maximum press force
and the use of the multiplication unit. The press
stroke was 1300 mm: it corresponds to a final height
between dies 1253 mm. (Fig. 3) It is possible to
describe further parameters except for material
behaviour, for example the temperature of preforging
1150°C, the temperature of dies 250°C, the period of
remeshing after 30 steps, the quality of mesh at 0,4,
the characteristic element length 135mm and so on.
3.3 GRAPHICS INTERPRETATION OF
RESULTS OF UPSETTING
Temperature distribution on ingot surface after a
stroke of 886 mm and time of 88 s is illustrated as an
example in Fig. 3. Final shape of ingot after
upsetting is shown on Fig. 4 and Fig. 5.
4. SIMULATION OF ELONGATION
The elongation follows the upsetting process in the
technological chain and is composed of a set of
particular strokes, which step by step reduce the
diameter of preforging.
Fig. 2: Stress - Strain curves for 34CrNiMo6 steel
Fig. 3: Temperature distribution on ingot surface afte
r
stroke 866mm [K]
3/5
4.1 MODELS OF PREFORGING AND DIES
A deformed mesh is taken into the process with
temperature and strain distribution, which is a result
of upsetting simulation. The model has in total 6183
nodes and 2011 nodes on the surface. Our example
is accomplished with flat dies, which is intended for
the beginning of elongation. A disadvantage of these
dies is their high sensitivity ensuing from the
unsuitability of computation especially on the existing
inequality between interacting surfaces.
4.2 KINEMATIC PARAMETERS OF ELON-
GATION
The same hydraulic presses are used for elongation
and upsetting. The dies move symmetrically, one
against another at a velocity of 10 mm.s –1 and
direction (0, ±, 0). After one stroke is finished, the
dies return to the former position and the
semiproduct turns to an angle of 45°. The stroke is
repeated eight times, which represents one
revolution of the semiproduct. It is necessary, due to
the aforementioned inequality of surface to fix a
motion of semiproduct axes in the specified
directions x and y. In practice, it is solved by hanging
the semi-product.
Fig. 6: Half of ingot presentation during elongation (A) with detailed distribution of equivalent strain [ - ] in axis
section (B). Reduction 100mm, width of dies 550 mm, overlap of dies to stroke 100 mm, number of elements
approx. 17,500, total length 3,400 mm, diameter 1,800 mm, initial diameter 2,500 mm. Total reduction of the ingo
t
is 110 mm. The forging around is accomplished by 5 reductions at 6 turns .
Fig. 5: Representative problems of elongation A) start of the simulation B) slippage of the ingot at the die C) sepa-
ration from the backplate during working
Fig. 4: Comparison of ingot shape A) before and B) afte
r
first cycle of elongation operation
4/5
4.3 GRAPHIC INTERPRETATION OF COMPU-
TED RESULTS
The cited result (Fig. 4) represents the 66th step of
computation. This operation caused a total elonga-
tion of 6 mm to a momentary length of 3,143 mm. It
is a very small absolute change of the dimensions,
when considered from the standpoint of a real
technological process. Nevertheless, the manage-
ment of the complex data transfer during simulation
is indispensable. In addition, a number of other
problems had to be solved to preserve the stability of
computation, from a suitable mesh up to the
prevention of tripping the round semiproduct out of
the dies. (Fig. 5) The complete process of elongation
was simulated by this developed and evaluated
manoeuvre. (Fig. 6)
5. SIMULATION OF FORGING OF STROKES
Free forging of strokes of crankshaft belongs to the
most significant operations that influence the
success of production. That is why new processes
are sought out all the time to make the production
more effective and pre-eminent.
It is achieved by means of simple figures and
sophisticated processes with higher accuracy and
smaller material allowances. (Fig .7)
It is also feasible to use FEM simulation for the
design of non-conventional technological processes
that were designed and arise from brain-storming.
(Fig. 8)
6. CONCLUSION
The management of simulation of technological
operation chains during free forging is documented
by very different subsequent technological opera-
tions. They are representative of the forging of big
forging with characteristic longitudinal axis and a
shape complexity. Correct computation technology,
data design and transition between operations during
simulation are the most complicated problem areas.
(Fig. 9) Big amount of data transfers is involved. It is
necessary to use suitable programs and to interpret
partial results in the correct way.
Continuous control of the computation results with
reality is also essential. (Fig. 10)
Fig. 7: A multinumerous dies stroke study
Process of Simulation
Action Program File
Geometry cr eating and meshing
HyperMesh *.hm
Export of the mesh in exchange format
HyperMesh *.pat
Import of the mesh and quality control
Interg *.dou
3D Meshing
Meshv *.ma
y
Action Program File
Geometry cr eating and meshing
HyperMesh *.hm
Export of the mesh in exchange format
HyperMesh *.pat
Import of the mesh and quality control
Interg *.dou
3D Meshing
Meshv *.may
Model of the Ingot
Model of the Dies
Geometry cr eating and meshing
HyperMesh *.hm
Export of the mesh in exchange format
HyperMesh *.pat
Import of the mesh and quality control
Interg *.dou
Geometry cr eating and meshing
HyperMesh *.hm
Export of the mesh in exchange format
HyperMesh *.pat
Import of the mesh and quality control
Interg *.dou
Problem set up
Object Adjustment
Creout
Motion set up
Creout *.out
Object Adjustment
Creout
Motion set up
Creout *.out
Material Properties
Other Parameters
Computation
Results Visualisation
Results Analysis
Fig. 9: Illustration of technology simulation process
Fig. 8: Study of crankshaft forming by cross rolling
technology (A- before, B- after forming) [2]
Fig.10: Representative computation of one-piece
crankshaft. The length is 9 m, model 1:5. Total
length of the model is 1,800 mm. Computation
deviation to the real forging is less then 30 mm, i.e. it
represents error 1.6% and 5 mm deviation in one
stroke
5/5
ACKNOWLEDMENTS
This development has been supported by:
1) MŠMT ČR Project „250“ Modelling of Heteroge-
nous Functional Structured Materials, University
of West Bohemia, Pilsen, 1995 - 2000, CZ
2) Eureka project 1869 Forming, Optimisation of the
Forming of Special Alloys Products, Škoda
Research Ltd. EU
3) University of West Bohemia - West Bohemia
Supercomputer Centrum, CZ
4) DFG - SFB 283 „Prozeßketten der Massiv-
umformung“ TU Chemnitz 1995 - 2000, D
5) Škoda Steel Ltd. Pilsen, CZ
REFERENCES:
[1] Mašek, B.; Kešner, D.; Nový, Z.: Forge 3 -
Möglichkeiten am Beispiel eines Gesenkeschmiede-
stückes, in: conference MEFORM 99 Freiberg 1999,
D
[2] Mašek, B.; Nový, Z.; Kešner, D.: Possibilities of
FEM Simulation by Large Crankshaft Production, in:
international conference ICTP 99, Nurmberg 1999, D
[3] Mašek, B.; Hartwig, H.: Comparison of
Computer Simulation Results and the Experiment for
the Ring Compression Test, in: international
conference THER TECH FORM ´99, Tále 1999, SK
[4] Meyer, L. W.; Mašek B.: Ermittlung von
Werkstoffkenndaten für ein FEM Modell des Stahls
34NiCrMo6, Untersuchungsbericht UB 06/99 LWM
TU Chemnitz 1999, D
[5] Koucký, V.; Basl. J.; Mašek, B.: Rekonstrukce
elektrické vyhodnocovací části zařízení pro zkoušení
tvařitelnosti krutem, in: international symposium
Applied Electronics, Plzeň, 1995, CZ
[6] Koucký, V.; Basl. J.; Mašek, B.: Monitorování
zkoušky plasticity na zkrutovém plastometru, in:
Seminář metalurgů a technologů kováren s
mezinárodní účastí, ZČU Plzeň , 1995, CZ
[7] Mašek, B.; Nový, Z.; Koucký, V.; Doubek, J.;
Švantner, M.: Vyhodnocování tvařitelnosti a určování
reologických konstant pro program Forge, zpráva
projektu MŠMT ČR „250“ Modelování heterogenních
funkčně strukturovaných materiálů, ZČU Plzeň 1997,
CZ
[8] Mašek, B.; Kešner, D.: Analýza materiálového
toku v patní oblasti ingotu při pěchovacích operacích
pomocí metody konečných prvků, in: international
metallurgical symposium Metal 98, Tanger, Ostrava,
1998, CZ
[9] Meyer, L. W.; Mašek B.: Einsatz der
Umformdilatometrie zur Entwicklung technologischer
Prozesse des Umformens, in: 76. seminář
technologů a metalurgů volných kováren, Valašské
Klobouky - Jelenová, VTS Vítkovice, 5/2000, CZ