materials

Article
Determination of Actual Friction Factors in Metal
Forming under Heavy Loaded Regimes Combining
Experimental and Numerical Analysis
Ana María Camacho 1, *, Mariano Veganzones 1 , Juan Claver 1 , Francisco Martín 2 ,
Lorenzo Sevilla 2 and Miguel Ángel Sebastián 1
1 Department of Manufacturing Engineering, Universidad Nacional de Educación a Distancia (UNED),
Madrid 28040, Spain; mveganzon6@alumno.uned.es (M.V.); jclaver@ind.uned.es (J.C.);
msebastian@ind.uned.es (M.Á.S.)
2 Department of Civil, Material and Manufacturing Engineering, University of Malaga, Malaga 29071, Spain;
fdmartin@uma.es (F.M.); lsevilla@uma.es (L.S.)
* Correspondence: amcamacho@ind.uned.es; Tel.: +34-913-988-660

Academic Editor: Magd Abdel Wahab

Received: 8 August 2016; Accepted: 29 August 2016; Published: 1 September 2016

Abstract: Tribological conditions can change drastically during heavy loaded regimes as experienced
in metal forming; this is especially critical when lubrication can only be applied at the early stage of
the process because the homogeneous lubricant layer can break along the die-workpiece interface.
In these cases, adopting a constant friction factor for the lubricant-surface pair may not be a valid
assumption. This paper presents a procedure based on the use of dual friction factor maps to
determine friction factors employed in heavy loaded regimes. A finite element (FE) simulation is
used to obtain the friction factor map for the alloy UNS A96082. Experiments were conducted using
four lubricants (aluminum anti-size, MoS2 grease, silicone oil, and copper paste) to determine the
actual friction curves. The experimental procedure is based on the application of lubricant only at the
beginning of the first stage of ring compression, and not at intermediate stages as is usual in typical
ring compression tests (RCTs). The results show that for small reductions (rh < 20%), the conventional
RCT can be applied because the tribological conditions remain similar. For large reductions
(rh > 20%), it is recommended to obtain an average value of the friction factor for every
lubricant-surface pair in the range of deformation considered.

Keywords: friction; lubricant; FE; tribology; metal alloys; forming; large deformations; experiments

1. Introduction
Friction reduction has typically been investigated due to its importance in mechanical processes
performance, systems performance, and wear prevention. An effective technique for improving
tribological performance is surface texturing, as demonstrated by Lu et al. in [1] where friction
reduction is achieved by generating square dimples of different sizes and geometries at the contact
surface. New coatings to prevent wear under severe conditions are also being investigated, as observed
in the work from Vandoni et al. [2], where the use of fiber laser sources for surface texturing of very thin
TiN coatings is presented as a good option for heavy loaded sliding regimes. Reduction of friction and
wear has been proposed in the recent work of Yazawa et al. [3] through a hybrid tribofilm consisting of
both coating and lubricant. In metal forming, friction is a very complex phenomenon due to the variety
of technological factors involved and their own interrelation. Not only friction reduction, but also
friction characterization has grown recently because tribological conditions between workpieces and
tools heavily influence material flow and tool life [4], as well as required loads, energy consumption,
surface quality, and the internal microstructure of the products obtained by plastic deformation.

Materials 2016, 9, 751; doi:10.3390/ma9090751 www.mdpi.com/journal/materials

Materials 2016, 9, 751 2 of 16

Recent studies published by Hua et al. [5] and Zhang and Ou [6] show the enthusiastic interest of
the scientific community to develop new methods for friction characterization and to discover new
relationships between classical friction models, respectively. The most widely accepted methodology
to characterize friction between two surfaces is to define a friction factor at the work piece-die interface.
To evaluate the behavior of lubricants in metal forming processes, it is common to quantify the friction
factor using different indirect techniques, such as the open-die backward extrusion test [7], double cup
extrusion test [8], or ring compression test, among others. The ring compression test (RCT) has been
widely used and successfully applied since its initial development by Kunogi [9]; this method was
improved by the well-known work of Male and Cockcroft [10]. After its publication, many researchers
have subsequently justified its validity; one example is the work of Hawkyard and Johnson [11]
that analyzed the problem assuming homogeneous deformation, neglecting strain hardening, and
assuming a constant Tresca friction factor at the interface. Male performed a study to determine the
variations of the friction factor of metals during the compression processes at room temperature [12],
and later on, the same author investigated the applicability of the RCT to conventional metal forming
processes [13]. Carter and Lee [14] simulated the RCT using a finite element model, generating the
friction factor curves for a particular material, and they observed some differences compared to the
curves obtained by Male and Cockcroft in their 1965 work. These differences were related to the
assumption of homogeneous strain and a constant value of the friction factor at the workpiece-die
interface. Some years later Wang and Lenard [15] concluded that the temperature and strain rate were
the two most influential parameters on the Tresca friction factor form results of hot RCTs. Other works
have attempted to determine the dependence of friction on the material properties, strain rate, and
non-homogeneous deformation, using both metallic [16] and non-metallic materials [17]. From these
works it can be concluded that the RCT is an effective method for the determination of the friction
factor in the metal forming processes. However, it is inadvisable to use generic friction maps and tables
independent of the type of material and operating conditions, as demonstrated by several authors
in the scientific literature [7,18]. Particularly, during any transient metal forming operation, such us
forging, tribological conditions change during the process; this is especially critical when lubrication
can only be supplied at the early stage of the process and heavily loaded conditions are applied due
to the extreme changes during the forming process that occur to the lubricant layer at the interface.
In these cases, the consideration of the initial friction factor for the lubricant-surface pairs applied
to the entire operation can lead to an underestimation of the required forces and, consequently, of
the required energy to finish the operation. To determine the actual friction factors found during
these operating conditions, numerical simulation is required as a complementary analysis tool to
experiments. The Finite Element Method (FEM) is widely used in metal forming simulation because its
analysis capability has been probed under large deformations. The combination of experimental tests
with analytical and numerical techniques has become the most powerful methodology for researchers
in metal forming. As an example, Shahriari et al. [19] have studied the hot RCT of the superalloy
Nimonic 115 by combining simulation techniques and experimental tests while using a profile projector
as a method of measurement of ring dimensions. Zhu et al. [20] also determined the friction factor
of Ti-6Al-4V titanium alloy in hot forging by combining RCT results and finite element simulations.
Other analytical techniques such as the Upper Bound Theorem (UBT) also have considerable potential,
as demonstrated by Bermudo et al. in their recent work [21]. This paper presents an alternative use
of the RCT (by means of dual friction factor maps) to determine friction factors more adapted to
heavy loaded regimes that occur during metal forming by combining experiments with numerical
simulations by FEM.

2. Materials and Methods

2.1. Approach
In this work, a combination of experiments with numerical simulations by the FEM has been
selected to perform the analysis and the procedure followed is outlined in Figure 1.
Materials
Materials2016, 9, 9,
2016, 751751 3 of 1616
3 of

Figure 1. Outline of the methodology. RCT refers to ring compression test.

Figure 1. Outline of the methodology. RCT refers to ring compression test.

This
Thismethodology
methodologywill
willbebe
explained inin
explained detail hereafter,
detail and
hereafter, it comprises
and thethe
it comprises following:
following:
•  Experimental
Experimentaldetermination
determinationofofthe the plastic
plastic flow curve for for the
thematerial
materialUNSUNSA96082
A96082used used in in
the
theRCTs.
RCTs.
•  Finite
Finite Element
Element modelling
modelling toto simulate
simulate RCTs,
RCTs, considering
considering thethe flow
flow curve
curve previously
previously obtained.
obtained.
•  Experimental
Experimental performance
performance of RCTs
of RCTs according
according to thetoalternative
the alternative procedure
procedure presented
presented in this
in this paper.
paper.
This This procedure
procedure is based isonbased on the application
the application of lubricant
of lubricant only at the onlybeginning
at the beginning of the
of the first first
stage
stage during the compression of the rings and not at intermediate
during the compression of the rings and not at intermediate stages as usual in typical RCTs. stages as usual in typical
• RCTs. representation of the dual friction factor map, which includes both the friction curves
Graphical
 obtained
Graphicalby representation
experiments inof the dual friction
laboratory factor map,simulations.
and by numerical which includes both the friction curves
• Selection of the friction factor according to the deformationsimulations.
obtained by experiments in laboratory and by numerical stage.
 Selection of the friction factor according to the deformation stage.
2.2. Determination of the Plastic Flow Curve
2.2. Determination of the Plastic Flow Curve
Before starting the numerical simulation, the flow curve of the material UNS A96082 (Sanmetal
Before starting
SA, Zaragoza, the numerical
Spain) used simulation,
in the experimental thehad
tests flow
to curve of the material
be determined. UNSCorrection
The Bulge A96082 (Sanmetal
Factor
SA, Zaragoza,
Method (BCFM) [22]Spain)
wasused
usedintothe experimental
obtain the plastic tests had toofbe
flow curve thedetermined.
aluminum alloyThe UNS
BulgeA96082
Correction
by
Factor
means ofMethod (BCFM)
compression tests[22]
underwascontrolled
used to obtain the plastic
conditions flow curveDuring
in the laboratory. of theuniaxial
aluminum alloy UNS
compression
A96082
tests, by meansundergoes
the specimen of compression tests under once
plastic deformation controlled
the yieldconditions in the laboratory.
stress is achieved, During
after the elastic
uniaxial
regime [23].compression
In the plastictests,
regime,thethere
specimen undergoes
is a uniform plastic at
deformation deformation once the friction
first; afterwards, yield stress
at the is
achieved, after the elastic
workpiece-compression regime
dies’ [23]. causes
interfaces In the plastic regime, there
a non-uniform is a uniform
deformation, deformation
so a correction at first;
factor is
required to calculate
afterwards, the flow
the friction atstress
the in correctly [24].
workpiece-compression dies’ interfaces causes a non-uniform
To accomplish
deformation, so a this goal, controlled
correction formingto
factor is required under quasi-non-friction
calculate the flow stressconditions was
in correctly performed
[24].
in the laboratory.
To accomplishDuringthisthe compression
goal, controlled of each specimen,
forming appropriate measurements
under quasi-non-friction conditionswere gathered
was performed
toin
obtain the flow curve
the laboratory. according
During to the BCFM. of
the compression In Figure 2, both curves
each specimen, (before and
appropriate after application
measurements were
ofgathered
BCFM) are represented.
to obtain the flow curve according to the BCFM. In Figure 2, both curves (before and after
application of BCFM) are represented.
Materials 2016, 9, 751 4 of 16

Figure 2. Plastic flow curve of the Aluminum alloy UNS A96082.

To characterizethe
To characterize thematerial
materialandand to implement
to implement its plastic
its plastic properties
properties into
into the the element
finite finite element
model,
model, the flow curve according to Hollomon’s expression is presented (Equation
the flow curve according to Hollomon’s expression is presented (Equation (1)): (1)):
 (MPa) = 500·0.1 0.1 (1)
σ (MPa) = 500·ε (1)
where K = 500 MPa is the strength coefficient and n = 0.1 is the strain hardening exponent. They have
where K = 500 MPa
been determined as isa the strength
result coefficient
of potential andofn the
fitting = 0.1 is the
data strain hardening
obtained exponent.
in the plastic zone ofThey the
have been determined
stress-strain as a result
curve by uniaxial of potential
compression testsfitting of the data
of cylindrical obtained in the plastic zone of the
specimens.
stress-strain curve by uniaxial compression tests of cylindrical specimens.
2.3. Lubricants
2.3. Lubricants
Lately, there is a growing interest in the use of green lubricants due to environmental concerns
Lately,
[25,26]. there is
Lubricants areacommonly
growing interest
applied in the form
in the use ofof liquid
green or lubricants
solid films due to environmental
at the die-workpiece
interface to minimize adhesion, the interaction between surfaces and, therefore,solid
concerns [25,26]. Lubricants are commonly applied in the form of liquid or films
friction at The
[27]. the
die-workpiece interface to minimize adhesion, the interaction between
most common lubricants in use are oils and greases. Oils are basically composed of a base oil and surfaces and, therefore,
friction [27].
specific The most
additives. These common lubricants
additives in use
are added toare
theoils
baseandoilgreases. Oils the
to provide are basically
lubricantcomposed of a
its properties
and performance characteristics. On the other hand, grease comprises a base oil, additives, andits
base oil and specific additives. These additives are added to the base oil to provide the lubricant a
properties
thickener. Theand thickener
performance maycharacteristics.
be any material Onthat,
the inother hand, grease
combination withcomprises a base
the base oil, willoil, additives,
produce the
and a thickener.
solid to semi-fluidThe thickener
structure.may Thebemain
any material
thickeners that,used
in combination
in greases with the base
include oil, will
lithium, produce
aluminum,
the solid to semi-fluid structure. The main thickeners used in greases include
calcium soaps, and clay, either alone or in combination. Lithium soap is the most common thickener lithium, aluminum,
calcium
in soaps, and
use today. clay, either
However, manyalone or in combination.
applications use solidLithium soapas
lubricants is the mostcase
in the common thickener in
of molybdenum
use today. However,
disulphide manynitride,
(MoS2), boron applications use solid lubricants(PTFE)
polytetrafluoroethylene as in theor case
Teflon,of molybdenum
and graphite.disulphide
These are
(MoS2 ), boron nitride, polytetrafluoroethylene (PTFE) or Teflon, and graphite. These are mainly used
mainly used in warm and hot forming where liquid lubricants are not recommended. These
in warm and hot forming where liquid lubricants are not recommended. These lubricants are widely
lubricants are widely used in the metal-mechanical industry [28].
used in the metal-mechanical industry [28].
To simulate different friction conditions during experimental tests, four lubricants (namely
To simulate different friction conditions during experimental tests, four lubricants (namely
aluminum anti-size, MoS2, silicone oil, and copper paste) were used and characterized by means of
aluminum anti-size, MoS2 , silicone oil, and copper paste) were used and characterized by means of
the RCT. Aluminum anti-size has excellent lubricating, anticorrosive, and anti-seize properties due
the RCT. Aluminum anti-size has excellent lubricating, anticorrosive, and anti-seize properties due to
to the aluminum structure that provides excellent behavior in high temperature conditions (until 600 °C).
the aluminum structure that provides excellent behavior in high temperature conditions (until 600 ◦ C).
It is recommended to prevent early wear of surfaces. MoS2 grease contains a mineral oil as a base, an
It is recommended to prevent early wear of surfaces. MoS2 grease contains a mineral oil as a base, an
organic thickener, and additives for use in high pressure conditions; it is widely used in cases where
organic thickener, and additives for use in high pressure conditions; it is widely used in cases where
oscillations, vibrations, and impact loads are encountered under moderate temperatures. Silicone oil
oscillations, vibrations, and impact loads are encountered under moderate temperatures. Silicone oil
can be used for metallic and non-metallic components; it is characterized by a low viscosity level and
can be used for metallic and non-metallic components; it is characterized by a low viscosity level and a
a high resistance against decomposition by heat. The last lubricant is the copper paste; it is typically
high resistance against decomposition by heat. The last lubricant is the copper paste; it is typically
used to prevent wear by corrosion in high temperature and high load applications.
used to prevent wear by corrosion in high temperature and high load applications.
Materials 2016, 9, 751 5 of 16
Materials 2016, 9, 751 5 of 16

2.4. Finite Element Model

2.4. Finite Element
A finite Model
element model has been developed in DEFORM F2™ (Scientific Forming Technologies
Corporation, Columbus,
A finite element OH,has
model USA)
beentodeveloped
accomplish in the goal ofF2™
DEFORM creating a friction
(Scientific factorTechnologies
Forming map for the
specific material UNS A96082. This FE code is a computer aided engineering
Corporation, Columbus, OH, USA) to accomplish the goal of creating a friction factor map software specifically
for the
designed for metal forming analysis. The DEFORM (Scientific Forming Technologies
specific material UNS A96082. This FE code is a computer aided engineering software specifically Corporation,
Columbus,for
designed OH, USA)
metal preprocessor
forming analysis.uses
Thea DEFORM
graphical user interface
(Scientific to assemble
Forming the dataCorporation,
Technologies required to
Columbus, OH, USA)
run the simulation. preprocessor
Input uses a graphical user interface to assemble the data required to run
data includes:
the simulation. Input data includes:
 Object description: all data associated with an object, including geometry, mesh, temperature,
• Object
material,description:
etc. all data associated with an object, including geometry, mesh, temperature,
 material, etc.
Material data: data describing the behavior of the material under the conditions which it will
• Material
reasonably data: data describing
experience the behavior of the material under the conditions which it will
during deformation.
 reasonably experience during deformation.
Inter object conditions: describes how the objects interact with each other, including contact and
• Inter object
friction conditions:
between objects.describes how the objects interact with each other, including contact and
friction between objects.
 Simulation controls: definition of parameters such as discrete time steps to model the process.
• Simulation controls: definition of parameters such as discrete time steps to model the process.
The main concepts of the preprocessing stage are explained in detail hereafter.
The main concepts of the preprocessing stage are explained in detail hereafter.
The geometrical relationships of the dimensions of the rings and the operating parameters
The geometrical relationships of the dimensions of the rings and the operating parameters
involved in the process were based on those used by Sofuoglu and Rasty in their tests [17]. The most
involved in the process were based on those used by Sofuoglu and Rasty in their tests [17]. The
common
most dimensional
common ratio used
dimensional ratio in thisintype
used thisof problem
type is the relation
of problem between
is the relation outer diameter:
between inner
outer diameter:
diameter:
inner height,
diameter: 6:3:2,6:3:2,
height, respectively,
respectively,commonly
commonly named
named thethe“canonical
“canonicalaspect
aspect ratio”. Specific
ratio”. Specific
dimensions used in the rings tested are shown
dimensions used in the rings tested are shown in Figure 3.in Figure 3.

Figure
Figure 3.
3. Geometry
Geometry of
of the
the rings
rings tested
tested by
by RCT.
RCT.

Symmetry conditions are typically applied in the literature about RCTs due to the axisymmetric
Symmetry conditions are typically applied in the literature about RCTs due to the axisymmetric
nature of this problem. This can be observed in the finite element modelling of RCT in works of
nature of this problem. This can be observed in the finite element modelling of RCT in works of
reference such
reference such as
as the
the ones
ones realized
realized byby Sofuoglu
Sofuoglu et et al.,
al., where
where bi-dimensional
bi-dimensional models
models using
using finite
finite
element codes
element codesANSYS
ANSYS [7] Abaqus
[7] and and Abaqus[16,17] [16,17]
have been have been developed.
developed. ConsideringConsidering this, an
this, an axisymmetric
axisymmetric
model modelwith
was created wasthe
created
finitewith the finite
element element
software DEFORM software
F2™ DEFORM F2™
(Scientific (Scientific
Forming Forming
Technologies
Technologies Columbus,
Corporation, Corporation, Columbus,
OH, OH, USA);
USA); symmetry symmetry
conditions conditions
were imposed, were imposed,
so only half ofsothe
only half
model
of the
was model was analyzed.
analyzed.
Flat platens
platens are
aremodelled
modelledasasrigid
rigidparts
partsand thethe
and workpiece
workpieceis modelled as aasdeformable
is modelled a deformable body. The
body.
workpiece
The workpiece has been meshed
has been with with
meshed first order continuum
first order continuumelements of quadrilateral
elements shapeshape
of quadrilateral and the mesh
and the
contains approximately 2200 elements.
mesh contains approximately 2200 elements.
Regarding the
Regarding the material,
material, each
each ring
ring has
has been modelled with
been modelled with aluminum
aluminum alloy
alloy UNS
UNS A96082,
A96082, whose
whose
plastic flow curve was determined previously from the compression test, as explained above. The flow
plastic flow curve was determined previously from the compression test, as explained above. The
stress data are introduced as tabular data because this is the most highly recommended method to
flow stress data are introduced as tabular data because this is the most highly recommended method
follow the true behavior of the material:
to follow the true behavior of the material:
 
 . 
 σ= σ, ε,,ε,TT (2)
(2)
where
where
Materials 2016, 9, 751 6 of 16

Materials 2016, flow

σ: effective 9, 751 stress, 6 of 16

ε: effective plastic strain,


.
ε: : effective
effective flowrate,
strain stress,
 work
T: temperature
: effective plastic strain,
 : effective
In this paper
straincold
rate,forming conditions are considered, so the flow stress does not depend on the
strain as the temperature is considered constant and equal to 20 ◦ C.
ratetemperature
T : work
A constant
In this paper friction model orconditions
cold forming shear friction model was assumed
are considered, for stress
so the flow a range of friction
does factors,
not depend on“m”,
the
from 0 to 0.6. This friction model considers a friction factor, m,
strain rate as the temperature is considered constant and equal to 20 C.to quantify the interface friction and its
analytical expression
A constant is (Equation
friction model or (3)):
shear friction model was assumed for a range of friction factors,
“m”, from 0 to 0.6. This friction model considers τ =a m ·k
friction factor, m, to quantify the interface friction(3)
and its
Thisanalytical expression
model assumes thatisfriction
(Equation (3)):is constant and it only depends on the shear flow stress,
stress
k. For example, for perfect lubrication (m =0) friction m k stress is null, whereas for sticking conditions (3)
(m = This
1), friction
model stress
assumesequals
that the shear
friction flowisstress.
stress constant Thisandmodel has
it only been demonstrated
depends to bestress,
on the shear flow more
realistic
k. than Coulomb’s
For example, for perfectfriction model(m
lubrication in =metal forming
0) friction analysis
stress because
is null, normal
whereas pressures
for sticking are often
conditions
higher
(m = 1),than flowstress
friction stresses so the
equals theCoulomb’s
shear flow friction
stress.model provides
This model hasfriction stresses highertothan
been demonstrated shear
be more
flow stresses.
realistic than Coulomb’s friction model in metal forming analysis because normal pressures are
oftenRegarding
higher than simulation controls,
flow stresses so theDEFORM
Coulomb’s F2™ is a numerical
friction code offriction
model provides implicit methodology
stresses that
higher than
uses the
shear flowNewton-Raphson
stresses. method for solving the equations. The model includes 200 steps and the
step Regarding
increment is defined ascontrols,
simulation 10. The DEFORM
number ofF2™ stepsisisa given by Equation
numerical code of (4):
implicit methodology that
uses the Newton-Raphson method for solving the equations.
x The model includes 200 steps and the
n =
step increment is defined as 10. The number of steps (4)
V · is
∆tgiven by Equation (4):
x
where n (4)
V  t
where
n: number of steps,
n:
x: number of steps,of the primary die,
total movement
x: total movement of the primary die,
V: primary die velocity,
V:
∆t:primary die increment
is the time velocity, per step
t: is the time increment per step
The
The completed
completed model
model after
after preprocessing
preprocessing is
is presented
presented in
in Figure
Figure 4,
4, where contact
contact nodes
nodes at the
beginning of the compression process are shown at the die-workpiece interfaces.
interfaces.

Figure
Figure 4.
4. Finite
Finite Element
Element model
model by
by DEFORM
DEFORM F2™
F2™ after
after preprocessing.
preprocessing.

The deformed rings

The deformed rings after
aftercompression
compressionand andextreme
extremefriction
friction conditions
conditions areare shown
shown in Figure
in Figure 5b
5b for
for comparison.
comparison. As As explained
explained in [18],
in [18], whenwhen a ring
a ring preform
preform is compressed
is compressed in the
in the plastic
plastic fieldfield between
between flat
flat platens, a high friction factor results in an inward flow of material (m = 0.6); however, a low
friction factor results in an outward flow of material (m = 0.04).
 Materials 2016, 9, 751 7 of 16

platens, a high friction factor results in an inward flow of material (m = 0.6); however, a low friction
Materials 2016, 9, 751 7 of 16
Materials 2016, 9, 751
factor results in an
outward flow of material (m = 0.04). 7 of 16

(a)(a) (b)(b)
Figure
Figure
Figure 5.5.5. Finite
Finite
Finite Element
Element
Element Method
Method
Method (FEM)
(FEM)
(FEM) simulation
simulation
simulation ofofof RCT
RCT
RCT for
for
for different
different
different friction
friction
friction conditions.
conditions.
conditions. (a)
(a)
(a) Initial
Initial
Initial
mesh (axisymmetrical
mesh(axisymmetrical
mesh model);
(axisymmetricalmodel);
model);(b) (b) Deformed
(b)Deformed samples
Deformedsamples
samplesfor for different
fordifferent friction
differentfriction conditions
frictionconditions (top: m =0.6;
(top:mm==0.6;
conditions(top: 0.6;
bottom:
bottom: m = 0.04).
bottom:mm==0.04).
0.04).

2.5. Experimental
2.5. Procedure
2.5.Experimental
ExperimentalProcedure
Procedure
Experimentaltests
Experimental
Experimental tests
tests were
were
were conducted
conducted
conducted totodetermine
to determinedetermine
the dualthethedual
dual
friction friction
friction
factor factor
factor
map of mapofofthe
map
the aforementioned the
aforementioned
aforementioned
lubricants lubricants
lubricants
for material UNSforfor material
material
A96082 UNS
UNS
under A96082
A96082
specific under
under specific
specific
forming forming
forming
conditions. conditions.
conditions.
Accordingly, Accordingly,
Accordingly,
experimental
experimental
experimental compression
compression compression tests
tests were
tests were realized were realized
realized
using using rings
using rings
rings treated treated
with treated with
with the
the lubricants the lubricants
lubricantsfor
presented; presented;
presented; forfor
this purpose,
thisuniversal
the
this purpose,test
purpose, thethe universal
machine
universal HOYTOMtestmachine
test machine
HM-100kN HOYTOM
HOYTOM (HOYTOMHM-100kN
HM-100kN (HOYTOM
S.L., Leioa,
(HOYTOM S.L.,Leioa,
Spain)S.L.,
(Figure Leioa, Spain)
6a) and flat
Spain)
(Figure
platens6a)
(Figure 6a) and
as forming flat platens as
tools (Figure
and flat platens forming
6b) were
as forming tools
used.
tools (Figure 6b) were
(Figure 6b) were used. used.

(a)(a) (b)(b)
Figure
Figure 6.6.6. Equipment
Equipment andand forming tools. (a) Universal test machine HOYTOM HM-100kN; (b) Detailed
Figure
view of Equipment
flat platens. andforming
formingtools.
tools.(a)
(a)Universal
Universaltest
testmachine
machineHOYTOM
HOYTOMHM-100kN;
HM-100kN;(b)
(b)Detailed
Detailed
view
viewof
offlat
flatplatens.
platens.

ToToconduct
conductthetheRCTs,
RCTs,each
eachringringwas wasplaced
placedononthe thebottom
bottomplateplateofofthe thetesttestmachine
machine
To conduct
(compressionarea) the RCTs,
area)after each ring
afterlubricationwas placed
lubricationofofthe on
thecontact the bottom
contactsurfaces; plate
surfaces;then,of the
then,the test
thetop machine
topplate
platewas (compression
waspositioned
positioned
(compression
area) aftercontact
making lubrication
with of
thethe contact
upper surfaces;
surface of thethen,
ring. the
Oncetopcontact
plate was
waspositioned
made, an making
increment contact
of thewith
load
making contact with the upper surface of the ring. Once contact was made, an increment of the load
the
wasupper surface
applied of the
causing a ring. Onceincontact
reduction height was
and made,
the an increment
deformation of of
thethe loadradius.
inner was applied
According causing
to a
the
was applied causing a reduction in height and the deformation of the inner radius. According to the
alternativeapproach
alternative approachpresented
presentedininthis
thispaper,
paper,the thelubricant
lubricantwas
wasonly
onlyapplied
appliedat atthethefirst
firststage,
stage,asas
explained previously.
explained previously.
Materials 2016, 9, 751 8 of 16

reduction in9,height
Materials 2016, 751 and the deformation of the inner radius. According to the alternative approach 8 of 16
presented in this paper, the lubricant was only applied at the first stage, as explained previously.
The RCTs
The RCTswere wereconducted
conducted onon three
three different
different ringsrings for lubricant
for each each lubricant
in orderintoorder to generate
generate enough
enough points to create the friction map. Compressions were applied to reach loads
points to create the friction map. Compressions were applied to reach loads of 30, 50, and 70 of 30, 50,kN.
andThe
70
kN. velocity
ram The ramofvelocity of the
the upper plateupper
of theplate of the test
test machine wasmachine
2.5 mm/swasin 2.5 mm/s
all of in all The
the cases. of the cases. The
conditions of
conditions of the process are specified
the process are specified in Table 1. in Table 1.

Table 1.
Table 1. Ring compression
compression test
test conditions.
conditions.

Temperature Ram Velocity 1st Stage Load 2nd Stage Load 3rd Stage Load
Ram Velocity 1st Stage Load 2nd Stage Load 3rd Stage Load
Temperature
(°C) (◦ C) (mm/min) (kN) (kN) (kN)
(mm/min) (kN) (kN) (kN)
20 2.5 30 50 70
20 2.5 30 50 70

Rings at the end of every compression stage are presented in Figure 7.

Rings at the end of every compression stage are presented in Figure 7.

1st stage

2nd stage

3rd stage

Figure 7. Rings after each compression stage using four lubricants. From left to right: copper paste,
Figure 7. Rings after each compression stage using four lubricants. From left to right: copper paste,
silicone oil, MoS2 grease, aluminum anti-size.
silicone oil, MoS2 grease, aluminum anti-size.

In this
thisfigure, thethe
figure, dimensional changes
dimensional of the of
changes ringthe
for ring
each lubricant
for each during the three
lubricant duringcompression
the three
stages can be observed.
compression stages canThese changes will
be observed. be changes
These explainedwillin detail in a subsequent
be explained section.
in detail in a subsequent
section.
2.6. Finite Element Model Validation
2.6. Finite
The Element Model Validation
finite element model has been validated with theoretical results obtained by
De Pierre
The finite element model has beenmethod
et al. [29] by an analytical validated forwith
the theoretical
canonical results
aspect obtained
ratio (6:3:2).
by DeDue
Pierretoetthe
al.
dependency of the method with the material, results for the same finite element
[29] by an analytical method for the canonical aspect ratio (6:3:2). Due to the dependency of the model and different
metallic alloys
method with theare presented
material, in Figure
results for the8.sameThese alloys
finite are (form
element modelleftandtodifferent
right): copper
metallicalloy
alloysUNS
are
C11000 and aluminum alloys UNS A92024 (both flow curves are imported from
presented in Figure 8. These alloys are (form left to right): copper alloy UNS C11000 and aluminum the materials library
of DEFORM
alloys F2™) and
UNS A92024 UNS
(both flowA96082
curves(used in this work).
are imported from the materials library of DEFORM F2™) and
The biggest differences
UNS A96082 (used in this work). are found for the friction factor m = 0.5. However, these differences
are minimum
The biggestfor the friction
differences coefficients
are found for themost
frictioncommonly
factor m = used in cold these
0.5. However, forming (m < 0.2).
differences are
Additionally, a comparison of the dimensional changes in diameter obtained
minimum for the friction coefficients most commonly used in cold forming (m < 0.2). Additionally, by experiments anda
finite element
comparison ofsimulation is shown
the dimensional in Figure
changes 9. In this
in diameter figure, it
obtained bycan be seen graphically
experiments and finite how the
element
inner diameter decreases in both cases during the compression process (lubricants: silicone oil
simulation is shown in Figure 9. In this figure, it can be seen graphically how the inner diameter
and aluminum anti-size).
decreases in both cases during the compression process (lubricants: silicone oil and aluminum
anti-size).
Materials 2016, 9, 751 9 of 16
Materials
Materials 2016,
2016, 9,
9, 751
751 99 of
of 16
16

(a)
(a) (b)
(b) (c)
(c)
Figure 8.
Figure 8.
Figure Finite
8. Finite element
Finite element model
model validation
element model validation with
validation with theoretical
with theoretical results
theoretical results obtained
results obtained by
obtained by De
by De Pierre
De Pierre et
Pierre et al.
et al. [29].
al. [29].
[29].
(a) Copper
(a)Copper
(a) alloy
Copperalloy UNS
alloyUNS C11000;
UNSC11000; (b)
C11000;(b) Aluminum
(b)Aluminum alloy
Aluminumalloy UNS
alloyUNS A92024;
UNSA92024;
A92024;(c)(c) Aluminum
(c)Aluminum alloy
Aluminumalloy UNS
alloyUNS A96082.
UNSA96082.
A96082.

(a)
(a) (b)
(b)
Figure
Figure 9.
9. Comparison
Comparison of
of experimental
experimental observations
observations and
and numerical
numerical results.
results. (a)
(a) Lubricant:
Lubricant: silicone
silicone oil;
oil;
Figure 9. Comparison of experimental observations and numerical results. (a) Lubricant: silicone oil;
(b)
(b) Lubricant:
Lubricant: aluminum
aluminum anti-size.
anti-size.
(b) Lubricant: aluminum anti-size.

Results
Results from
from the
the simulation
simulation are
are in
in good
good agreement
agreement with
with both
both analytical
analytical and
and experimental
experimental
Results
results, so from
the the simulation
numerical model are
can in
be good agreement
considered with
validated.
results, so the numerical model can be considered validated. both analytical and experimental results,
so the numerical model can be considered validated.
Materials 2016, 9, 751 10 of 16
Materials 2016, 9, 751 10 of 16
3. Results and Discussion
3. Results and Discussion
3.1. Determination of Dimensional Variables
3.1. Determination
Once RCTs haveof Dimensional Variables
been conducted in the laboratory, it is possible to create the friction factor map;
to accomplish
Once RCTsthis,
havethebeen
reductions
conductedin height (Equation (5))
in the laboratory, it isand innertoradius
possible create(Equation (6))
the friction that map;
factor each
ring experienced under different lubricants have to be obtained:
to accomplish this, the reductions in height (Equation (5)) and inner radius (Equation (6)) that each
h 
ring experienced under different lubricants have h1 obtained:
0 to be
r (%)   100 (5)
h h
h00 − h1
r h (%) = · 100 (5)
 d  dh0 
 i ,0 i,1 
r (%)      100 (6)
d  ddii,0 − d i,1

r d (%) =  ,0  · 100 (6)
di,0
where di,0 is the inner diameter of the ring at the beginning, di,1 is the inner diameter of the ring at the
end of dthe
where i,0 is the inner
stage, h0 is diameter
the initialofheight
the ring at the
of the ring, and h1 is dthe
beginning, i,1 is the height
final inner diameter
at every of the ring stage.
reduction at the
end of the stage,
To measure h0 is the initialchanges
the dimensional height of ofthe
thering,
innerand h1 is thedifferent
diameter, final height at every
methods reduction
have stage. To
been proposed.
measure
Wang and theLenard
dimensional changes of
[15] measured thethe inner diameter,
diameter with calipersdifferent methods
in two have been
directions at theproposed.
middle and Wangon
and Lenard
the end of the[15] measuredHartley
specimen. the diameter
et al. [30] with calipers in
proposed a newtwo method
directions byatusing
the middle and on the
a ball-bearing end of
of known
the specimen.
diameter, andHartley et al. [30]Shahriari
more recently, proposedeta new method
al. [19] combinedby using thea aforementioned
ball-bearing of known
method diameter,
with a
and
profile projector equipped with an X-Y micrometer table; however, this technique stillprojector
more recently, Shahriari et al. [19] combined the aforementioned method with a profile offered
equipped with an X-Y
problems because micrometer
the method table;the
assumes however,
deformed thisprofile
technique still offered
possesses problems
a circular shape because
which does the
method assumes the deformed profile possesses a circular shape which
not match perfectly to reality. Goetz et al. [31] demonstrated that there are more types of profiles does not match perfectly to
reality. Goetz et al. [31] demonstrated that there are more types of profiles
besides concave and convex, which reduces the feasibility of the measurement process. For the sake besides concave and convex,
which reducesand
of simplicity, the feasibility
as there isofnothe measurement
perfectly process.
established For the sake
procedure, of simplicity,
a profile projector,and as there
TESA VISIO is
no perfectly established procedure, a profile projector, TESA VISIO (TESA
(TESA SA, Renens, Switzerland) (Figure 10), has been used in this paper to take the measurements of SA, Renens, Switzerland)
(Figure
the middle10), plane.
has been used in this paper to take the measurements of the middle plane.

Figure 10. Details of the inner diameter measurement with profile projector TESA VISIO.
Figure 10. Details of the inner diameter measurement with profile projector TESA VISIO.

Table 2 shows the measurements obtained after performing the RCT with each ring in every
Table 2 shows the measurements obtained after performing the RCT with each ring in every stage
stage of the compression process. Each value of the diameter is the mean of three measurements, and
of the compression process. Each value of the diameter is the mean of three measurements, and each
each measurement is provided by the profile projector selecting eight points at 45 around the
measurement is provided by the profile projector selecting eight points at 45◦ around the diameter; the
diameter; the diameter of the circle is determined by fitting the data by the least-square method.
diameter of the circle is determined by fitting the data by the least-square method.
Materials 2016, 9, 751 11 of 16

Materials 2016, 9, 751 11 of 16

Table 2. Dimensional measurements after conducting the RCTs.

Table 2. Dimensional measurements after conducting the RCTs.
Lubricants
Lubricants Forming Stage
Forming Stage Inner
Inner Diameter
Diameter (mm) OuterOuter
(mm) Diameter
Diameter (mm)(mm)
11 5.1495.149 10.989
10.989
Aluminum anti-size
Aluminum anti-size 22 4.8034.803 12.510
12.510
33 4.2944.294 13.680
13.680
11 5.2595.259 11.080
11.080
MoSMoS
2 grease
2 grease 22 4.8784.878 12.754
12.754
3 4.476 13.735
3 4.476 13.735
11 5.0935.093 10.842
10.842
Silicone oil oil
Silicone 22 4.6384.638 12.383
12.383
3 4.002 12.982
3 4.002 12.982
11 5.1025.102 10.943
10.943
Copper paste 2 4.800 12.538
Copper paste 2 4.800 12.538
3 4.377 13.598
3 4.377 13.598

AsAs it can
it canbebeseenseenininTable
Table2,2,the
theouter
outerdiameter
diameter expands
expands in in all
allcases;
cases;while
whilethe theinner
inner diameter
diameter
increases
increases at at
thethefirst stage
first stageand
anddecreases
decreasesatatthethelast
last stages
stages during the compression
during the compressionprocess.
process.
Images
Images (from
(fromthe theprofile
profileprojector)
projector)of of the
the rings’
rings’ surfaces thathave
surfaces that havebeenbeeninincontact
contact with
with thethe
flatflat
platens
platensforfor
all all
lubricants
lubricants areare
shown
shown in in
Figure
Figure 11.11.
The
The profile projector
profile projector allows
allowsfor
forthe
theopportunity
opportunitytotosee
thesee
contact surfaces
the contact of the workpiece
surfaces after each
of the workpiece aftercompression
each compressionstage. stage.
In all the cases
In all the there
cases isthere
an external
is an
ring belonging to the lateral surface of the ring that establishes new contact
external ring belonging to the lateral surface of the ring that establishes new contact once the once the compression takes
place; this will lead
compression takesto an inhomogeneous
place; contact
this will lead to an surface andcontact
inhomogeneous it will surface
influence
andthe friction
it will factorthe
influence so it
canfriction
be onefactor
reason sonot to be
it can consider this factor
one reason m constant
not to consider thisthroughout the process.
factor m constant throughout the process.

Figure 11. Rings tested in the laboratory after each forming stage. Images obtained by TESA VISIO
Figure 11. Rings tested in the laboratory after each forming stage. Images obtained by TESA VISIO
profile projector.
profile projector.
Materials 2016, 9, 751 12 of 16
Materials 2016, 9, 751 12 of 16

3.2. Dual
3.2. Friction
Dual FrictionFactor
FactorMap
Map
The friction
The factor
friction map
factor waswas
map obtained by a by
obtained finite element
a finite modelmodel
element according to the material
according properties
to the material
of properties
the ring (aluminum alloy UNS A96082). By overlapping the friction curves obtained
of the ring (aluminum alloy UNS A96082). By overlapping the friction curves obtained numerically
and through experimental
numerically and throughtesting, a quantitative
experimental testing, and qualitativeand
a quantitative assessment of the
qualitative behaviorofofthe
assessment each
lubricant
behaviorcan be realized,
of each lubricantand
can some considerations
be realized, about alternative
and some considerations uses
about of RCTsuses
alternative are of
described.
RCTs
Figure 12 showsFigure
are described. the dual frictionthe
12 shows factor
dualmap obtained.
friction factor map obtained.

Figure12.
Figure 12.Dual
Dualfriction
friction factor
factor map
map for
for UNS
UNSA96082
A96082experimental
experimentalresults.
results.

The dual friction factors map plotted for alloy UNS A96082 (Figure 12) shows that the value of
theThe dualfactor
friction friction factors experimentally
obtained map plotted for foralloy
eachUNS A96082
lubricant can(Figure
only be12) shows that
considered the value
constant for of
thesmall
friction factor obtained experimentally for each lubricant can only be considered
reductions in height (lower than 20% in most cases). Table 3 presents the friction factorsconstant for small
reductions
estimatedinbyheight (lowerthe
comparing than 20% in most
experimental andcases). Tableresults.
simulated 3 presents the friction
However, factors estimated
for reductions higher
bythan
comparing
this percentage, the assumption of a constant value for the friction coefficient throughout thethis
the experimental and simulated results. However, for reductions higher than
percentage,
RCT is nothe assumption
longer of aTable
acceptable. constant valuethe
3 shows for ranges
the friction coefficient
of friction throughout
factors encounteredthefor
RCT is no
each
longer acceptable.
lubricant, and theTable 3 shows
average thecoefficient
friction ranges offorfriction factors encountered for each lubricant, and the
each range.
average friction coefficient for each range.
Table 3. Friction factors determined by the RCT according to reduction in height.
Table 3. Friction factors determined by the RCT according to reduction in height.
Aluminum
Friction Factor (m) MoS2 Grease Copper Paste Silicone Oil
Anti-Size
Aluminum
Friction
SmallFactor
height(m)reduction MoS2 Grease Copper Paste Silicone Oil
0.12 Anti-Size
0.20 0.23 0.24
(rh < 20%)
Small height
0.12 0.20 0.23 0.24
reduction (rh < 20%)
Large height reduction 0.12 < m < 0.24, 0.20 < m < 0.28, 0.23 < m < 0.27, 0.24 < m < 0.39,
Large height 0.12 < m < 0.24, mavg 0.20 < m < 0.28, mavg 0.23 < m < 0.27, mavg 0.24 < m < 0.39, mavg
reduction (r > 20%)
(rh > 20%) ** mavg = 0.18
= 0.18 m=avg0.24
= 0.24 mavg==0.25
0.25 mavg = =0.32
0.32
h
* Upper value: friction coefficients range; lower value: average friction coefficient.
* Upper value: friction coefficients range; lower value: average friction coefficient.
Materials 2016, 9, 751 13 of 16
Materials 2016, 9, 751 13 of 16

This
Thisbehavior
behavior is reasonable because
is reasonable the tribological
because conditionsconditions
the tribological are changingareat changing
the dies-workpiece
at the
interfaces during
dies-workpiece deformation
interfaces duringcaused by heavily
deformation causedloaded regimes,
by heavily loaded asregimes,
observed in real transient
as observed in real
metal forming
transient metaloperations, such as open
forming operations, such asdieopen
forging or stamping.
die forging On one
or stamping. Onhand, new contact
one hand, areas
new contact
corresponding
areas correspondingto the tolateral surfaces
the lateral of the of
surfaces ring
theare
ringestablishing new contact,
are establishing and, on
new contact, and,the
onother hand,
the other
hand,
the the lubricant
lubricant layer duringlayer large
during largereductions
height height reductions is not to
is not expected expected to be homogeneously
be homogeneously distributed
atdistributed
the contact at area
the contact areato
according according to the inhomogeneous
the inhomogeneous contact pressures
contact pressures profile [32].profile
As an[32]. As an
example,
example,
Figure 13a Figure
shows 13a shows the
the contact contact
pressure pressure
profile for a profile for a slim
slim preform, as itpreform,
occurs inastheit occurs in theofearly
early stages open
stages
die of open
forging die forgingcomponents,
of cylindrical of cylindricalandcomponents,
Figure 13b and Figure
presents the13b presents
same profiletheforsame profile for a
a flattened-shape
flattened-shape
preform, as in thepreform, as in
late stages the late
during stages during
compression compression
of cylindrical of cylindrical parts.
parts.

(a) (b)
Figure 13. Contact pressure profiles for different friction conditions. (a) Slim preform (d = 10 mm and
Figure 13. Contact pressure profiles for different friction conditions. (a) Slim preform (d = 10 mm and
h = 20 mm); (b) Flattened-shape preform (d = 10 mm and h = 3 mm).
h = 20 mm); (b) Flattened-shape preform (d = 10 mm and h = 3 mm).

As observed in the first case (Figure 13a), a minimum is detected at the center; this configuration
can As observed in
be favorable to the firstthe
entrap caselubricant
(Figure 13a), at thisa minimum
area, resultingis detected at theresponse
in positive center; this configuration
against friction.
can
However, in the second case (Figure 13b) the maximum of the profile is presented around the friction.
be favorable to entrap the lubricant at this area, resulting in positive response against center
However, in the second
of the workpiece, case (Figure
as expected from13b) the maximum
classical analysis of the under
[33]; profilethis
is presented around
configuration thethelubricant
center of
the workpiece,
layer at the centeras expected
could break fromand classical analysis the
slip towards [33];edge
under of this configuration
the workpiece. In the
bothlubricant
cases, everylayer
atcontact
the center could break and slip towards the edge of the workpiece.
pressure profile has a different friction condition associated with it and vice versa. This In both cases, every contact
pressure
means that profile
duringhas any
a different
transient friction
metalcondition associatedsuch
forming operation, withas it and vicetribological
forging, versa. This conditions
means that
during
changeany transient
during metal forming
the process as a result operation, such as
of different forging,
stress tribological
states, conditions change
so the establishment during
of the initial
the process as a result of different stress states, so
friction factor for the entire operation, as usual, should be reconsidered. the establishment of the initial friction factor for the
entireFurthermore,
operation, as in usual, should be reconsidered.
metal forming processes several types of lubrication can be distinguished [27].
In dryFurthermore,
forming there in metal
is no forming
lubricantprocesses severalsotypes
at the interface of lubrication
friction is high. In can be distinguished
boundary lubrication[27]. thereIn
dry
is a forming
thin filmthere is no lubricant
of lubricant, generally at organic,
the interface that issophysically
friction isabsorbed
high. In boundary
or chemically lubrication
adheredthere to theis
amaterial
thin filmsurface.
of lubricant,
As in the generally
former case,organic, thatremains
friction is physically absorbed
high. Full or chemically
film lubrication occursadhered
when thereto the
material
is a thicksurface.
layer ofAssolid in the former case,
lubricant; in this friction
case, theremains high. Full
conditions film lubrication
affecting friction are occurs
governed when bythere
the
isshear
a thick layer of solid
strength lubricant;
the lubricant in this
film. When case,
theretheisconditions
a thick layer affecting
of liquidfriction are governed
lubricant, by the
the lubrication
shear strength
conditions are of the lubricant in
hydrodynamic; film.
thisWhen there isisa governed
case, friction thick layer byofthe
liquid lubricant,
viscosity of the the lubrication
lubricant and
conditions
the slidingare hydrodynamic;
velocity at the contactin this interface
case, friction is governed
allowing frictionbytothe beviscosity
considered of the lubricantlow.
relatively and
the sliding velocity
However, the mostatcommon the contact interface
situation allowing
in metal formingfriction to be considered
operations is mixed-layerrelatively low. However,
lubrication; this is
the most common
because hydrodynamic situation in metal forming
conditions cannot be operations
maintained is mixed-layer
at high pressureslubrication;
andthis lowis because
sliding
hydrodynamic
velocities. Taking conditions
this into cannot be maintained
account, it is necessary at high topressures
establish and a newlowprocedure
sliding velocities. Taking
to determine
this into account,
conditions it is in
of friction necessary
metal formingto establish
processes a new procedure
closer to thoseto determine
actually conditions of friction in
encountered.
metalAs forming
indicatedprocesses
by Mandićcloser and
to those actually[28],
Stefanović encountered.
modelling results and their transfer onto real
As indicated
processes by Mandić
highly depends upon and theStefanović
similarity [28], of themodelling resultsconditions
contact friction and their in transfer
modelling ontoand real
processes
during thehighly depends
real process. upon thetosimilarity
According of the contact
this, the friction factor atfriction
the very conditions
beginninginofmodelling
any forming and
during
process,the real friction
when process.conditions
According cantobe this, the friction
significantly factor at
different to the ones
very resulting
beginning atof
theany
endforming
of the
process, can be far from reality. For small height reductions (lower than 20% in the case of the
Materials 2016, 9, 751 14 of 16

process, when friction conditions can be significantly different to the ones resulting at the end of the
process, can be far from reality. For small height reductions (lower than 20% in the case of the material
considered in this paper, UNS A96082) friction curves obtained experimentally for each lubricant are
clearly similar to some of the curves obtained by the finite element simulation. However, for high
height reductions (higher than 20% in the case of UNS A96082), this behavior is not observed, and
the experimental results fall into a friction factor range for each lubricant. In this case, the use of a
single average value of the friction factor assigned to every lubricant-component pair in the range of
deformation is recommended.

4. Conclusions
This paper shows that friction factors obtained for each lubricant-surface pair by the alternative
approach of the RCT technique seem to vary during the compression process, as observed in real
transient metal forming operations; thus, the points obtained do not entirely match a single curve of
the friction factor map plotted from the finite element simulation.
Based on these observations, the alternative approach of the RCT presented in this paper can be
successfully applied in the determination of friction conditions that are better adapted to real processes.
To obtain a friction factor closer to the actual one present in real forming processes, the new procedure
described in this paper is recommended:

• For small deformation stages (rh < 20% in this work), the conventional RCT [10] can be applied, as
the tribological conditions will remain similar during the operation (Table 3). Particularly, in this
paper, friction factors obtained for small reductions were 0.12, 0.20, 0.23, and 0.24 for lubricants
MoS2 , aluminum anti-size, copper paste, and silicone oil, respectively.
• For large deformations-reductions (rh > 20% in this work), it would be advisable to use the
alternative approach of the RCT presented in this paper (without any lubrication between steps),
and to obtain an average value of the friction factor assigned to every lubricant-component pair
in the range of deformation considered (Table 3). This average value was 0.18, 0.24, 0.25, and
0.32 for lubricants MoS2 , aluminum anti-size, copper paste, and silicone oil, respectively. Some
differences were also observed in the width of the range obtained for each lubricant, being small
for aluminum anti-size and copper paste (0.08 and 0.04, respectively) and bigger for MoS2 and
silicone oil (0.12 and 0.15, respectively). In these last two cases the determination of an average
value is more critical.

The methodology presented in this paper can be used in the real practice of forming processes as
follows. In incremental processes and those ones where lubrication can be applied between stages of
small deformations, the conventional RCT is advisable for taking into account that friction conditions
are more homogeneous with the application of incremental loads [34]. However, in those metal forming
operations where heavy loaded regimes occur and the lubricant cannot be applied in intermediate
stages, the lubricant layer is not constant during the forming process, and the tribological conditions
change throughout the operation. When a transient forming process, such as forging or stamping, is
being performed, the use of the alternative RCT approach is recommended. In this case, an average
value for the whole friction range should be estimated instead of applying the initial friction factor for
the entire operation, as usual. For more precise results, the friction factor should be selected according
to the deformation stage in each particular case. Thus, the friction factor values should be based on the
deformation level of each particular forming process, and this alternative approach of the RCT can
be valuable.
To help in this matter, dual friction factor maps, as presented in this work, can be an interesting
tool to determine friction coefficients at die-workpiece interfaces closer to those found in real
process conditions.

Acknowledgments: The authors would like to take this opportunity to thank the Research Group of the UNED
“Industrial Production and Manufacturing Engineering (IPME)” for the given support during the development of
Materials 2016, 9, 751 15 of 16

this work. This work has been financially supported by the funds provided through the Annual Grant Call of the
E.T.S.I.I. of UNED of reference 2016-ICF04.
Author Contributions: A.M.C. and M.A.S. conceived and designed the experiments; M.V. and J.C. performed the
experiments; A.M.C., F.M., and L.S. analyzed the data; and A.M.C. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Lu, P.; Wood, R.J.K.; Gee, M.G.; Wang, L.; Pfleging, W. The friction reducing effect of square-shaped surface
textures under lubricated line-contacts: An experimental study. Lubricants 2016, 4, 26–39. [CrossRef]
2. Vandoni, L.; Demir, A.G.; Previtali, B.; Lecis, N.; Ugues, D. Wear behavior of fiber laser textured tin coatings
in a heavy loaded sliding regime. Materials 2012, 5, 2360–2382. [CrossRef]
3. Yazawa, S.; Minami, I.; Prakash, B. Reducing friction and wear of tribological systems through hybrid
tribofilm consisting of coating and lubricants. Lubricants 2014, 2, 90–112. [CrossRef]
4. Groche, P.; Muller, C.; Stahlmann, J.; Zang, S. Mechanical conditions in bulk metal forming tribometers-part
one. Tribol. Int. 2013, 62, 223–231. [CrossRef]
5. Hu, C.L.; Ou, H.A.; Zhao, Z. An alternative evaluation method for friction condition in cold forging by ring
with boss compression test. J. Mater. Process. Technol. 2015, 224, 18–25. [CrossRef]
6. Zhang, D.W.; Ou, H.A. Relationship between friction parameters in a coulomb-tresca friction model for bulk
metal forming. Tribol. Int. 2016, 95, 13–18. [CrossRef]
7. Sofuoglu, H.; Gedikli, H. Determination of friction coefficient encountered in large deformation processes.
Tribol. Int. 2002, 35, 27–34. [CrossRef]
8. Barcellona, A.; Cannizzaro, L.; Forcellese, A.; Gabrielli, F. Validation of frictional studies by double cup
extrusion tests in cold-forming. CIRP Ann. Manuf. Technol. 1996, 45, 211–214. [CrossRef]
9. Kunogi, M. A new method of cold extrusion. J. Sci. Res. Inst. 1956, 50, 215–246. [CrossRef]
10. Male, A.; Cockcroft, M. A method for the determination of the coefficient of friction of metals under
conditions of bulk plastic deformation. J. Inst. Met. 1965, 93, 38–45.
11. Hawkyard, J.; Johnson, W. An analysis of the changes in geometry of short hollow cylinder during axial
compression. Int. J. Mech. Sci. 1967, 9, 163–182. [CrossRef]
12. Male, A.T. Variations in friction coefficients of metals during compressive deformation. J. Inst. Met. 1966, 94,
121–125.
13. Male, A. The determination of coefficients of friction during compressive deformation. Proc. Inst. Mech. Eng.
1968, 182, 64–67. [CrossRef]
14. Carter, W.T.; Lee, D. A finite-element analysis of cylinder and ring compression and its
experimental-verification. Comput. Struct. 1985, 21, 1–19. [CrossRef]
15. Wang, F.; Lenard, J.G. An experimental-study of interfacial friction-hot ring compression. J. Eng. Mater.
Technol. ASME 1992, 114, 13–18. [CrossRef]
16. Sofuoglu, H.; Gedikli, H.; Rasty, J. Determination of friction coefficient by employing the ring compression
test. J. Eng. Mater. Technol. ASME 2001, 123, 338–348. [CrossRef]
17. Sofuoglu, H.; Rasty, J. On the measurement of friction coefficient utilizing the ring compression test.
Tribol. Int. 1999, 32, 327–335. [CrossRef]
18. Camacho, A.M.; Torralvo, A.I.; Bernal, C.; Sevilla, L. Investigations on friction factors in metal forming of
industrial alloys. Procedia Eng. 2013, 63, 564–572. [CrossRef]
19. Shahriari, D.; Amiri, A.; Sadeghi, M.H. Study on hot ring compression test of nimonic 115 superalloy using
experimental observations and 3D FEM simulation. J. Mater. Eng. Perform. 2010, 19, 633–642. [CrossRef]
20. Zhu, Y.C.; Zeng, W.D.; Ma, X.; Tai, Q.G.; Li, Z.H.; Li, X.G. Determination of the friction factor of Ti-6Al-4V
titanium alloy in hot forging by means of ring-compression test using FEM. Tribol. Int. 2011, 44, 2074–2080.
[CrossRef]
21. Bermudo, C.; Martin, F.; Martin, M.J.; Sevilla, L. Experimental validation of the new modular application of
the upper bound theorem in indentation. PLoS ONE 2015, 10, 1–15. [CrossRef] [PubMed]
22. Bridgman, P. Studies in Large Plastic Flow and Fracture; McGraw-Hill: New York, NY, USA, 1952.
23. Altan, T.; Oh, S.; Gegel, H. Metal Forming Fundamentals and Applications; ASM International: Materials Park,
OH, USA, 1983.
Materials 2016, 9, 751 16 of 16

24. Fereshteh-Saniee, F.; Fatehi-Sichani, F. An investigation on determination of flow curves at room temperature
and under forming conditions. J. Mater. Process. Technol. 2006, 177, 478–482. [CrossRef]
25. Bay, N.; Nakamura, T.; Schmid, S. Green lubricants for metal forming. In Proceedings of the 4th International
Conference on Tribology in Manufacturing Processes, Nice, France, 13–15 June 2010.
26. Shih-Chen, S.; Fu-I, L. Biopolymer green lubricant for sustainable manufacturing. Materials 2016, 9, 1–11.
27. Gariety, M.; Ngaile, G. Cold and Hot Forging. Fundamentals and Applications; ASM International: Materials
Park, OH, USA, 2005.
28. Mandić, V.; Stefanović, M. Friction studies utilizing the ring-compression test-part ii. Tribol. Ind. 2003, 25,
76–82.
29. De Pierre, V.; Gurney, F.; Male, A.T. Mathematical Calibration of the Ring Test with Bulge Formation; Westinghouse
Electric Corporation: Savannah, OH, USA, 1972.
30. Hartley, R.S.; Cloete, T.J.; Nurick, G.N. An experimental assessment of friction effects in the split hopkinson
pressure bar using the ring compression test. Int. J. Impact Eng. 2007, 34, 1705–1728. [CrossRef]
31. Goetz, R.L.; Jain, V.K.; Morgan, J.T.; Wierschke, M.W. Effects of material and processing conditions upon ring
calibration curves. Wear 1991, 143, 71–86. [CrossRef]
32. Veganzones, M.; Camacho, A.M.; García-Prada, J.C.; Sebastián, M.Á. Contact pressure profiles in
axisymmetric compression considering friction and geometrical factors. Procedia Eng. 2014, 69, 72–80.
[CrossRef]
33. Rowe, G. An Introduction to the Principles of Metalworking; Edward Arnold Limited: London, UK, 1971.
34. Valero, J.; Marín, M.; Camacho, A. Influence of load application methodology in the performance of ring
compression tests. Procedia Eng. 2015, 132, 306–312. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).