Vol. 38, No.

1 (2016) 127-132

Tribology in Industry

RESEARCH
www.tribology.fink.rs

Computer Simulation of Continuous SPD Process

Commercially Pure Titanium Using Virtual Full
Factorial Experiment with the Influence of
Friction Factors

V.I. Semenov a, S.-J. Huang b, L.Sh. Shuster c, A.G. Raab c

a Instituteof Oil and Gas Technologies and Novel Materials, Ufa, Russia,
b National Taiwan University of Science and Technology, Taiwan,
c State Aviation Technical University, Ufa, Russia.

Keywords: ABSTRACT
Computer modeling
With the use of computer modeling in the environment of the software
Commercially pure titanium
DEFORM-3D virtual conducted a full factorial experiment process, equal
A virtual full factorial experiment
channel angular pressing (ECAP) by method "Conform" of commercially
The friction factor
pure titanium, which evaluated the effect of independent parameters
The deformation forces
(speed of rotation of the impeller, the friction factor on the side surfaces of
the impeller and the factor of friction between the billet and the matrix).
Corresponding author: As a result of the experiment obtained the regression equation and
V.I. Semenov identified the most important individual factors and their mutual
Institute of Oil and Gas Technologies combinations that affect the response parameter are the deformation
and Novel Materials, Ufa, Russia. forces.
E-mail: [email protected] © 2016 Published by Faculty of Engineering

1. INTRODUCTION

Currently, there is interest in research in the field

of increasing the strength of metals by shredding
patterns to submicrocrystalline (SMC) size by
processing under conditions of severe plastic
deformation (SPD) [1]. One of the SPD methods is
equal channel angular pressing (ECAP) [2,3] and
its development - ECAP-Conform [4], was
developed to obtain long length billets with bulk
SMC structure and allows you to create Fig. 1. Scheme one of the methods SPD - equal
preconditions for the implementation of the SPD channel angular pressing - Conform (ECAP -
process. In Fig. 1 is a diagram of the process for the Conform): 1 - stationary matrix; 2 - billet; 3 -
implementation of the ECAP-Conform. impeller-punch.

127
 V.I. Semenov et al., Tribology in Industry Vol. 38, No. 1 (2016) 127-132

Process technology is based on the refinement of The advantage of FFE is the ability to describe
the structure by the method of SPD and the process in full compliance with the algorithm
implemented on the installation of the ECAP- of physical experiment, based on established
Conform is an effective way to increase the assumptions. FFE is the most easy to implement
strength of metals and alloys. However, for the numerous methods of physical experiment.
manufacture of long length semi-finished When using FFE purpose is to obtain a linear
products according to the above scheme, it is mathematical model of the process, which will
necessary to solve the problem caused by allow defining future strategy for conducting a
identified by contradiction. It is that for feeding real experiment.
the billet in the deformation zone, you must use
active friction force on the side surfaces of the Thus, the purpose of modeling is holding a
impeller, i.e. have a maximum friction coefficient virtual process SPD by ECAP-Conform method
(f1). At the same time to implement directly the with the use of FFE, and the identification of
deformation process and ensure high-quality rational speed processing mode in combination
semi-finished product with a defect-free surface with the universal training of the billet surface in
in the deformation zone is required to ensure the condition of receiving lengthy SMC semi-
the lowest value of the friction coefficient (f2). finished product.

The use of fragmentary coating of lubricant only

on those surfaces where it is necessary to have a 2. RESEARCH METHODOLOGY
low friction coefficient leads to lower
productivity and mechanization of the process of It is possible to obtain more complete
SPD. It is more expensive and not so cheap for information about the investigated
the whole process. Thus, to improve the dependencies when you run the simulation, the
efficiency of the process SPD method ECAP- authors have used FFE. Experiment planning is
Conform is necessary to find a compromise the procedure of choice in the number and
solution, which would enable the use of one conditions of the experiments, a necessary and
variant of the preparation of the billet surface sufficient to obtain a mathematical model of the
before deformation processing, allowing process [9]. It is important to consider the
ensuring the supply of billet in the deformation following: a desire to minimize the number of
zone and obtaining semi-finished products of the experiments; the simultaneous variation of all
required quality in the deformation process. the variables that define the process; the choice
of a clear strategy that allows you to make
In scientific and practical activities, in particular, informed decisions after each series of
in the analysis of tribological systems, a experiments. Prior planning of the experiment it
significant numerical methods for the study of is necessary to gather additional information
complex processes, including computer about the test object, which uses the skills and
modeling using the latest software products [5- knowledge obtained in previous studies, or
7]. The efficiency of methods for modeling and described in the literature [10].
solving of engineering problems increase
significantly, if at the stage preceding the design Experiment planning was conducted on the
of the actual manufacturing process to create basis of modeling the process of obtaining long
conditions to assess the impact of the most length semi-finished products of commercially
important independent parameters. pure titanium method ECAP-Conform. Diagram
of the device represented in Fig. 1. The object of
The application of mathematical methods is one study is commercially pure titanium VT1-0 flow
of the most rational approaches to solving characteristics of which were laid during the
problems in assessing the effectiveness of non- development of the numerical model [11].
standard processing pressure. In this regard, it
seems reasonable to conduct numerical For the purposes of numerical simulation used a
simulation using virtual planning of a full standard software package (SP) “DEFORM-3D”.
factorial experiment (FFE) [8]. The purpose of the simulation and factorial
experiment in SP “DEFORM-3D” were pre-

128
 V.I. Semenov et al., Tribology in Industry Vol. 38, No. 1 (2016) 127-132

generated three-dimensional model in the Table 1. Factor levels.

software product “Kompas-3D”. X3
Factors X1 X2
(V, m/min.)
Assumptions Datum level (Xi) 0,50 0,50 20
Variability interval (Xi) 0,25 0,25 10
Higher level (xi = 1) 1 1 30
1) Material of the billet in the initial state is Lower level (xi = – 1) 0 0 10
isotropic and there are no initial stress and
strain;
2) The temperature of deformation is assumed 3. EXPERIMENTAL RESULTS AND DISCUSSION
to be 200 oC;
The number of experiments N was determined
3) The angle of intersection of the channels 120°; by the number of factors k in accordance with
4) The tool is absolutely rigid, and accounting the expression:
for the geometry of the tool is performed N  2 k  23  8 (1)
automatically; It is required to determine such values f1, f2, V,
5) The material of the billet took plastic; which will ensure the lowest force deformation P.
6) For the simulation was chosen 100 steps, The mathematical model after the
taking into account the full passage of the billet implementation of experiments full factorial
in spinnerets and obtaining a stable result; experiment is:
7) The billet is divided into 43553 trapezoidal y=b0+b1x1+b2x2+…+b12x1x2+b13x1x3+
elements. +b23x2x3…+b123x1x2x3, (2)

In preparation for the task of modeling believe where bi – regression coefficient.

that the most significant factors affecting the
production of defect-free products in the For calculating the coefficients of this model
conditions of intensive deformation at a using the extended planning matrix and results
temperature of 200 °C are factors of friction of the experiments (Table 2).
(contact options) billets with different parts of
Table 2. Extended matrix of plan 23 and the results of tests.
the tool and the rate of deformation, due to the
№ test х0 х1 х2 х3 х1х2 х1х3 х2х3 х1х2х3 У(кН)
speed of rotation of the impeller. In this regard,
1 + + + + + + + + 13,3
it was decided to hold a virtual FFE using a two- 2 + - + + - - + - 16,5
level model with three variables on a number of 3 + + - + - + - - 10,7
variables, followed by the formalization of the 4 + - - + + - - + 12,7
results in the form of the regression equation 5 + + + - + - - - 23,2
and the optimization of the selected factors. 6 + - + - - + - + 19,8
7 + + - - - - + + 13,5
8 + - - - + + + - 12,9
Thus, as independent variables during the
drawing process with a shift to describing the In Fig. 2 the problem of numerical simulation of
execution of the process and its effectiveness the process of ECAP-Conform, this resulted in
from the point of view of the deformation forces the minimum of the strain rate.
chose the friction factor from the upper and
lower surfaces of the impeller, which determines
the efficiency of feed of the billet in the
deformation zone f1 (X1), the friction factor from
the forming tool f2 (X2), the strain rate (the speed
of rotation of the impeller) V (X3). Parameter
response (dependent parameter) determined
the effect of strain rate P (Y).

The factors were varied at two levels. The Fig. 2. The result of the simulation process, the ECAP-
intervals of variation of the variables and their Conform: field distribution of the accumulated plastic
values in the natural scale are shown in Table 1. deformation. Strain rate Рaverage = 10,7 kN.

129
 V.I. Semenov et al., Tribology in Industry Vol. 38, No. 1 (2016) 127-132

The regression coefficients were calculated by optimal combination adopted in this study, the
the formula: independent parameters.
N

x y ,
Is of practical interest, the solution of the
i i
bi  i 1 (3)
N optimization problem for determining the actual
where i = 0, 1, 2, …, 8. values of the independent parameters
considered in the virtual experiment, numerical
On the basis of calculations received the following simulation and provides minimum value of the
General form of linear regression equation: deformation forces to implement the ECAP-
Conform. This task is solved by the method of
y=15,33X0 – 0,15X1+2,88X2–2,03X3+0,20X1X2– "steep ascent" [8].
1,15X1X3 – 1,28Х2Х3 – 0,50Х1Х2Х3 (4)
From equation (4) shows that the most Steps in the change factors were calculated in
significant influence on the strength of the natural scale. For this, we first identified the
deformation have the factor of friction in sliding product of the coefficients on the corresponding
contact of the billet and tool f2 (X2) and the rate of intervals of varying factors, i.e. biXi, then in
deformation V (X3). Moreover, from the proportion to these works prescribed steps.
coefficients of the regression equation shows that Using the values biXi, identified the steps in
the force of deformation will decrease with the changing the factors as follows. From
increase of both factors. A much smaller impact technological considerations chose a step change
on deformation force has an active factor of in the friction factor from the upper and lower
friction f1 (X1) from the upper and lower surfaces surfaces of the impeller (1 = 0.05). Steps for the
of the impeller, which is supplied to the billet in other factors received from the proportions of:
the deformation zone. Despite the fact that the
b1X 1 1 ; b1X 1 1 , (5)
greatest and unidirectional influenced by factors  
b 2 X 2  2 b3X 3 3
X2 and X3 can select variants universal
preparation of the surface of the billet. It should The sequence of stages steep ascent is presented
be noted that double and triple interaction have in Table 3. Some mental experiments
multiple interpretations and therefore a complex implemented in a computer model (Tab. 3).
interaction should be analyzed separately, and Experiment planning using the steep ascent
with reference to specific operating conditions of showed that under these conditions the strength
a multicomponent system. of the deformation will be minimal at high friction
from the upper and lower surfaces of the impeller
A priori, we can say that in the considered (f1 ≈ 1.00), the friction tending to the minimum
conditions the minimum value of the values from the forming tool (f2 ≈ 0.00), as well as
deformation forces can be obtained when the at high strain rate (V ≈ 25 m/min).

Table 3. Steep ascent.

Х1 (the friction factor
X2 (the friction factor X3 (the rate of Y (the force of
from the upper and
Factors from the forming deformation, V deformation P,
lower surfaces of the
tool, f2) m/min) kN)
impeller, f1)
bi -0.15 2,875 -2,025
bi Xi -0.15 1.44 - 40.5
Step -0.05 0.05 - 5.0
Step after rounding -0.05 0.05 - 5,0
Basic level (Xi) 0,5 0,5 20
Mental experience 0,45 0,3 15
Implemented experience 0,45 0,3 20
0,70
Mental experience 0,55 0,7 15
Mental experience 0,55 0,3 25
Implemented experience 0,55 0,7 15
0,90
Mental experience 0,45 0,7 25
Mental experience 0,55 0,3 15
Implemented experience 0,55 0,3 20 1,20
Implemented experience 0,55 0,3 15 1,50

130
 V.I. Semenov et al., Tribology in Industry Vol. 38, No. 1 (2016) 127-132

If all these values of independent parameters it tool f2 (X2) and the rate of deformation V (X3). It
is possible to ensure the strength of the was found that the active factor of friction f1 (X1)
deformation P ≈ 10.7 kN (Fig. 2). However, the from the upper and lower surfaces of the
objective of the study was to provide a process impeller, which is supplied to the billet in the
SPD processing method ECAP-Conform with the deformation zone, has a much smaller impact.
minimum possible force deformation at the
universal condition of preparation of the surface Virtual full factorial design method a steep
of the billet. ascent in the process of numerical simulation
allowed us to determine the numerical values of
By solving the inverse problem were able to friction factors from the upper and lower
choose the option of a universal surface surfaces of the impeller, f1 and forming parts of
preparation and rate of deformation, which the instrument, f2, which are universal for the
achieved a value of the deformation forces P ≈ SPD process of commercially pure titanium by
12.5 kN, which is quite acceptable. method ECAP-Conform.

Herewith want to ensure f1 = f2 = 0.3 and the rate For the practical implementation process of
of deformation of V ≈ 30 m/min. ECAP-Conform commercially pure titanium can
offer the option of preparing the surface of the
In Fig. 3 shows the simulation results for the billet, which combines the application of under
above values of variables in the context of the lubricating coating and technological lubricant.
stated objectives of the study.

Acknowledgement

The author A.G. Raab accord thanks for the

financial support allocated for research in the
framework of the project part of the State
assignments in the academic sphere, the Russian
Ministry of education and science No.
11.729.2014K.
Fig. 3. The result of the simulation process, the ECAP-
Conform after solving the inverse problem: the field The authors V.I. Semenov and L.Sh. Shuster
distribution of the accumulated plastic deformation. accord thanks for the financial support allocated
Force deformation Рaverage = 12,5 kN. for research in the framework of the joint
Russian-Taiwanese research project RFBR-
Thus, universal preparation of the surface of the MOST No. 15-58-52015 NNС_a.
billet with minimum value of the force of
deformation. On this basis, for the practical S.-J. Huang gratefully acknowledges the financial
implementation process of ECAP-Conform support allocated for research in the framework
commercially pure titanium can offer the option of the project MOST- 104-2923-E-011-003-MY3.
of preparing the surface of the billet, combining
under lubricating coating and technological
lubricant. Thus the rheological properties of REFERENCES
such combination must match the material with
high shear stress in the area of sliding contact. [1] R.Z. Valiev, Nanostructure bulk metallic
This assumption requires further research. materials: production, structure and properties.
R.Z. Valiev, I.V. Alexandrov – Moscow: ICC
"Akademkniga", 2007. (in Russian)
CONCLUSIONS [2] V.M. Segal, V.I. Reznikov, V.I. Kopylov, Processes
of plastic structure formation of metals/Minsk:
The result is a virtual full factorial experiment Science and technology, 1994. (in Russian)
was established that the most significant [3] V.M. Segal, Engineering and commercialization
influence on the strength of deformation have the of equal channel angular extrusion, Mater. Sci.
factor of friction in sliding contact of the billet and Eng. A., vol. 386, no. 1-2, pp. 269–276, 2004.

131
 V.I. Semenov et al., Tribology in Industry Vol. 38, No. 1 (2016) 127-132

[4] G.I. Raab, R.Z. Valiev, Equal channel angular Frictional Contact, Tribology in Industry, vol. 36,
extrusion of long billets, Nonferrous metallurgy, no. 1, pp. 49-66.464, 2014.
vol. 5, pp. 50-53, 2000. (in Russian)
[8] F.S. Novick and Y.B. Arsov, Process Optimization
[5] O.I. Abdullah and J. Schlattmann, Finite Element technology metals by methods of experiments
Analysis of Temperature Field in Automotive planning. M: Mechanical Engineering; Sofia:
Dry Friction Clutch, Tribology in Industry, vol. 34, Technique, 1980. (in Russian)
no. 4, pp. 206-216, 2012.
[9] Yu.P. Adler, Predplanirovanie experimenta. -M.:
[6] S. Petrović Savić, D. Adamović, G. Devedžić, B. Znanie, 1978. (in Russian)
Ristić and A. Matić, Contact Stress Generation on
[10] Yu.P. Adler, Introduction to design of
the UHMWPE Tibial Insert, Tribology in Industry,
experiments. M.: Metallurgy, 1969. (in Russian)
vol. 36, no. 4, pp. 354-360, 2014.
[11] P.I. Polukhin, G.J. Gun and A.M. Galkin, Resistance
[7] A. Belhocine, A.R. Abu Bakar and M. Bouchetara,
to plastic deformation of metals and alloys. M.:
Numerical Modeling of Disc Brake System in
Metallurgy, 1976. (in Russian)

132