Available via license: CC BY 4.0
Content may be subject to copyright.
ARCHIVES OF METALLURGY AND MATERIALS
Volume 60 2015 Issue 2
DOI: 10.1515/amm-2015-0217
Z. GRONOSTAJSKI∗,], M. HAWRYLUK∗, M. KASZUBA∗, G. MISIUN∗, A. NIECHAJOWICZ∗, S. POLAK∗, M. PAWEŁCZYK∗∗
AN ANALYSIS OF THE INDUSTRIAL FORGING PROCESS OF FLANGE IN ORDER TO REDUCE THE WEIGHT OF THE INPUT
MATERIAL
ANALIZA PRZEMYSŁOWEGO PROCESU KUCIA ODKUWKI KOŁNIERZA W CELU ZMNIEJSZENIA MASY MATERIAŁU
WSADOWEGO
This paper presents an analysis of the industrial process of hot forging a flange. The authors developed several ther-
momechanical models of the forging process for which they carried out computer simulations using the MSC.Marc 2013
software. In the Jawor Forge flanges with a neck are manufactured by hot forging in crank presses with a maximum load of
25 MN. The input material, in the form of a square bar, is heated up to a temperature of 1150◦C and then formed in three
operations: upsetting, preliminary die forging and finishing die forging. The main aim of the studies and the numerical analyses,
in which the geometry of the tools would be modified, was to maximally reduce the amount of the input material taking into
account the capabilities of the Jawor Forge, and consequently to significantly reduce the production costs. Besides the Forge’s
equipment resources, the main constraint for modifications was the flange-with-neck forging standard which explicitely defines
the tolerances for this element. The studies, which included numerical modelling, infrared measurements and technological
tests, consisted in changing the geometry of the tools and that of the forging preform. As a result, the optimum direction for
modifications aimed at reducing the mass of the input material was determined. The best of the solutions, making it possible
to produce a correct forging in the Jawor Forge operating conditions, were adopted whereby the weight of the preform was
reduced by 6.11%. Currently research is underway aimed at the application of the proposed and verified modifications to other
flange forgings.
Keywords: numerical modeling, forging, optimisation
Artykuł dotyczy analizy przemysłowego procesu kucia matrycowego na gorąco odkuwki typu kołnierz. Autorzy publikacji
opracowali szereg modeli termomechanicznych procesu kucia, dla których przeprowadzili symulacje komputerowe z zastoso-
waniem oprogramowania MSC.Marc 2013. Przemysłowy proces wytwarzania kołnierzy z szyjką w kuźni Jawor wykonuje się
metodą kucia na gorąco na prasach korbowych pracujących z maksymalnym obciążeniem 25 MN. Materiał wsadowy w postaci
pręta kwadratowego jest nagrzewany do temperatury 1150◦C i następnie kształtowany z wykorzystaniem trzech operacji –
spęczania, matrycowania wstępnego oraz matrycowania wykańczającego. Głównym celem przeprowadzanych badań i analiz
numerycznych polegających na modyfikacji geometrii narzędzi było jak największe ograniczenie wielkości materiału wsadowego
przy jednoczesnym uwzględnieniu możliwości technologiczno-technicznych Kuźni Jawor. Miało to pozwolić na znaczącą reduk-
cję kosztów produkcji. Za największy ogranicznik przeprowadzanych modyfikacji poza parkiem maszynowym zakładu przyjęto
normę kucia kołnierzy z szyjką, która w jednoznaczny sposób definiuje tolerancje wykonywanego elementu. Przeprowadzone
badania polegające na zmianie geometrii narzędzi i przedkuwki, obejmujące modelowanie numeryczne, pomiary termowizyjne
oraz próby technologiczne pozwoliły określić optymalny kierunek wprowadzanych modyfikacji procesu umożlwiający redukcję
masy materiału wsadowego. Przyjęcie przez autorów najlepszego z możliwych do uzyskania w warunkach Kuźnia Jawor
rozwiązań pozwoliło na zmniejszenie masy wstępniaka o 6,11% w stosunku do masy wyjściowej. Obecnie trwają dalsze prace
badawcze związane z przełożeniem zaproponowanych i zweryfikowanych zmian na inne odkuwki kołnierzowe.
1. Introduction
Because of the increasingly free flow of goods and the
difficult situation on the market, forges put great emphasis on
competitiveness. According to the data contained in [2], in
2013 39% of the production came from China which country,
owing to relatively cheaper labour and less restrictive environ-
mental standards, poses serious competition to the domestic
plants. The simplest way to attract potential customers is to
ensure a low price of the product while maintaining its high
quality. These are the guidelines which the larger European
manufacturers follow [8]. It is estimated that over 50% of the
total product costs are material costs [4], about 20% of which
are material losses [6, 10, 11, 13]. Therefore the reduction
of material costs is regarded to be the most effective way of
reducing the price of forgings at a minimum change in their
∗WROCŁAW UNIVERSITY OF TECHNOLOGY, 25 WYBRZEŻE WYSPIAŃSKIEGO STR., 50-370 WROCŁAW, POLAND
∗∗ FORGE JAWOR S.A.
]Corresponding author: zbigniew.gronostajski@pwr.edu.pl
- 10.1515/amm-2015-0217
Downloaded from PubFactory at 08/07/2016 12:12:33AM
via free access
850
properties [1, 7]. Today IT tools based on computer simula-
tion play a major role in the design of not only engineering
constructions, but also industrial processes. As the power of
computing machines increases, software based on advanced
numerical methods, such as the finite element method, has
considerably developed in recent years [3, 5, 9, 12, 14, 15].
Today there is a relatively wide choice of software dedicated
to nonlinear problems, such as the plastic deformation of met-
als. The computer programs become indispensable tools for
engineers and can contribute to the improvement of the exist-
ing technological solutions. Thanks to the use of FEM-based
computing software packages by process and design engineers
the latter can quickly analyze forging processes and determine
their key features, such as: the distribution of stress and strain,
the way the material flows and the process force parameters.
Today forging plants in collaboration with scientific and re-
search centres (equipped with much more advanced IT tools
and laboratories) can optimize their industrial processes not
only to make them more efficient and to improve the quality
of the forgings, but also to lower the production costs.
2. Description of process
The hot forging of a flange with a Ø80 neck was select-
ed to be studied. Flange Ø80 is typical axisymmetrical part
used to connect pipes. Forging part drawing with exemplary
dimensions has been show on Figure 1.
Fig. 1. Final forging with exemplary dimensions
The forging process is conducted in a crank press with a
maximum pressure of 25 MN. The preform is heated up to a
temperature of 1150◦C and then shaped in three operations:
upsetting, preliminary die forging and finishing die forging.
Figure 2 shows the preforms after the first two operations, and
the finished product. In the analyzed process the main aim of
upsetting is to give the blank a preliminary shape to make
it easier to fill the die impression in the next operations and
remove the scale. The product gets the initial proper shape in
the course of preliminary die forging while finishing forging
is responsible for its sizing to the final dimensions (mainly
the radii of the roundings, and selected dimensions) consis-
tent with the customer’s specifications. The material properties
for steel C.22 for forging part were taken from the software
MSC.Marc 2013 database.
3. Numerical modelling of analyzed process
Computer simulations were run, using the MSC MARC
2013 software, in order to determine the principal parameters
of the forging process. A 3D model and Tetra 4 elements were
used in the simulation, which was divided into three stages
corresponding to the particular operations.
Fig. 2. Shape of forging after particular forging operations: a) upset-
ting, b) preliminary die forging c) finishing die forging
The tools were assumed as stiff bodies with constant tem-
perature of 300◦C and according to symmetry of process the
analysis was limited to 1/8 of the model. The constant friction
factor of 0.35 has been assumed based on author experience.
Heat transfer coefficient between workpiece and tool has been
set to 20000 W/(m2*K) and between workpiece and environ-
ment to 35 W/(m2*K). Such factors as: the temperature of the
material, the plastic strains in the forging, the load of the press,
the shape of the flash and the filling of the die impressions,
were taken into consideration.
Figure 3 shows load press versus press slide for the par-
ticular operations obtained in modelling. During preliminary
forging and finishing forging the press forces increase sharply
at the final moment of filling the die impression, reaching 25
MN. In the case of upsetting, the increase is much gentler and
it does not exceed 3 MN. It is apparent from the diagram that
the forces do not significantly exceed the maximum load of
crank press, which means that the original process has been
correctly designed for the press with a nominal force of 25
MN. The characteristic raggedness of the upper part of the
diagram at the maximum load is due to the use a remeshing
algorithm.
Fig. 3. Load of press versus distance from bottom dead centre
Also the distribution of the temperature in the forging was
modelled and towards the end of the second operation a con-
siderable cooling was observed in the lower part of the neck
(the area marked in Fig. 4), where the temperature reaches
- 10.1515/amm-2015-0217
Downloaded from PubFactory at 08/07/2016 12:12:33AM
via free access
851
540◦C. It was found that the drop in temperature may signif-
icantly contribute to the increase in the forces in the second
and third operations and result in the incomplete filling of the
die impression. This was confirmed by observations of the real
forgings in which underfills would occur in those places.
Fig. 4. Distribution of temperature in forging after second operation,
with marked area of greatest overcooling
A considerable increase in the length of the flash was also
observed between the second operation and the third opera-
tion, which is due to the large differences between the shapes
of the impressions, leading to the considerable flow of the
material in the third operation (intended to perform mainly
the calibrating function) (Fig. 5).
Fig. 5. Flash length after: a) preliminary die forging, b) finishing die
forging
4. Assumptions for modifications
The main aim of the modifications was to maximally
reduce the amount of input material in order to lower the pro-
duction costs. Besides the technical capabilities of the Jawor
Forge, the tightest constraint on any modifications was stan-
dard PN-1092-1 for the forging of flanges with a neck, which
explicitly defines the tolerances for this element.
It was decided to reduce selected forging dimensions and
the maximum forging force. For this purpose the temperature
of the flange neck was to be increased by lowering the bottom
and reducing the angle of inclination of the bottom knock-outs
and the thickness of the bottom (functioning as an internal
flash) situated in the centre of the forging. In addition, in order
to maximally reduce the flash forming in the third operation,
the difference in the die inserts between the second operation
and the third operation was reduced.
5. Modifications made
In order to gain maximum insight into the effect of the
modifications on the process the former were introduced grad-
ually and it was analyzed how each of them changed the forg-
ing conditions. Figure 6 schematically shows the original and
modified profiles of the impressions of the tools used for pre-
liminary forging and finishing forging.
Fig. 6. Comparison of shape of die impressions before and after
modification for preliminary die forging and finishing die forging
(modified dies are in red)
First the angle of inclination (α) of the bottom knock-outs
was reduced by 4 degrees for both preliminary forging and
finishing forging. Considering that the temperature of the ma-
terial being in contact with the knock-outs is relatively high,
this modification should not have a significant effect on the
load of the press while the presence of the knock-outs in the
two operations should prevent the forgings from wedging on
the inserts. The obtained results confirmed the insignificant
effect of the inclination of the load of the press (Fig. 7).
Fig. 7. Comparison of press load as function of distance from bottom
dead centre for preliminary die forging (a) and finishing die forging
(b)
Then the bottom was shifted by 10 mm towards the neck
and its thickness was reduced by 4 mm (denoted as X and G
in figure 6). The aim was to reduce the amount lost as flash
and at the same time to increase the temperature of the most
overcooled point, which was to compensate for the increase in
load resulting from the thinning of the bottom. Owing to this
modification the area with lowered temperature was reduced
and its temperature was increased by about 50◦C. A compar-
ison of the effect of the modifications on the distribution of
temperature is shown in Fig. 8.
- 10.1515/amm-2015-0217
Downloaded from PubFactory at 08/07/2016 12:12:33AM
via free access
852
Fig. 8. Distribution of temperature in lower part of forging after
finishing operation for a) model before modification and b) model in
second stage of modification
The change in bottom thickness with the simultaneous
increase of neck temperature had a beneficial effect on the
load of the press (fig. 9) and together with the change of
the inclination angle resulted in both an overall reduction in
the weight of the charge by 3.98% and a reduction of the
maximum force loading the press.
Fig. 9. Comparison of press load as function of distance from bottom
dead centre for preliminary die forging (a) and finishing die forging
(b) in second stage of modification
The final modification consisted in reducing the thickness
of the flange in the second operation by 0.7 mm (denoted as
Y3in the figure). The aim was to reduce the difference in
shape between preliminary forging and finishing forging and
thereby limit excessive flash after the third operation. In order
to prevent an excessive increase in the force the bridge opening
was increased for the two die forging operations. In addition,
because of the intensive wear of the bridge during preliminary
forging its width was increased by 5 mm.
After a thorough analysis of the degrading mechanisms
which occur in these tools a decision was made to slightly
increase the thickness of the bottom of the die inserts in order
to prevent their excessive abrasive wear and fracturing. The
totality of the changes is denoted with the letter Y in the
figure.
Owing to bridge opening the increase in the share of
forming in preliminary forging did not contribute to any in-
crease in the load of the press in this operation whereby
the forces in the final operation decreased considerably: after
the modifications their maximum value does not exceed 20
MN (Fig. 10). Moreover, the difference in length between the
second- and third-operation flashes was reduced whereby the
total charge weight was reduced by 6.11%.
Fig. 10. Comparison of press load versus distance from bottom dead
centre for preliminary die forging (a) and finishing die forging (b) in
third stage of modification
6. Verification of modelling by experiment
Forging tests (taking into account the proposed modifi-
cations) in industrial conditions were carried out in order to
verify the results of the FEM analysis. In case of computing
errors, preforms with an ever smaller mass, beginning with the
nominal mass specified in the technology, were prepared for
the tests. In the course of the technological tests the tempera-
ture of the tools was controlled by means of a thermal imaging
camera so that it was possible to determine the temperature
of forgings in various areas (Fig. 11). On the presented ther-
mograms the temperature in subsequent forging operations is
rapidly declining, which is confirmed by the results of numer-
ical modeling (Fig. 8).
Fig. 11. Thermogram from thermal imaging camera after particular forging operations: a) upsetting, b) preliminary die forging c) finishing
die forging
- 10.1515/amm-2015-0217
Downloaded from PubFactory at 08/07/2016 12:12:33AM
via free access
853
In the tests full tool impression filling was obtained for
each of the preform masses. Moreover, no other irregularities
(e.g. material jamming) were found. The shape of the flanges
with a flash after the second and third operation is compared
in Fig. 12.
Fig. 12. Comparison of shape of upper part of flange with flash after
second operation (top row) and third operation for for every mass
lowering (in kilograms)
The dimensional and shape examination of the finished
product revealed that in some cases a burr had formed in
the upper part of the flange neck, probably as a result of the
lowering of the bottom (leading to material burring during die
shearing, Fig. 13). But this defect does not affect the quality of
the product and it would occur also in the “original process”
and the place where it occurs is anyway machined by turning.
Fig. 13. Burr forming as result of bottom shearing
7. Conclusion
The main parameters of the forging process were deter-
mined through FEM-based numerical modelling and techno-
logical tests. The key factors having influence on the load and
the final shape of the product were identified and on this basis
several modifications were gradually introduced whereby their
effectiveness and effect on the whole process could be stud-
ied. The best of the solutions, making it possible to produce a
correct forging in the Jawor Forge operating conditions, were
adopted whereby the weight of the preform was reduced by
6.11%.
However, due to the imperfection of the finite element
method and to the adopted simplifications it was necessary
to carry out a verification, which confirmed the satisfactory
effectiveness of the modifications. Full filling of the die im-
pression was obtained for each of the masses.
Still, many important factors, such as the stresses gen-
erated in the tools and the wear of the latter, were not taken
into account. The modelling and the tests merely indicated the
direction of the introduced modifications is correct. In order
to fully validate the modifications they need to be implement-
ed in the actual manufacturing process and then their mass
production effect should be analyzed.
Currently research is underway aimed at the further mod-
ification of the process, using the Rozenbrock optimization
method, and at the application of the proposed and verified
modifications to other flange forgings.
Acknowledgements
This research has been funded under the project entitled: “De-
velopment and implementation of precise forging in Jawor Forge
PLC” being carried out under priority 1 – Research and development
of modern technologies, Measure 1.3 – Support for R&D projects for
entrepreneurs carried out by scientific entities, Submeasure 1.3.1 –
Development projects, the Operational Programme Innovative Econ-
omy, in the years 2007-2013, co-funded from the European Regional
Development Fund.
REFERENCES
[1] N.R. Bhutta, N.R. Chitkara, Int. J. Mech. Sci. 38, 891-916
(1996).
[2] http://www.euroforge.org/fileadmin/user upload/Downloads/
World-tons-2013.pdf
[3] M. Geiger, Cirp. Ann.-Manuf. Techn. 44, 581-588 (1995).
[4] W.A. Knight, Int. J. Adv. Manuf. Tech. 7, 159-167 (1992).
[5] R. Kopp, J. Mater. Process. Tech. 60, 1-4 (1996).
[6] G. Samołyk, Eksploat. Niezawod. 22, 50-54 (2004).
[7] V. Vazquez, B. Painter, T. Altan, T. Takemasu, J. Mater. Process.
Tech. 59, 95-105 (1996).
[8] http://www.euroforge.org/fileadmin/user upload/eLibrary/ Ex-
tra Info 44 Meist besser und selten teurer.pdf
[9] S.H. Zhang, Z.R. Wang, Z.T. Wang, Y. Xu, K.B. Chen, J. Mater.
Process. Tech. 151, 39-47 (2004).
[10] M. Hawryluk, M. Marciniak, G. Misiun, Eksploat. Niezawod.
16, 600-607 (2004).
[11] Z. Gronostajski, M. Hawryluk, J. Krawczyk, M. Marciniak. Ek-
sploat. Niezawod. 13, 129-133 (2013).
[12] J. Tomczak, Z. Pater, T. Bulzak, Arch. Civ. Mech. Eng. 14,
104-113 (2014).
[13] Z. Gronostajski, M. Kaszuba, M. Hawryluk, M. Zwierzchowski,
Arch. Civ. Mech. Eng. 14, 528-539 (2014).
[14] D. Pociecha, B. Boryczko, J. Osika, M. Mroczkowski, Arch.
Civ. Mech. Eng. 14, 376-382 (2014).
[15] P. Preś, W. Skoczyński, K. Jaśkiewicz, Arch. Civ. Mech. Eng.
14, 622-635 (2014).
Received: 20 April 2014.
- 10.1515/amm-2015-0217
Downloaded from PubFactory at 08/07/2016 12:12:33AM
via free access
