Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
1
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
2
Sankt Lorenzen 36, 8715, Sankt Lorenzen, Austria
Mechanics, Materials Science & Engineering Journal
May 2016
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
3
Mechanics, Materials Sciences & Engineering Journal
, Austria, Sankt Lorenzen, 2016
Mechanics, Materials Science & Engineering Journal (MMSE Journal) is journal that deals in peer-
reviewed, open access publishing, focusing on wide range of subject areas, including economics,
business, social sciences, engineering etc.
MMSE Journal is dedicated to knowledge-based products and services for the academic, scientific,
professional, research and student communities worldwide.
Open Access model of the publications promotes research by allowing unrestricted availability of
high quality articles.
All authors bear the personal responsibility for the material they published in the Journal.
The Journal Policy declares the acceptance of the scientific papers worldwide, if they passed the
peer-review procedure.
Editor-in-Chief Mr. Peter Zisser
Dr. Zheng Li, University of Bridgeport, USA
Prof. Kravets Victor, Ukraine
Ph.D., Shuming Chen, College of Automotive Engineering, China
Dr. Yang Yu, University of Technology Sydney, Australia
Prof. Amelia Carolina Sparavigna, Politecnico di Torino, Italy
ISSN 2412-5954
e-ISSN 2414-6935
Design and layout: Mechanics, Materials
Science & Engineering Journal, www.mmse.xyz
Technical support: hotmail@mmse.xyz
©2016, Magnolithe GmbH
© Copyright, by the authors
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
4
CONTENT
I. Materials Science MMSE Journal Vol. 4 ..................................................................................... 6
Experimental Characterization of Innovative Viscoelastic Foams. Massimo Viscardi,
Maurizio Arena .................................................................................................................................... 7
Optimization Of Tribological Properties Of Aluminium Honeycomb Reinforced Polymeric
Composites Using Grey Based Fuzzy Algorithm. K.Panneerselvam, K.Lokesh, Chandresh D.,
T.N.S.Ramakrishna ............................................................................................................................ 15
Formation And Distribution of Brittle Structures in Friction Stir Welding of AA 6061 To
Copper. Influence of Preheat. Seyed Vahid Safi, Hossein Amirabadi, Mohammad Kazem
Besharati Givi .................................................................................................................................... 25
Isothermal Pneumo-Forming of Hemispherical Parts Made Out of Anisotropic Materials In
Short-Term Creep Mode. S.N. Larin, V.I. Platonov, Nuzhdin G.A. ................................................ 34
MHD Stagnation Point Flow in a Boundary Layer Of a Nano Fluid Over a Stretching Sheet
in the Presence of Viscous Dissipation and Chemical Reaction. Ch. Achi Reddy, B. Shankar .... 45
Quality Characteristics of Cutting Surfaces in the Milling of the Titanium Alloy
Ti10V2Fe3Al. Michael Storchak, Lucas Saxarra, Like Jiang, Yiping Xu, Xun Li ............................ 57
A Comparison between Dual Phase Steel and Interstitial Free Steel Due To the Springback
Effect. E.A. Silva, L.F.V.M. Fernandes, N.A.S. Sampaio, R.B. Ribeiro, J.W.J. Silva, M.S. Pereira . 71
II. MECHANICAL ENGINEERING & PHYSICS MMSE JOURNAL VOL. 4 .......................................... 81
On Application of the Ground Effect For Highspeed Surface Vehicles. Kravets Viktor V.,
Kravets Vl.V. & Fedoriachenko S.A. ................................................................................................. 82
On the Comparison Between the Approximate And Precise Methods of Piled Raft
Foundation Analysis. Abbasali Taghavi Ghalesari ......................................................................... 88
Dynamic Stress and Strain Analysis for 8x4 Truck Frame. Nagwa Ahmed Abdel-halim ...... 96
Identification of Standing Pressure Waves Sources in Primary Loops of NPP with WWER
and PWR. K.N. Proskuriakov, A.I. Fedorov, M.V. Zaporozhets and G.Y. Volkov ......................... 109
Methods for Solving a Stress Behaviour of Welded Joints under Repeated Loads. Semrád K.,
Čerňan J. .......................................................................................................................................... 128
Numerical Modelling of Basin Type Solar Stills. Nguyen The Bao ....................................... 133
VII. ENVIRONMENTAL SAFETY MMSE JOURNAL VOL. 4 ............................................................. 148
HAVS and HAV-nots: Investigating Resonance in the Human Arm Caused by Contact
with Machinery. Irina Viktorova, Matthew Fleck and Muhammed Kose ...................................... 149
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
5
Monitoring the Natural Factors Influence on Vegetation Development by Using Moderate-
Resolution Imaging Spectroradiometer (Modis) Images with OBIA Method in Uzbekistan.
Sh. B. Akmalov, J. V. Gerts, D. B. Omonov ..................................................................................... 156
An Empirically Derived Arc Flash Discharge Energy Model and Comparison to Established
Safety Codes. Irina Viktorova & Michael Bates ............................................................................. 160
On Development of a New Filtering Half-Mask. S.I. Cheberyachko, D.I. Radchuk, Y.I.
Cheberyachko, M.O. Ziborova ......................................................................................................... 164
Fluid Injection Induced Seismicity in the Oil and Gas Field Areas: Monitoring and
Modelling. A. Zabolotin, A.V. Konovalov, A.A. Stepnov, A.S. Sychov, D.E. Tomilev ..................... 170
The Effect of Microwave Radiation on the Properties of Canola Seeds.
Roudane M., Hemis M...................................................................................................................... 179
IX. ECONOMICS & MANAGEMENT MMSE JOURNAL VOL. 4........................................................ 187
Human Capital and Growth of E-postal Services: A cross-country Analysis in Developing
Countries. Dalibor Gottwald, Libor Švadlenka, Hana Pavlisová ................................................ 188
X. PHILOSOPHY OF RESEARCH AND EDUCATION MMSE JOURNAL VOL. 4 ................................. 210
Systematic Analysis and Synthesis of Integral Estimations of Bachelors’ Training in the Field
of Financial Monitoring. Alena Gaibatova, Grigory Krylov1, Ilya Seryy, Anastasiia Vorobeva,
Konstantin Vorobev.......................................................................................................................... 211
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
6
I. M a t e r i a l s S c i e n c e
M M S E J o u r n a l V o l . 4
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
7
Experimental Characterization of Innovative Viscoelastic Foams
Massimo Viscardi
1
, Maurizio Arena
1
1 University of Naples “Federico II”, Department of Industrial Engineering, Via Claudio, 21 80125 Napoli, Italy
DOI 10.13140/RG.2.1.5150.6325
Keywords: noise control, automotive, damping, foam, viscoelasticity.
ABSTRACT. The evolutionary trend in the automotive industry has produced over time numerous performance and
aesthetic innovations, however, the exponential development related to transportation technologies also introduced new
requirements concerning the environmental impact [1]. The awareness of ecological issues has led to a reorganization of
the evaluations and the vehicle design, currently aimed at reducing the problems that have emerged in empirical
investigations and the parallel increase in environmental solutions. The vehicle renewal process involves targeted
technical mutations both to observance of ecology as to the safety and comfort of the driver. New recyclable materials
and more resistant have been developed in order to minimize the environmental impact of the vehicle even at the end of
the operating life of its components, as well as solutions relating to the reduction of noise pollution generated as a response
to the requirements of comfort. Modern research programs on a global scale have set themselves the objective of
exploiting the potentiality of innovative technologies in the optimization of vehicles efficiency, the noise reduction and
in the consequent reduction of fuel burn. One of the crucial topics in the greening of the new generation automotive sector
is therefore the use and development of high vibro-acoustic performance materials. The goal of this research is properly
focused on the analysis of viscoelastic materials appointed to increase the damping of the vibrations generated in a vehicle.
The use of a viscoelastic material in this context is due to its high property to convert vibrational energy into heat,
providing a significant dissipation of the vibrations. Trade-off analyses are performed in order define the stiffness and
damping capacity of several viscoelastic foams with different thickness and density.
Introduction. The purpose of the present work is the
experimental investigation of new materials and
technologies to reduce the noise and vibrations
produced inside motor vehicles. It is known, that there
are several noise causes of the vehicles, both in the
cockpit and in the environment: those of great impact
are related to the engine and to the tyre/road
interaction processes, that induce noise into the
cockpit both through a structure-borne path
contribution (mainly the vibration of the car floor
induced by the mechanical forcing of the car body
floor) and an air-borne path contribution. In Figure 1
these noise sources are evident for a moving car.
The new low-consumption engines also provide a design, which causes a higher specific noise than
the previous ones: new soundproofing solutions are therefore necessary. In this research contest,
innovative materials, targeted at the reduction of the car-body floor, will be analysed and compared
to “standard one”; in the specific, viscoelastic foams will be investigated as a valid alternative to
conventional add-on damping element as generally used in these applications [2]. The investigation
of this class of viscoelastic material in this context is due to the high ability to convert vibration
energy into thermal energy, providing a significant increase of vibration damping, hence this aspect
will be study in terms of both static and dynamic stiffness and damping factor [3-6].
Nomenclature
Logarithmic decrement
FE
F
F
FRF
Finite Element
Force
Frequency
Frequency Response Function
g-force
k
t
w
ζ
Static stiffness
Time
Static deflection
Damping ratio
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
8
Fig. 1. Vibration distribution color map
Fig. 2. Technological application
Standard foams are already used as a part of the car-body carpet element, but their role is mainly the
decoupling of the carpet from the floor; basic idea of the research is to force this element to strongly
contribute to the vibrational energy dissipation.
Within the paper, viscoelastic foams with different physical properties (density, thickness, structure)
will be compared. The first experimental step concerns static stiffness measurements of the several
viscoelastic foams.
In the second phase, damping characteristics of each foam have been carried out by the use of modal
testing. Furthermore, in some cases, the dynamic stiffness has been measured for comparison with
the static one.
Experimental measures. The experimental measurements performed in this research have allowed
estimating the properties of stiffness and damping of innovative viscoelastic foams with different
densities and thicknesses.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
9
Static stiffness. In this section, the results of laboratory tests in order to measure the stiffness
coefficient (1) will be explained.



.
(1)
The experimental investigation concerns the comparison of original foam (intended as the standard
foam already used in most of the automotive applications) 65-30 and two other viscoelastic foams
65-30, 75-30 where these two digit represent respectively density (Kg/
) and thickness (mm).
The static stiffness of each viscoelastic foam was evaluated in correspondence of different
compression load settings by means of a test facility, shown in Fig. 3.
Fig. 3. Static test facility
In Figure 4 the trends of load versus static displacement have been plotted with reference to each case
of investigation.
Fig. 4. Load-Displacement Curve
0
1
2
3
4
5
6
0 2 4 6 8
Load (Kg)
Displacement, w (mm)
Original
65-30
75-30
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
10
Table 1. Static stiffness measure
ID Foam
Slope
(Kg/mm)
Static stiffness, k
(N/mm)
Original 65-30
3
29.43
Visco 65-30
0.7
6.867
Visco 75-30
9
88.29
Dynamic stiffness. The dynamic stiffness is defined as the ratio between the dynamic force and the
dynamic displacement: it is the quantity that expresses the elastic capacity of a material subjected to
a harmonic stress. An excitation source, an accelerometric transducer, a test material (viscoelastic
foam), and a rigid support plate are necessary to perform a dynamic test. The dynamic stiffness is
comparable with the static value as a result of spectral analysis, as can be seen in Figure 5 for the
Visco 75-30 foam.
Fig. 5. Dynamic stiffness, Viscoelastic Foams 75-30
For the characteristic curve, 1024 points were chosen in a 0-100 Hz frequency range. It is clear that
the stiffness of the foam 75-30 in correspondence of 50 Hz (first mode of vibration of the structure),
is about 93500 N/m, next to the respective static stiffness value.
Modal Analysis. For the purposes of the frequency response measurement an LMS TestLab system
has been used; 9 acquisition points have been defined for the mode shape reconstruction by the use
of the rowing hammer technique. The goal of the spectral test is to compare the structure without
material (Baseline) with the coated structure (viscoelastic foam) so as to discriminate the dynamic
response of each foam. In Figure 6 the setup made for the dynamic test, is shown.
0
20000
40000
60000
80000
100000
120000
0 20 40 60 80 100
Dynamic Stiffness (N/m)
Frequency (Hz)
75-30
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
11
Fig. 6. Riding plate in free-free conditions
In Figure 7 the first elastic mode shape of the metal plate, which the resonance frequency is, about
48 Hz is represented.
Fig. 7. Baseline configuration first mode shape, f = 47.8 Hz
The following figures show the most significant frequency responses (FRF) of the tested materials.
Fig. 8. Frequency Response Function (FRF)
0
10
20
30
40
50
60
70
80
0 20 40 60 80 100
Amplitude (g/N)
Frequency (Hz)
Baseline
Original
65-30
75-30
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
12
Fig. 9. Zoom about the first resonance frequency bandwidth
The results processing leads to observe a significant reduction of the resonance peak mainly due to
the 65-30 and 75-30 viscoelastic foams. The added mass is perceived as a shift of the transfer function
curve in virtue of (2):

(2)
Half-Power bandwidth method. The system response close to the resonance region is strictly
dependent on the damping. To estimate damping factor from frequency domain, the half-power
bandwidth method is usable. In this method, two point corresponding to 3 dB down from the
resonance peak are considered.
The damping factor ζ, is so defined as:
 
(3)
Where f
1
and f
2
represent the cut-off frequencies at the two points with an amplitude of 3 dB under
the resonance value, f
n
is the value of the natural frequency. In Table 2 the damping coefficients
obtained by that method for some of the foams are reported.
Table 2. Trade-Off damping coefficients, Half-Power Bandwidth
Baseline
Original Foam
Visco Foam
65-30
Visco Foam
75-30
Frequency (Hz)
47.8
47.3
44.7
44.5
Max Amplitude (g/N)
55.4
43.1
15.7
14.5
Damping Ratio, ζ
0.11
0.19
0.56
0.60
0
10
20
30
40
50
60
70
80
40 42 44 46 48 50
Amplitude (g/N)
Frequency (Hz)
Baseline
Original
65-30
75-30
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
13
Logarithmic decrement method. To estimate the value of damping factor in time domain through
logarithmic decrement method it is necessary to know peak amplitude in two consecutive points, Y
1
and Y
2
.
δ = ln
.
(4)

.
(5)
Fig. 10. Time History, Baseline
Fig. 11. Time History, Original Foam
Fig. 12. Time History, 65-30
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
14
Fig. 13. Time History, 75-30
Table 3. Trade-Off damping coefficients, Logarithmic Decrement
Baseline
Original Foam
Foam 65-30
Foam 75-30
Logarithmic decrement,
δ
0.77
0.96
3.01
3.52
Damping Ratio, ζ
0.12
0.15
0.48
0.56
Summary. Acoustic and vibrational aspects are becoming central in many engineering field as those
including automotive application and many transportation systems where the research of light-
weighting construction solutions is a demanding aspect. Along the presented research, some visco-
elastic foams have been studied as a possible mean to reduce the vibration induced noise inside a
vehicle; the use of these foams could lead to the overall weight reduction because of the elimination
of extra treatment nowadays used for this specific target.
It has been assessed that the use of viscoelastic materials brought significant benefits in terms of
vibration’s damping that has been approximately measured four times than the standard commercial
solution. As regard the weight aspects, these viscoelastic foams are porous by nature, they have low
density, and therefore they are particularly suitable for light-weighting application.
For the next developments, a numerical FE model will be developed to be correlated with the
experimental one. Also direct acoustic measurements (to directly evaluate the radiated power from
the panel) will be performed through a piezo-electric as an input source and a microphone as a
transducer for the acquisition.
References
[1] D. Roylance, Mechanical properties of materials, pp. 37-40, 2008.
[2] D. Roylance, Engineering viscoelasticity, pp. 8-14, October 24, 2001.
[3] M. Carfagni, E. Lenzi, M. Pierini, The loss factor as a measure of mechanical damping.
[4] M. Akay, Introduction to polymer science and technology, 2012.
[5] Ricci, F., Viscardi, M., Dynamic behaviour of metallic and composite plates under in-plane loads
(2000) Proceedings of the International Modal Analysis Conference - IMAC, 1, pp. 99-103.
[6] Siano D., Viscardi M., Napolitano P.,Panza, M.A., Numerical and experimental acoustic
performance investigations of a high-speed train composite sandwich panel, WSEAS Transactions
on Applied and Theoretical Mechanics, Volume 9, Issue 1, 2014, Pages 290-300.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
15
Optimization Of Tribological Properties Of Aluminium Honeycomb Reinforced
Polymeric Composites Using Grey Based Fuzzy Algorithm
K.Panneerselvam
1
, K.Lokesh
1
, Chandresh D.
1
,T.N.S.Ramakrishna
1
1 Department of Production Engineering, National Institute of Technology, Tiruchirappalli-620015, India
DOI 10.13140/RG.2.1.4683.1762
Keywords: fuzzy logic, grey relational analysis (GRA), nylon 6, optimization, polypropylene (PP), tribology.
ABSTRACT. In this research two composite materials were fabricated by using two matrix materials and one common
reinforcement material. The two matrix materials were Polypropylene and Nylon 6. Reinforcement material was
Aluminium honeycomb core. Compression moulding machine was used for the fabrication of these composite materials.
Two body abrasive wear experiments were conducted by using a pin-on-disc Tribotester under dry sliding condition and
at room temperature. The design process parameters for two-body abrasive wear test were normal load, sliding velocity,
sliding distance and abrasive paper grit size. The output responses were Coefficient Of Friction (COF) and Specific Wear
Rate (SWR). The design of experiments is based on L
9
Taguchi orthogonal array. Grey fuzzy logic algorithm was used
for the optimization of input process parameters. For polypropylene composite material the highest Grey Fuzzy Reasoning
Grade (GFRG) is obtained at 30 N normal load, 0.523 m/s sliding velocity, 450 m sliding distance, 320 grit size of abrasive
paper and these are the optimum level of process parameters. For nylon composite material highest GFRG is obtained at
30 N normal load, 1.046 m/s sliding velocity, 150 m sliding distance, 400 grit size of abrasive paper and these are the
optimum level of process parameters. The optimum level of process parameters were also validated with conformation
experiments.
Introduction. Literatures revels that there are different approaches to handle prediction and multi-
objective optimization problems. The common approaches includes Response Surface Methodology
(RSM), [1,2], Artificial Neural Network (ANN), Genetic Algorithm (GA), [3], Fuzzy regression, [4]
and Desirability Function (DF) approach, [5,6]. From last few years, polymeric composite materials
have been widely used in many industries. Some of polymer materials, mainly thermoplastics are
polypropylene (PP), Nylon etc., have shown grater improvement in their tribological and mechanical
properties. The major advantage of polymer composite from a tribological point of view is their low
wear rate [7]. Incorporation of inorganic particles into the polymers is best and an effective way to
fabricate polymer composite materials with improved tribological properties. The degree of the
reinforcement of the filler is depends up on the many factors like composition of the filler material,
size of the filler, shape of the filler, matrix and filler bonding etc. [8]. The incorporation of the fillers
can improve the tribological application depends on the type of application i.e., friction coefficient
and wear resistance were not the properties of the base material. In particular brake pads and clutches
require high coefficient of friction and low wear resistance whereas gears and bearings require low
coefficient of friction and wear rate [9]. The major advantage of the polymer composites are of their
self-lubricating property in tribological point of view.
The present research work analysis based on fuzzy-logic finds applications in uncertain environment
conditions. From last few years, fuzzy-logic-based multi-criteria optimizatin methods were mostly
using in optimization of tribological parameters. It is also shoewξ that application of fuzzy logic
technique improves the multi objective responses. Grey Relational Analysis (GRA) has strong
potential to improve the capability of fuzzy-logic in multi-objective optimization problems. In grey
fuzzy logic technique the optimization of multiple responses can be effectively changed into single
Grey Fuzzy Reasoning Grade (GFRG).
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
16
Experimental Details. Material preparation. Polypropylene and Nylon 6 were used as matrix
materials and aluminium honeycomb was used as a reinforcing material. Aluminium honeycomb core
made of alloy 3003 H18 of thickness 4.9 mm and cell size of 6 mm were used. Compression Moulding
was carried out to produce plates of dimension 150 x 150 x 9 mm
3
. Initially the empty die was placed
between the top and bottom heaters and it was pre heated up to 200°C for polypropylene and 250°C
for Nylon 6 material. When the required temperature was reached the Aluminium honeycomb core
and Polypropylene sheet materials were kept inside the die and heated for 10 mins. The compression
moulding pressure was set to 150 bars. Then pressure was applied in such a way that Aluminium
honeycomb core can be reinforced between Polypropylene sheets. When the required temperature
was reached the AHC and Nylon 6 GF composite granules materials were kept inside the die and
heated for 10 mins. The compression moulding pressure was set to 150 bars. Then pressure was
applied in such a way that Aluminium honeycomb core can be reinforced with Nylon 6 GF composite
granules. After heating for the pre-set time it was air cooled for 24 hours for both materials.
Experimental design. The process parameters considered for friction and wear test were normal load
(N), sliding velocity (m/s), sliding distance (m), abrasive paper grit size (µm) and the output responses
were coefficient of friction (µ) and specific wear rate (mm³/N-m). Total of 9 experimental runs were
conducted to take an interaction of the process parameters. The process parameters and their different
levels for L
9
orthogonal design is given in the Table 1.
Table1. Factors and different levels for L
9
orthogonal design.
Process Parameters
Levels
1
2
3
A
Applied Load (N)
10
20
30
B
Sliding Velocity (m/s)
0.523
1.046
1.569
C
Sliding Distance (m)
150
300
450
D
Abrasive Paper Grit Size
180
320
400
Experimental setup. Friction and wear test for Aluminium honeycomb reinforced PPG composite
material was conducted by using DUCOM Pin-On-Disc Tribotester as shown in the Fig. 1 according
to G99-05 standard in dry sliding conditions at room temperature. The samples were cut in the
dimensions of 9×9×4 mm. The material specimens were glued to the metal pin. By using loading
lever load can be applied on the specimen. Force sensor can measure the frictional force. The samples
were weighted before and after the experiments by using electronic balance with an accuracy of
0.0001 g. The formula used for calculation of specific wear rate was,
K
0
(w
1
-w
2
)/ (ρ×S
d
×L) (1)
Where, w
1
and w
2
are the weight of the sample before and after the abrasion test in gm, ρ is the density
of the composite material, K
0
is the specific wear rate in mm
3
/N-m, S
d
is the sliding distance in meters,
and L is the load in N. The Experimental results were shown in the Table 2 and Table 3 for
polypropylene and Nylon 6 composite materials respectively.
Grey based fuzzy logic. Grey relational analysis (GRA). The Grey relational analysis is a sub part
of grey system which was proposed by Deng in 1982 [10,11]. Taguchi method is used to optimize
single response characteristics while Grey relational analysis was used to multi objective
optimization. Therefore Grey relational analysis is somewhat complicated [12,13]. In grey relational
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
17
analysis the output responses obtained from the conducted experiments were normalized in between
0-1 using lower the better characteristics by using following equation.
Fig. 1. Schematic diagram of Pin-on-Disc Tribotester.
Table2. Experimental results of polypropylene composite material.
Expt.
No.
Applied
Load (N)
Sliding
Velocity(m/s)
Sliding
Distance(m)
Abrasive
paper
size(µm)
Coefficient of
friction(µ)
Specific wear
rate (mm
3
/N-m)
1
10
0.523
150
180
0.73
0.0629
2
10
1.046
300
320
0.64
0.0257
3
10
1.569
450
400
0.58
0.0140
4
20
0.523
300
400
0.66
0.0147
5
20
1.046
450
180
0.59
0.0090
6
20
1.569
150
320
0.54
0.0247
7
30
0.523
450
320
0.49
0.0068
8
30
1.046
150
400
0.45
0.0193
9
30
1.569
300
180
0.56
0.0086
x
ij
=










, (2)
where x
ij
the sequence after data processing;
η
ij
the original sequence of S/N ratio (where i=1,2,3…m, j= 1,2,3…n);
max η
ij
largest value of η
ij
;
min η
ij
smallest value of η
ij
.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
18
Table3. Experimental results of Nylon 6 composite material.
Expt.
No.
Applied
Load (N)
Sliding
Velocity(m/s)
Sliding
Distance(m)
Abrasive
paper
size(µm)
Coefficient of
friction(µ)
Specific wear
rate (mm
3
/N-m)
1
10
0.523
150
180
0.62
0.0707
2
10
1.046
300
320
0.55
0.0287
3
10
1.569
450
400
0.51
0.0145
4
20
0.523
300
400
0.59
0.0151
5
20
1.046
450
180
0.54
0.0095
6
20
1.569
150
320
0.49
0.0267
7
30
0.523
450
320
0.45
0.0078
8
30
1.046
150
400
0.38
0.0204
9
30
1.569
300
180
0.51
0.0092
In the second step grey grey relational coefficient is calculated using Eq. (3)
ξ
ij
=
















(3)
where
ideal normalized S/N ratio for the ίth performance characteristic;
ζ distinguishing coefficient which is in the range 0ζ 1.
The value of the grey relational coefficient is higher it implies that the corresponding process
parameter is closer to the optimum value.
Finally the grey relational grade can be calculated by using the Eq. (4)

(k), (4)
where
grey relational grade;
n number of process responses.
Rule based Fuzzy modelling. In GRA, each and every response is characterized as either ‘nominal-
the-better’ or ‘lower-the-better’ or ‘higher-the-better’ quality characteristics and this results indicates
some level of uncertainty. This level of uncertainty can be efficiently investigated with help of fuzzy-
logic technique [14,15]. Thus complex multi-objective optimization problem can be solved by
combining fuzzy-logic and GRA techniques. Fuzzy-logic technique consists of a fuzzifier,
membership functions, fuzzy rule base, inference engine and defuzzifier [16]. Out of these elements,
the fuzzifier uses membership functions to fuzzify the GRC of input parameters, (X1 = grey relation
coefficient for coefficient of friction, X2 = grey relation coefficient for specific wear rate).
Membership function was used to map the values of inputs (X1 and X2) and output (Y1=GFRG)
parameters in the range of 0 to 1. The structure of two inputs and one output Fuzzy logic unit as
shown in the Fig. 2 Fuzzy reasoning can be performed by the inference engine to generate a fuzzy
value of the fuzzy rules. The rules writing procedure can be described as follows. In this experimental
work nine fuzzy rules can be written for two inputs and one output are developed for inference.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
19
Rule 1 : if x1 is A1;and x2 is B1; then y is C1;else
Rule 2 : if x1 is A2;and x2 is B2; then y is C2; else.. .
Rule n : if x1 is An;and x2 is Bn; then y is Cn
Ai, Bi, and Ci are fuzzy subsets as per the corresponding membership functions.
Finally the fuzzy value can be converted into the crisp output with the help of deffuzifier by using
centroid defuzzification method (Eq. 5).
Yo =


, (5)
where Yo Grey Fuzzy Reasoning Grade (GFRG).
Fig. 2. Structure of two input and one output Fuzzy logic unit.
Fig. 3. Grey fuzzy logic method.
The steps of grey-fuzzy-logic method are illustrated in Fig. 3 and described as follows:
1. The experimental values of coefficient of friction and specific wear rate are normalized in
between 0 and 1.
2. Grey relational coefficient (GRC) is calculated for each and every response.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
20
3. Then fuzzy-logic technique is applied. In this technique the fuzzifier uses the membership
functions to fuzzify the Grey relational coefficient (GRC) of each response.
4. Fuzzy rules (if-then control rules) can be generated and finally fuzzy predicted value converted
into Grey Fuzzy Reasoning Grade (GFRG) by using defuzzifier.
Results and Discussion. For ecah and every combination of parameters overall grey relational grade
and grey relational coefficients were calculated. By using MATLAB (R2015a) grey-fuzzy reasoning
grade is obtained with help of fuzzy logic tool box. Grey relational coefficients of coefficient of
friction and specific wear rate were the inputs to the fuzzy logic tool box. For fuzzy modelling,
triangular shaped membership function was used. The names of membership functions are LOWEST,
LOWER, LOW, LOW MEDIUM (LM), MEDIUM, HIGH MEDIUM (HM), HIGH, HIGHER, and
HIGHEST are used to indicate the grey relational coefficients (GRC) of input variables and output
variable grey relational grade (GRG). The membership functions used in this fuzzy logic tool box can
be shown in Fig. 4 and Fig. 5. The fuzzylogic rule viewer can be shown in Fig. 6. In fuzzy rule viewer
nine rows indicates the nine fuzzy rules and the first two columns indicates two input variables i.e.
grey relational coefficients of coefficient of friction and specific wear rate respectively. The final
column of fuzzy rule viewer gives the grey-fuzzy reasoning grade (GFRG) with defuzzification.
Aluminium honeycomb reinforced with Polypropylene Composite.
Table 4. Grey relational coefficients and grey relational grade.
Expt. No
Grey relational coefficient
Grey relational grade
Rank
COF
SWR
1
0.3333
0.3333
0.3333
9
2
0.4242
0.5974
0.5108
8
3
0.5185
0.7957
0.6571
5
4
0.4000
0.7802
0.5901
7
5
0.5000
0.9272
0.7136
4
6
0.6089
0.6104
0.6095
6
7
0.7777
1.0000
0.8888
1
8
1.0000
0.6917
0.8458
2
9
0.5600
0.9396
0.7498
3
Fig. 4. Membership function for coefficient of
friction and specific wear rate.
Fig. 5. Membership function for Grey relational
grade.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
21
Fig. 6. Fuzzy logic rules viewer.
Table 5. Grey Fuzzy reasoning grades and their Rank.
Expt. No
Grey Fuzzy Reasoning Grade (GFRG)
Rank
1
0.355
9
2
0.542
8
3
0.681
5
4
0.611
7
5
0.749
4
6
0.620
6
7
0.865
1
8
0.819
2
9
0.750
3
The GFRG values of all the experiments were shown in Table .5 From Table. 5 , it shows that the
seventh experiment combination of parameters will be the optimum combination of parameters
having highest grey-fuzzy reasoning grade. So, the optimum combination of parameters were 30 N
normal load, 0.523 m/s sliding velocity, 450 m sliding distance, 320 grit size of abrasive paper.
Confirmation test was conducted to validate the experimental results. The predicted GFRG for the
optimal combination of the parameters is calculated by using following equation:
Y
0
= Y
0m
+
 

(6)
where Y
0
estimated GFRG;
Y0
m
total mean GFRG;
Y0i mean value of GFRG at the optimal level;
n number of parameters affecting the multiple performance characteristics.
Table 6. Conformation test results.
Optimum process Parameters
Prediction
Experiment
Parameter Levels
A3B1C3D2
A3B1C3D2
GFRG
0.865
0.888
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
22
Conformation test results are shown in the Table. 6. At optimum combination of parameters
(A3B1C3D2) the estimated GFRG is 0.865 and experimental GFRG is 0.888. Therefore gain can be
observed in GFRG which shows that we acn use grey fuzzy logic algorithm for the multi objective
optimization.
Aluminium honeycomb reinforced with Nylon 6 Composite
Fig. 7. Membership function for coefficient
friction and specific wear rate.
Fig. 8. Membership function for Grey relational
grade.
Table 7. Grey relational coefficients and grey relational grade.
Expt. No
Grey relational coefficient
Grey relational grade
Rank
COF
SWR
1
0.3333
0.3333
0.3333
9
2
0.4137
0.6007
0.5072
8
3
0.4800
0.8243
0.6521
5
4
0.3636
0.8116
0.5876
6
5
0.4285
0.9487
0.6886
4
6
0.5217
0.6246
0.5731
7
7
0.6315
1.0000
0.8157
2
8
1.0000
0.7139
0.8569
1
9
0.4800
0.9573
0.7186
3
Fig. 9. Fuzzy logic rules viewer.
The GFRG values of all the experiments were shown in Table .8 From Table. 8 ,it shows that the
experiment 8 combination of parameters will be the optimum combination of parameters having
highest grey-fuzzy reasoning grade. So, the optimum combination of parameters were 30 N normal
load, 1.046 m/s sliding velocity, 150 m sliding distance, 400 grit size of abrasive paper.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
23
Table 8. Grey Fuzzy reasoning grades and their Rank.
Expt.No
Grey Fuzzy Reasoning Grade (GFRG)
Rank
1
0.353
9
2
0.530
8
3
0.661
5
4
0.595
6
5
0.674
4
6
0.585
7
7
0.792
2
8
0.834
1
9
0.704
3
Table 9. Conformation test results.
Optimum process Parameters
Prediction
Experiment
Parameter Levels
A3B2C1D3
A3B2C1D3
GFRG
0.834
0.856
Conformation test results are shown in the Table. 9. At optimum combination of parameters
(A3B2C1D3) the estimated GFRG is 0.834 and experimental GFRG is 0.856. Therefore gain can be
observed in GFRG which shows that we acn use grey fuzzy logic algorithm for the multi objective
optimization.
Summary. In present investigation an algorithm used which is a combination of fuzzy logic and grey
relational analysis for the optimization of multiple response process parameters. The following
conclusions can be drawn from the present research work.
The grey fuzzy based system indicates that the highest GFRG can be obtained at 30 N normal
load, 0.523 m/s sliding velocity, 450 m sliding distance, 320 grit size of abrasive paper for
Aluminium honeycomb reinforced with Polypropylene Composite material.
The grey fuzzy based system shows that the highest GFRG can be obtained at 30 N normal
load, 1.046 m/s sliding velocity, 150 m sliding distance, 400 grit size of abrasive paper for
Aluminium honeycomb reinforced with Nylon 6 Composite material.
The confirmation test also reveals that the grey-fuzzy logic system is applicable for
optimization of the multi responses in between the ranges of the process parameters.
References
[1] Gunaraj V, Murugan N, Prediction and optimization of weld bead volume for the submerged arc
process, Part-1, Journal of Welding 79 (10), 2000, 286-294.
[2] Gunaraj V, Murugan N, Prediction and optimization of weld bead volume for the submerged arc
process, Part-2, Journal of Welding 79 (11), 2000, 331-338.
[3] Al-Aomar, Raid, A robust simulation-based multi criteria optimization methodology, Simulation
Conference, Proceedings of the winter, Vol. 2. IEEE, 2002.
[4] Wang Jen-Ting, Jean Ming-Der, Optimization of cobalt-based hard facing in carbon steel using
the fuzzy analysis for the robust design, International Journal of Advanced Manufacturing
Technology 28, 2006, 909-18.
[5] Asiabanpour B, Palmer K, Khoshnevis B, An experimental study of surface quality and
dimensional accuracy for selective inhibition of sintering, Journal Rapid Prototyping, 2004, 10(3),
181-92.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
24
[6] Wu Ful-Chiang, Optimization of correlated multiple quality characteristics using desirability
function, Quality Engineering, 2005, 17(1), 119-26.
[7] Bekhet N.E, Tribological behaviour of drawn polypropylene, Journal of Wear, 1999, 5561.
[8] Wetzel, Bernd, F. Haupert, K. Friedrich, M. Q. Zhang, M. Z. Rong, Mechanical and tribological
properties of microparticulate and nanoparticulate reinforced polymer composites, Proceedings of the
ICCM-13, 2001, ID1021.
[9] Malucelli, Giulio, Francesco Marino, Abrasion resistance of polymer nanocompositesa
review, Abrasion resistance of materials, Intech-Open Access Publisher, Rijeka, 2012, 1-18.
[10] J.L. Deng, Control problems of grey systems, Systems & Control Letters, 1982, 288294.
[11] J.L. Deng, Introduction to grey system, Journal of Grey System, 1989, 124.
[12] P.N. Singh, K. Raghukandan, B.C. Pai, Optimization by grey relational analysis of EDM
parameters on machining Al10% SiCP composites, Journal of Materials Processing Technology,
2004, 16581661.
[13] M. Kurt, S. Hartomacioglu, B. Mutlu, U. Koklu, Minimization of the surface roughness and form
error on the milling of free-form surfaces using a grey relational analysis, Journal of Materials science
and Technology, 2012, 205213.
[14] S. Pattnaik, D.B. Karunakar, P.K. Jha, Multi-characteristic optimization of wax patterns in the
investment casting process using grey-fuzzy logic, International Journal of Advanced Manufacturing
Technology, 2013, 67 (5-8), 1577-1587.
[15] P. Asokan, R. Ravi Kumar, R. Jeyapaul, M. Santhi, Development of multi-objective optimization
models for electrochemical machining process, International Journal of Advanced Manufacturing
Technology, 2008, 39, 5563.
[16] Y. Tzeng, F. Chen, Multi-objective optimisation of high-speed electrical discharge machining
process using a Taguchi fuzzy-based approach, Journal of Materials and Design 28 (4) 2007, 11.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
25
Formation And Distribution of Brittle Structures in Friction Stir Welding of AA
6061 To Copper. Influence of Preheat
Seyed Vahid Safi
1,a
, Hossein Amirabadi
2
, Mohammad Kazem Besharati Givi
3
1 M.Sc. student. Mech. Eng., University of Birjand, Birjand, Iran
2 Assoc. prof., Mech. Eng., University of Birjand, Birjand, Iran
3 Assoc. prof., Mech. Eng., University of Tehran, Tehran, Iran
a vahid.safi@gmail.com
DOI 10.13140/RG.2.1.2620.9684
Keywords: friction stir welding, mechanical properties, preheat, intermetallic.
ABSTRACT. In this paper, apart from introducing brand new warm friction stir welding (WFSW) method, the effect
of preheating on friction stir welded of copper and aluminum alloys sheets and its influence on improving the mechanical
properties of the weld were investigated. Sheets of aluminum alloy 6061 and copper with thickness of 5mm were used.
The tool was made of tool steel of grade H13 with a threaded cone shape. Rotational speeds (
) of 1200-1400 rpm and
traverse speeds (
v
) of 50-100 mm/min were used for better understanding the behavior of the tools during the heat input.
The sheets were kept in furnace with temperature of 75 ˚C and 125˚C and welding was done afterwards. At last, tensile
and micro hardness tests were done to compare the mechanical properties of the welds. Considering to the high thermal
conductivity of both copper and aluminum, the reason of increase in strength of the joints could be related to the low
temperature gradient between the weld zone and base metal because the heat gets out of the stir zone with lower steep. A
significant increase in hardness is observed in the SZ for the following reasons: (i) the presence of concentric grains with
intensely refined recrystallization and (ii) the presence of intermetallic compounds. The tensile test results showed 85%
increase in the strength of preheated joints. The maximum strength occurs for preheating of 75˚C, rotational speed of
1200 rpm and traverse speed of 50 mm/min. In the present study, intermetallic compounds and the precipitates are moved
to the grain boundaries during the welding process. These precipitates act as strong obstacles to the movements of
dislocations and increase the deformation resistance of material. This phenomenon may result in locking of grain
boundaries and consequently decrease of grain size. This grain refinement can improve the mechanical properties of
welds. Accordingly, hardness and strength of the material will be increased.
1. Introduction. FSW is a process including many actions and interactions between a series of
simultaneous thermodynamic processes. These reactions are the results of heat input rate, cooling
rate, flow and deformation, dynamic recrystallization and integration of mechanical joints [1].This
method of welding was invented in 1991 by Thomas Wayne et al. in TWI in Cambridge, England
[2].FSW is a relatively new method in metal joining that because of solid state nature of this method
(no melting happens at joining) it has some advantages comparing with other fusion welding methods.
Low distortion and shrinkage, good mechanical properties, fewer defects and the ability of welding
some metals that cannot be welded with fusion welding methods are among the most important
advantages of this method [3]. In last few years, this process has been used in aerospace and aviation
industries, automotive industries, fuel tanks etc. in most developed countries. During this process,
imposes high lateral forces to the material during deformation which causes temperature to increase
up to about 70-80% of its melting temperature [4, 5].Copper has a wide range of applications because
of its great thermal and electrical conductivity, corrosion and fatigue resistance and good flexibility.
Copper alloys have various set of properties, which is dependent on the addition of elements and heat
treatment [6].
Today, aluminum and its alloys have a wide range of applications in defense industry, aerospace,
transportation, marine industry, construction industry, packaging and containers etc. some of the
properties which made aluminum and its alloys as one of the most economical and popular group of
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
26
metals are: Light weight, flexibility, physical properties, mechanical properties, corrosion resistance
[7]. During the improvement of FSW process, the conventional fusion welding methods for similar
and dissimilar materials were used. There are a few studies about FSW for dissimilar materials at the
moment. FSW of dissimilar materials such as aluminum and copper causes the formation of
intermediate brittle phases of Al/Cu, which still needs to be studied in the future [8, 9].
Ouyang et al [10]. Also studied the FSW on dissimilar metals such as AA6061 (T6) and copper. They
showed that joining of AA6061 to copper due to the brittle nature of intermetallic compounds in the
weld nugget, is hard. Mechanical mixture zone of dissimilar joint of AA6061 to copper is mainly
consisted of various intermetallic compounds including CuAl2, CuAl and Cu9Al4 with few amount
of α-Al and FCC solid solution of aluminum in copper.
On the other hand Galvao et al [11]. observed that welding with higher ω/v, increase the heat input
and causes the formation of mixture zones from materials with higher size and more homogenous.
The morphology of stir zone, type and the amount of intermetallic structures which are formed
because of thermo mechanical induction solid state are highly dependent to welding parameters. In
FSW of dissimilar metals, pin is an important factor in compare to the two metals.
Ratnesh and Pravin [12] welded AA6061 to copper with FSW. They obtained good joints with
changing the central line of the pin into the copper and in the advancing direction. They reported that
ultimate tensile strength of aluminum in copper joint is low due to the presence of intermetallic
compounds. Increase in the rotational speed lowers the tensile strength, which is mainly due to the
increase in the amount of intermetallic compounds, which are formed in the aluminum-copper
interface. Higher hardness in the stir zone in compare to the base metals is due to the formation of
hard and brittle compounds such as CuAl2, CuAl and Cu9Al4.
Akinalbi et al [13]. observed that interface of the joint are recognized with lamellar mixture of
aluminum and copper which are apparent in the structure and are due to heat input from the stir action
of the pin during the FSW process. They also observed that decrease in grain size towards the stir
zone of the welds increases.
Recent researches show that with placing the copper sheet, which has higher melting temperature in
the advancing side, a weld with good compounds can be obtained. However, many of these studies
show the presence of intermetallic compounds during FSW of aluminum with copper, but more study
is needed on new formed phases for better understanding the weld joint. Optimizing the parameters
of the process to decrease the formation of intermetallic compounds in the interface is also required.
2. Experimental procedure. Dissimilar FSW of 6061 aluminum alloy and pure copper sheets both
with the thickness of 5 mm was performed. The specimens were cut to the dimension of 70 mm×200
mm by wire cut. Due to the different behaviors of aluminum alloy and pure copper, in addition to the
welding and also complication of condition simultaneous with applying preheating condition, it has
been attempted to obtain more appropriate analysis of condition dominant in dissimilar joints welding
by considering a set of parameters affecting welding process. That is why a tool with frustum threaded
pin and also pin tendency towards copper sheet (Fig. 1) in 3 different travel and rotation speeds have
been used.
Parameters employed in dissimilar joining of copper and 6061 aluminum alloy sheets are shown in
Table 1. The joints were performed in 3 different temperature domains including room temperature,
75 ˚C and 125 ˚C.
After welding, three tensile test samples were cut from each welded sheet in a vertical direction to
the weld line using wire cut to investigate the strength of the joints. The tensile test samples were
prepared according to ASTM E8M standard. The tests were carried out using a Hounsfield tensile
testing machine under the crosshead speed of 1 mm min
-1
at room temperature. Microstructural
analysis was performed on the cross section perpendicular to the welding direction. A solution of Fe

(5g) + HCL (25 ml) +
O (75 ml) was used to etch the copper side of the joints, while the
aluminum side was etched by a reagent with a solution of 25ml H
and 75ml
O for 60 s. Optical
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
27
and scanning electron microscopy (SEM) equipped with energy dispersive X-ray spectroscopy (EDS)
were used in order to observe the microstructure and characterize the intermetallic compounds at the
joint interface. Furthermore, EDS line scan analysis was performed to follow the compositional
change across the intermetallic compounds existing at the interface of aluminum and copper sides of
the interface. Micro hardness measurements were also taken from the cross sections of both aluminum
and copper sheets at the welded area perpendicular to the welding direction by applying load of 25 gf
for 20 s duration loading time.
Fig. 1. Schematic illustration showing set-up of dissimilar FSW.
Table
1. Parameters employed for dissimilar joining of copper and 6061 aluminum alloy sheets
Geometry of tool pin
Frustum threaded pin
Shoulder diameter, Pin diameter,
Pin length
18- 6 - 4.8 mm
Tool rotation speed
1400 -1300-1200 rpm
Tool travel speed
50-75-100 mm/min
Tilt angle
3 degree
Tool offset
2 mm toward copper sheet
3. Result and Discussion.
3.1 Joints appearance. Fig. 2 shows surface appearances and cross sections of Al/Cu joints produced
at different traverse and rotational speeds as it is observed, the weld surface does not contain any kind
of defect and onion ring structure is clearly observable as well.
As can be observed in Fig. (2-e), within the stir zone, the dominant phase is aluminum and fine and
coarse particles of copper have been distributed within the aluminum matrix. Full turbulence of
material and displacement of copper toward aluminum is indicative of adequate flow in the stirring
zone. Welding area of welded joints is similar to the common form of nugget zone in FSW process
that is as elliptical. In the Fig. (2-d), an image of optimal cross sections (
= 1200
v
= 50) can be
observed. The boundary of Al/Cu is completely continuous and free of any hole or discontinuities.
The presence of small particles of copper in the aluminum side indicates that conditions have been
favorable for adequate stirring by means of tool in the weld zone. However, in this case, because of
the placement of copper on the advance side, the situation is different (Fig. 2-f). In this case, the
continuity of copper and stir zone is not desirable and in the border of copper, discontinuity and crack
are observed. Moreover, there is several holes in the nugget zone. Because of the discontinuity in the
boundary between the copper and the stir zone, the created joints is so weak in terms of mechanical
property and can be easily fractured in the boundary of Al/Cu.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
28
Fig. 2. Macroscopic view of weld surface and cross sections of joints at the welding conditions of:
(a, d):
=1200
v
=50; (b, e):
=1300
v
=75 and (c, f):
=1400
v
=100.
3.2 Material flow. In order to study the flow behavior of materials more precisely, the
microstructures of the nugget zone were examined in detail by SEM and results are shown in Fig. 3.
Different phases were evidenced and corresponding EDS analysis were carried out. The EDS results
were presented in table 2.
Fig. 3. SEM images of the dissimilar Al/Cu joints interface
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
29
Table. 2. EDS analysis results of reaction layer pointed in Fig. 3 (at-%)
Position
Al
Cu
Possible phases
A
83/48
16/52
Al+Al
2
Cu
B
11
84/45
Solid Solution
Cu(Al)
C
39/43
60/57
Al
2
Cu
3
D
43/56
56/44
AlCu
E
32/15
64/44
Al
4
Cu
9
As can be observed in Fig. (3-a), rotation of tool in the interface of the joint has caused enough stirring
and mixing of the copper and aluminum, that this mixture along with heat caused by the process of
preheating sheets, plastic deformation and friction have caused the separation of coarse particles of
copper and locating them in aluminum matrix. The result of this reaction is penetration of copper and
aluminum into each other particles together and formation of intermetallic compounds. In Fig. (3- b),
which is an image with a larger magnification compare with Fig. (3- a), the intermetallic compounds
can be clearly observed. Compounds that as the small and scattered parts have been distributed in the
boundary of the joint. In Fig (3-c), the sequential layers of copper and intermetallic compounds are
observed that reflects this fact that the aluminum particles are isolated from the aluminum sheet and
they convert to intermetallic compounds in the side of copper sheet because of suitable condition in
the terms of the amount of heat and enough time to react. Fig (3-d) shows the presence of halos with
different color besides copper layers that indicates the formation of different intermetallic compounds
in those areas. By an exact investigation of the joint area using EDS of chemical composition of
present components, the presence of fine particles of metallic compounds in aluminum matrix that
have been distributed heterogeneously (Fig. 3-e). According to the results of the analysis carried out,
the points of B and C (Fig. 3-f) are two different intermetallic compounds. Point B contains 5.13 wt-
% of aluminum and 92.67 wt-% of copper, while point C includes 21.65 wt-% aluminum and 78.35
wt-% of copper that indicates the fact that the conditions in point C is much more favorable in terms
of time and the amount of heat to create intermetallic compounds.
Compounds created in the D and E is directly related to the aluminum particles in the copper (Figure
3. g). The reason for this phenomenon and formation of different intermetallic compounds can be
stated in a way, that atoms isolated from copper sheet in the aluminum side and isolated aluminum
atoms in the copper side, according to the condition of joint such as the inlet temperature and time,
react with each other in the interface of the joint and cause formation of intermetallic layers with
different colors, that this difference in contrast is contributed to the differences in the chemical
composition of these components. Then this difference in chemical composition occurs because of
the penetration atoms into the interface of the joint. Therefore, it can be concluded that formation of
intermetallic compounds depends on the permeability of atoms and welding conditions. The newly
formed isolated structures and solid solution may be related to the local diffusion induced by the
preheating effect on the nugget zone. Based on the EDS analysis results, typical intermetallic were
observed in the lower part of the weld, including AlCu (D), Al
2
Cu
3
(C), and Al
4
Cu
9
(E).
3.3 Mechanical properties.
3.3.1 Micro hardness distributions. The results of the dissimilar Al/Cu joints hardness test can be
observed in Fig. 4.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
30
Distance from weld center/mm
Fig. 4. Hardness profiles along the traverse cross section of Al/Cu joint (ω=1200, ν=50) at three
different preheating temperatures
The above values signify different welding states including constant traverse and rotational speed,
and varied preheating conditions. The results suggest that in contrast to the base metal a significant
increase in hardness is observed in the SZ for the following reasons: (i) the presence of concentric
grains with intensely refined recrystallization and (ii) the presence of intermetallic compounds.
TMAZ has less hardness compared to the SZ, but due to the presence of different grading, it enjoys
more hardness compared to the base metal and the HAZ zone. Observations are indicative of the fact
that, by the application of preheating conditions, the rate of hardness significantly increases in the stir
zone, and reaches its maximum value at 75 degrees centigrade. One of the mechanisms increasing
the hardness and strength of the materials is precipitation hardening. In this mechanism, some of the
alloying elements combine with each other and form fine precipitates. These precipitates act as strong
obstacles to the movements of dislocations and increase the deformation resistance of material.
Accordingly, hardness and strength of the material will be increased. In the present study,
intermetallic compounds and the precipitates are moved to the grain boundaries during the welding
process. This phenomenon may result in locking of grain boundaries and consequently decrease of
grain size. This grain refinement can improve the mechanical properties of welds. Locking of grain
boundaries decreases the grain growth and grain boundary movement during the plastic deformation,
which in turn delays the occurrence of recrystallization. Moreover, in this condition, the
recrystallization temperature is probably higher than the temperature at which the precipitates are
formed. Another possible explanation is that the recrystallization process needs noticeable amount of
time, which is not provided during the welding process. Preheating of the specimens at 75 C provides
the driving force for precipitates to move to the metallurgical defects such as grain boundaries and
prevent the grain growth during the welding process. Therefore, the joint strength will be increased
due to the formation of finer grain by increasing of the preheating temperature from 75 to 125 C.
3.3.2 Tensile strength. Tensile test results are presented in Fig. 5 according to the applied travel
speed, rotational speed and preheating temperature. By considering to the tensile test results given in
Fig. 5, it can be found that the maximum strength for the specimens welded at room temperature is
related to the rotation and travel speeds of 1300 rpm and 100 mm/min, respectively. Fig. 5 shows the
effect of traverse speed on UTS. Welding speed has an inverse effect on tensile strength. Lower
0
50
100
150
200
250
300
350
Hardness(HV,
25 g)
Cu Advance Al Retreat
room temp 75 deg 125 deg
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
31
welding speed leads to greater amount of heat supplied to materials and therefore, it improves plastic
deformation of the material and the formation of the effective weld joint. However, increase in heat
caused loss of strength. Different behaviors of aluminum alloys and pure copper toward preheating
can be articulated as the main reason of this phenomenon. Because in a condition that heat input is
suitable for copper, aluminum alloy shows negative reaction to this heat or in the similar case, when
the generated heat to the welding area is appropriate for aluminum alloy, this amount of heat is
inadequate for copper stir and consequently suitable joint is not achieved. As a result, for dissimilar
joining of 6061 aluminum alloy and pure copper, a mediate condition toward different behaviors of
welding alloys should be considered. It was found from the result that the welding speed have more
signification effect on tensile strength.
Fig. 5. Tensile test result for dissimilar Al/Cu joints
0
20
40
60
80
100
50 75 100
48
44
25
53
65
85
50
39
38
UTS(MPa)
traverse speed(mm/min)
R O O M T E M P
1200 1300 1400
0
20
40
60
80
100
50 75 100
89
75
65
32
55
60
26
33
34
UTS(MPa)
traverse speed(mm/min)
7 5 D E G
1200 1300 1400
0
10
20
30
40
50
50 75 100
50
27
26
15
20
24
8
19
21
UTS(MPa)
traverse speed(mm/min)
1 2 5 D E G
1200 1300 1400
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
32
According to previous researches, ratio of tool rotation speed to its travel speed plays an important
role on turbulence of the plasticized material and consequently on microstructure and mechanical
properties of joints due to having direct impact on rates of generated heat. That is why this parameter
is known as one of the most important process parameters [17]. Increase in
v
ratio causes the
decrease of tool advancing length in each rotation. In another word, in a specified length of joint, tool
will have more rotation number. This issue will have an especial effect on materials flux in stir zone.
In fact, stir of materials will be increased by decreasing the advancing speed and increasing the tool
rotation speed. Changes in
v
ratio in welding process will have a considerable impact on
temperature peak and also its distribution. In fact, this is an important factor in microstructural
changes and thereby the eventual properties of these kinds of joints. As a result, changes in
v
variable are along with changes in joint quality.
The reasons of these changes can be justified in a way that reduction of
v
ratio will lead to decrease
in amount of heat interred in welding zone. Because the tool rotation number in one millimeter of
weld will be decreased and so, friction heat and heat derived from plastic work (due to strain
decreasing derived from
decrease) will be reduced. In this case, material flux around the tool will
be decreased and extent of turbulence zone will be reduced.
Summary. In this study, effect of preheating on mechanical properties of dissimilar friction stir
welded Al/Cu joints have been investigated. The following results were obtained:
Preheating of Al/Cu dissimilar joints improved the mechanical properties of the welds.
The hardness at the copper side of the nugget was higher than that at the aluminum side and
preheating has an effective role to improve it as well.
Based on the EDS analysis results, typical intermetallic were observed in the lower part of the weld,
including AlCu, Al
2
Cu
3
and Al
4
Cu
9
.
WFSW process is a novel method to introduce metallurgical and mechanical changes in materials
because of the uniform distribution of the particles in the stir zone.
Full stir of aluminum and copper was observed at the interface. Due to the presence of aluminum
particles in copper on the advancing side, flow of materials is upward on the bottom of the welded
plates, while it is downward in the retreating side.
References
[
1
] R. Sakano, K. Murakami, K. Yamashita, T. Hyoe, M. Fujimoto, M. Inuzuka, U. Nagao,H. Kashiki,
Kobe, Japan,2001.Development of spot FSW robot system for automobile body members,
in:Proceedings of the 3rd International Symposium of Friction Stir Welding
[2] Thomas WM, Nicholas ED, Needham JC, Murch MG, Temple-Smith P, Dawes CJ. Friction stir
butt welding. International Patent Application No. PCT/GB92/02203; 1991.
[3] Montheillet, D. Allehaux, 2006, “Mechanical and thermal modelling of Friction Stir Welding”,
Journal of Materials Processing Technology, 171, pp.348357.
[4] S.Benavides, Y.Li, L.E.Murr, D.Brown, J.C.Mcclure, 1999, ScriptaMater.Vol 41, pp. 809
[5] Y. Li, E.A. Trillo, L.E. Murr, J. 2000, Mater. Sci. Lett.Vol 19, pp. 1047
[6]. Kundig K.J.A., Cowie J.G. Copper and copper alloys. In: Myer K, editor. Mechanical engineers’
handbook. Wiley Interscience; 2006. p. 117-220.
[7] T.Chen, “Process parameters study on FSW joint of dissimilar metals for aluminium-steel”,
journal of materials science, Vol.44, 2009, pp.2573-2580.
[8] B.S. Yilba, A. Z. Sahin, N. Kahraman , A. Z. Al-Garni, “Friction stir welding of St-Al and Al-Cu
materials” J. Mater. Process. Technol., 1995, 49, 431-443.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
33
[9] J. Ouyang , E. Yarrapareddy , R. Kovacevic , “Microstractural evolution in the friction stir welded
6061 aluminium alloy ( T6- temper condition) to copper” , J. Mater. Process. Technol., 2006, 172,
110-122.
[10] Jiahu Ouyang, Eswar Yarrapareddy, Radovan Kovacevic, Microstructural evolution in the
friction stir welded 6061 aluminum alloy (T6-temper condition) to copper” Journal of Materials
Processing Technology, 2006, 172, pp 110122.
[
1
1] I. Galvao, J. C. Oliveira, A. Loureiro and D. M. Rodrigues, “Formation and distribution of brittle
structures in friction stir welding of aluminium and copper: influence of process parameters” Science
and Technology of Welding and Joining, 2011, 16, No 8, pp. 681- 689.
[
1
2] Ratnesh K. Shukla , Pravin K. Shah , “ Investigation of Joint Properties of Friction Stir Welding
of Aluminum 6061 Alloy to Copper” International Journal of Engineering Research and Technology,
2010, 3, Number 3, pp. 613620.
[
1
3] Esther T. Akinlabi, Stephen A. Akinlabi, “Effect of Heat Input on the Properties of Dissimilar
Friction Stir Welds of Aluminium and Copper” American Journal of Materials Science, 2012, 2(5),
pp. 147- 152.
[
1
4] M. Abbasi. Gharacheh, A. H. Kokabi, G. H. Daneshi, B. Shalchi and R. Sarrafi, “ The Influence
of the Ratio of “Rotational Speed/Traverse Speed” (
v
) on Mechanical Properties of AZ31 Friction
Stir Welds” , International Journal of Machine Tools and Manufacture, Pages 1983-1987, December
2006.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
34
Isothermal Pneumo-Forming of Hemispherical Parts Made Out of
Anisotropic Materials In Short-Term Creep Mode
S.N. Larin
1
, V.I. Platonov
1
, Nuzhdin G.A.
2
1 Tula State University, Tula, Russia
2 quality assurance company “Konsersium”, Moscow, Russia
DOI 10.13140/RG.2.1.3033.6408
Keywords: anisotropy, mathematical model, high-strength materials, domed parts, pneumo-forming, creep, pressure,
stress, thickness, damage rate, failure.
ABSTRACT. Provided here are results of theoretical and experimental research of strained and stressed state, force
modes, geometrical sizes for the blanks, and limit possibilities of deformation during isothermal blow molding of
hemispheric parts of anisotropic material in creeping mode .Determined is the effect for the researched parameters of the
studied deformation process, produced by anisotropy of mechanical properties, loading conditions and blank’s geometric
dimensions. Comparison of the theoretical and experimental data regarding the relative blank thickness in the blank dome
and base points, and of data regarding the relative height of the blank, point to their satisfactory agreement (up to 10
percent). Recommendations have been developed regarding calculation of scientifically-based technological parameters
for operations of isothermal straining of semi-spherical components made out of highly strong anisotropic materials in
the mode of short-time creeping. The recommendations were used during development of technological processes of
manufacture -- in the mode of short-time creeping and out of highly strong anisotropic materials --.of semispherical
components conforming to the operational technical requirements. The technological processes provide for increasing
specific strength by 1,5 1,7 times, for decreasing the mass by 1,5 times, for reducing labor content by 2-3 times, and for
growth of capacity factor from 0,3 to 0,9.
Introduction. Spherical sheet domes make bodies for tanks for fuel and liquid nitrogen, which are
used in aviation and space technologies. The conventional methods of their manufacture are in the
mode of multistage squeeze drawing with transitional thermal treatment rounds, or else it is hammer
stamping in subpress dies, these being highly labor-consuming operations. Isothermal deformation of
domed parts with gas of sheet high-strength aluminum and titanium alloys has considerable advantage
over the traditional methods of processing, and it is quite promising for its industrial use [1-5].
Materials that are subjected to processes of plastic straining, possess, as a rule,
anisotropy of mechanical properties. Anisotropy of mechanical properties may produce both
positive or negative effect upon stable procedure of technological processes of plastic
deformation under various thermomechanical modes [1-19].
Equations. Let us consider straining of a round sheet blank with radius
0
R
and with thickness
0
h
by
free bulging in the mode of viscous flow of material as influenced by excessive gas pressure
t
n
p
p
app
0
into a spherical matrix (Fig. 1). Here
pp
nap ,,
0
are pressurizing constants.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
35
Fig. 1. Diagram for calculation of strained condition of middle surface of blank in meridian plane
The blank is fixed along the outer contour. The blank material is taken as transversal-isotropic with
anisotropy ratio
R
; the stressed state of the dome is plane, i.e., the stress is perpendicular to the plane
of the sheet and equals zero (
0
z
). We consider straining in the meridian plane of the dome as
straining a membrane. Because of the symmetry of the material’s mechanical properties, relative to
the blank axis and the character of action of the external forces, -- the strain stresses and speeds, that
are meridian, circumstantial and normal to the middle surface of the blank, are the major ones.
The middle surface of the blank remains a part of the spherical surface at any stage of straining.
Taking place in any meridian section of the dome is radial flow of material relative to the new center
at every stage of straining.
According to the adopted allowance, the radiuses of curvature of the meridian section
m
of the
middle surface and of the slashing of the dome by the taper surface that is perpendicular to the
meridian arc,
t
, are found according to the following formula.
H
RH
tm
2
2
0
2
, (1)
where
H
is the height of the dome at a given moment of straining.
Since the trajectories of the points of the middle surface are orthogonal at this moment
to the profile in formation, straining speeds in the pole of the meridian surface (point “c” ) shall
be determined in the meridian sections as
2
0
2
2
RH
HH
c
tc
;
2
0
2
2
RH
HH
c
mc
;
h
h
c
zc
, (2)
where
dtdHH
;
dtdhh
.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
36
The blank is fixed in the contour (point k”), i.e., the strain speed along the contour equals zero
0
c
tk
, and according to the associated flow law, we have
0
c
tk
;
R
R
mk
tk
1
;
c
mk
c
zk
, (3)
where
R
is the normal anisotropy factor for viscous flow of material.
Further in the text we make no limitations about changing the shell thickness along the circumference
arc in the meridian section. In this case the speed for the strain in the meridian
c
m
, circumferential
direction
c
t
and the thickness strain
c
z
of the shell is found after the following formulas accordingly
ctg
c
m
sin
sin
;
ctg
c
t
sin
cos
;
h
h
c
z
. (4)
where
is here actual angle between the blank’s vertical axis of symmetry and radius-vector
determining the position of point in the section of middle surface by diagonal plane;
dtd
.
It was assumed during the straining of the shell that there takes place in the meridian plane during
every stage of straining a radial flow of point of middle surface, that is relative to the new center at
moment
dtt
, i.e., in the direction of
d
.
Connection between angle
and straining time
t
, when functional connection is assumed
tHH
,
is found as follows
0
2
R
tH
arctg
. (5)
The thickness of the meridian plane dome of the shell (
0
) is found from the following formula
2
2
0
2
0
1
R
tH
hh
. (6)
The changing thickness of the shell from straining time
t
at the spot of its fixation (
) is found
after the following formula
0
2
0
2
0
0
1
R
H
arctg
R
tH
R
tH
hh
. (7)
Cutting elements from the membrane by meridian planes and cone surfaces in the vicinity of the point
under consideration, and assuming that the stresses are evenly distributed throughout the thickness of
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
37
the element, we write down the equation of equilibrium of momentless shell loaded with uniform
pressure
p
, as follows:
h
p
t
t
m
m
;
h
p
t
mx
2
. (8)
Solving them together and taking account of
tm
, we find
h
p
tm
2
. (9)
The equivalent speed of straining
c
e
and equivalent stress
e
in the apex of the dome (point с”)
and in the point of fixation of the shell over the entire circuit (point k”) are computed respectively
for anisotropic material after the following formulas:
c
mc
c
ec
R 2
3
2
;
mc
c
ec
R
22
3
; (10)
mk
c
ek
R
RR
2
1
12
12
3
2
;
mkek
RR
R
2
1
12
12
2
3
. (11)
Let us consider a case of slow isothermal straining of a shell made out of material for which hold
constitutive equations of the energy theory of creep and creep damage [4]
m
c
A
n
ee
c
e
B
1
0
;
c
nр
c
ee
c
A
A
, (12)
where
B
,
n
,
m
are material invariables depending on the test temperature;
c
A
is the material damage rate for viscous deformation in the energy failure models;
с
пр
A
is specific fracture work for mode of viscous flow;
tdd
c
A
c
A
/
;
с
е
and
e
are equivalent straining speed and strain;
0
e
is equivalent stress dividing the viscous and viscoplastic flows of material.
The value of the specific fracture work
с
пр
A
for viscous flow of anisotropic material is found from
coscoscos
3210
bbbbDA
с
пр
,
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
38
where
3210
,,,, bbbbD
are material constants;
3/)(
321
is mean stress;
21
,
and
3
- are primary stresses;
,
,
are orientation angles of the first main stress axis
1
as relative to the main axes of
anisotropy
yx,
and
z
respectively.
Since the value of pressure
p
at every moment of straining is distributed evenly over
the surface of the shell, we shall be finding its value in the apex of the shell dome (point “с”).
By substituting into the first material state equation (12) its constituent values
e
and
c
e
, found after
formulas (10), and taking account of formulas (1), (4), (9), we receive
1
2
0
2
2
1
1
2
1
22
0
3
221
n
n
nn
n
n
m
c
Ac
n
e
n
RHB
dHhHR
dtp
. (13)
Shell thickness
h
is found after formula (6).
We now find the value of the damage accumulation
c
Ac
. By substituting into the second state
equation (12) formulas (10), account being taken of (1), (4) and (9), we obtain
H
Ah
R
H
p
c
np
c
Ac
0
2
2
0
2
1
. (14)
This equation is fitfully used if the pressurizing provides for
constp
.
If we substitute the first state equation into the second one, we shall have a different type of equation
for finding the damage rate
nc
np
n
n
c
ec
nm
c
Ac
e
c
cA
BA
1
1
0
1
. (15)
This equation can be comfortably used if
const
c
e
c
ec
1
. In the latter case integration of equation
(15) results in a formula of the following type:
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
39
mn
n
nc
np
e
n
n
c
e
c
cA
BA
t
n
mn
1
0
1
1
11
. (16)
Destruction time
t
is found from assumption
1
c
Ac
n
n
c
e
e
nc
np
mn
nBA
t
1
1
0
1
. (17)
Pressure
p
, which is necessary for creating the straining conditions, shall be computed after the
following formula
. (18)
Functional connection
t
c
A
c
A
is found as per formula (16), while
tHH
can be found from
the following equation
c
e
RH
RH
R
t
1
2
0
2
0
2
0
2
ln
3
22
. (19)
Limiting dome height
H
can be found by equation (19) with
tt
.
Function setting
tHH
allows us to find
t
c
A
c
A
from formulas (15) or (16), whereas
function
tpp
is calculated after formula (13).
The same procedure is used for finding the stressed and strained conditions of the blank in the point
of fixation of the shell (point “k”), and also found are the principаl equations and correlations for the
solving of the problem with the assumption that the behavior of the material is subject to equations
of the kinetic theory of creep and creep damage under the known law of time-pressure
)(tpp
and
under permanent equivalent strain rate in the blank dome
1e
.
There have been developed the calculation algorithm for the force and strain parameters of the studied
technological process, and the computer software.
Discussion of results. Evaluated were the stressed and strained conditions, the material
flow kinematics, the force modes and frontiers of the studied process of strain as relating to
accumulated microscopic damages, and depending on anisotropy of the mechanical properties
p t
R Hh
B H R
e
Ac
c
m n
n
e
c
n
0
2
1 2
1
0
2
0
2
1
1
1 2 2
3
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
40
of the source material, on the stress sequence, and on the geometric dimensions of the blank
and of the ready item.
Calculation was performed for titanium alloy LT31 under
CT
860
, the behavior of which is
subject to the energy theory of creep and creep damage, and for titanium alloy 4AL-3MO-1V under
CT
950
, the behavior of which is subject to the kinetic theory of creep and creep damage.
Mechanical characteristics of these materials during deformation under conditions of viscous flow of
material are shown in the work [3, 4].
The curves for changing values of gas pressure
p
, for relative values of blank thickness in the dome
0
hhh
сс
and at the place of its attachment
0
hhh
kk
, for the height of the dome-shaped blank
0
/ RHH
, for strain time
t
for titanium alloy LT31 (
CT
860
) under permanent value of the
equivalent strain rate in the blank dome
1e
--are shown in Fig. 2. Experimental data are shown here
with dots.
Fig. 2. Changes of
p
,
H
and
h
in studied points of blank as depending on
t
of titanium alloy LT31
(
300
0
R
;
c
e
1002,0
1
)
It follows from analysis of the calculation results and graphic dependencies that, as there grows to a
certain limit the strain time
t
, there takes place a steep increase in the relative height of the blank
H
and reduction in the relative thickness of the blank in the dome
с
h
and in the place of its fixation
k
h
.
Further increase in the time of straining
t
results in a smooth changing of the values under study. At
time moment
t
, close to the destruction of the blank, there takes place a sudden change of the relative
values of
H
,
с
h
and
k
h
. This is linked to the intensive accumulation of micro-damages at the
concluding stage of the process.
It has been found that change of the relative thickness in the blank dome proceeds in a more intensive
manner as versus changing the relative thickness at the place of its fixation. As the
t
straining time
grows, that difference grows too and may reach as much as 50%.
It was shown that for the purpose of ensuring a permanent equivalent speed of strain in the blank
dome, the law of changing pressure
p
during time of straining
t
has a complicated character. At the
initial moment of deformation we observe a sudden growth of pressure
p
, since there takes place a
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
41
substantial change of half-sphere radius
m
. Further extension of the straining time
t
is accompanied
with reduction in the magnitude of the gas pressure
p
.
Comparison of theoretical and experimental data for the relative thickness in the dome of blank
с
h
and at the place of its fixation
k
h
, as well as the data for relative height of blank
H
points to their
satisfactory coincidence (up to 10%).
It is defined, that destruction of the blank during isothermal staining takes place in blank dome where
there happens the maximum thinning of the blank.
Dependence of change in destruction time
*
t
, of relative height
H
and of thickness in the blank
dome
h
at the moment of destruction as determined after the magnitude of accumulated micro-
damages with , -- on the magnitude of the standing equivalent speed of strain in the blank
dome
1e
and on anisotropy ratio
R
is shown in Fig.3 and Fig.4 respectively. It is shown that
increasing the parameters of stress sequence
р
a
,
р
n
and increasing the magnitude of the permanent
equivalent speed of strain in the blank dome
1e
, result in decreasing destruction time
*
t
and blank
relative height
H
, and result also in increasing the relative thickness in the blank dome
h
.
It follows from the curves’ analysis (Fig. 4) that normal anisotropy factor
R
produces a substantial
effect upon the magnitude of the destruction time and upon the relative values
H
,
h
. As there
grows the anisotropy factor
R
, the relative value
h
grows steeply, while destruction time
*
t
and
blank relative height
H
, abruptly drop in magnitude. It was found, that failure to take account of
anisotropy of blank’s mechanical properties during analysis of the process of isothermal deformation
of a spherical shell results in a measurement error in the evaluation of destruction time
*
t
in the order
of 35%, and this results in an error of 15% in the evaluation of the relative height
H
and of the
thickness in the blank dome
h
. during the destruction moment.
Fig. 3. Dependence on
e
of changing
t
and
hH ,
in blank dome, titanium alloy LT31 (
300
0
R
)
1
c
A
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
42
Fig. 4. Dependence on R of changing
t
and
hH ,
(kinetic theory;
300
0
R
;
;013,0
0
МПаp
;104
3
p
n
p
сМПаa
6,0
p
n
)
Calculation analysis shows that the limitations of deformation under conditions of viscous flow of
material whose behavior obeys the kinetic theory of creep and creep damage (alloy 4AL-3MO-1V),
do not depend on the loading conditions of the blank. Shown is the substantive dependence of
destruction time
*
t
on loading parameters
р
a
,
р
n
and on the magnitude of the permanent equivalent
strain rate
1e
. Increasing the loading parameter
р
a
from 0,2
p
n
сМПа
3
10
to
1,4
p
n
сМПа
3
10
, and increasing
р
n
from 0,5 to 0,8 under fixed other parameters, results in the
reduction of the destruction time
*
t
by 1,8 times.
There has been shown substantive effect of the geometric dimensions of the blank upon the values of
the destruction time
*
t
. It was found that increasing the relative size of the blank radius
000
/ hRR
from 200 to 800 results in reduction of the destruction time by more than 4 times over.
Summary. There has been carried out experimental research of operations of isothermal straining of
hemispherical parts of high-strength anisotropic materials in the mode of short-time creeping. It was
experimentally shown that in case of direct one-stage molding, there forms significant unevenness in
the width of the wall along the generatrix from the flange radius to the dome center (
0
hh
). In case
of shells of alloys LT31 it was on an average 0,4; and was 0,33 for shells made out of alloy AA5052.
In order to lessen the unevenness in the wall thickness, it is recommended to use two-stage stamping
with an insert (reversal). Comparison of theoretical and experimental data on relative thickness in the
blank dome and in the basic points, as well as comparison of the blank’s relative heights, points to
their satisfactory coincidence (up to 10 %).
Recommendations have been developed for calculating scientifically-based technological parameters
for operations of isothermal straining of semispherical components made out of highly strong
anisotropic materials in the mode of short-time creeping. The technological processes are based on
observing the proper sequence of operations with the provided blanks in one operating position:
heating, vacuuming forming thermal stabilization cooling.
The above recommendations were used for developing technological processes of manufacturing
semispherical components meeting the operational technological requirements, the components were
made out of highly strong anisotropic materials in the mode of short-time creeping. Samples of dome-
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
43
shaped critical-task components made out of titanium alloys 4AL-3MO-1V, manufactured by
isothermal stamping in the mode of short-time creeping, are shown in Fig. 5.
a)
b)
Fig. 5. Samples of dome-shaped components made out of titanium alloys 4AL-3MO-1V (а) and ВТ23
(b)
The described technological processes provide for: greater specific strength 1,5 ... 1,7 times; weight
reduction - 1,5 times; lower labor consumption - 2...3 times; greater capacity factor (from/to) - 0,3/0,9.
This research was carried out as per basic part of state task №2014/227 for scientific research, issued
by the Ministry of Education and Science of the Russian Federation for 2014-2020, and as per grant
RFFI № 14-08-00066 а, № 16-48-710016 and 16-08-00020 .
References
[1] K.I .Romanov, Mechanics of hot metal forming, M .: Engineering, 1993. 240 pp.
[2] N.N. Malinin, Applied theory of plasticity and creep, M .: Mechanical engineering, 1975. 400 p.
[3] S.P. Yakovlev, V.N .Chudin, S.S .Yakovlev, J.A. Sobolev, Isothermal deformation of high
anisotropic materials, M .: Engineering, 2004. 427 pp.
[4] S.S. Yakovlev, V.N. Chudin, J.A. Sobolev, S.P. Yakovlev, V.I. Tregubov, S.N. Larin, Isothermal
pneumo-forming of anisotropic high-strength sheet materials, M .: Engineering, 2009. 352 pp.
[5] E.I. Semenov, Forging and Stamping: Reference: In 4 volumes. V.4. Stamping, M .: Engineering,
2010. 732 pp.
[6] Y.M.Aryshensky, F.V. Grechnikov, Theory and calculations of plastic deformation of anisotropic
materials, M .: Metallurgy, 1990. 304 pp.
[7] F.V.Grechnikov, Straining of anisotropic materials, M .: Engineering, 1998. 446 pp.
[8] V.D. Golovlev, Calculation of stamping processes, M .: Engineering, 1974. 136 pp.
[9] S.P.Yakovlev, V.D. Kuhar, Punching anisotropic blanks, M .: Engineering, 1986. 136 pp.
[10] S,P.Yakovlev, S.S. Yakovlev , V.A. Andreichenko, Forming anisotropic materials, Kishinev:
Quant, 1997. 332 p.10.
[11] S.S.Yakovlev, O.V. Pylypenko, Isothermal extract anisotropic materials, M .: Engineering,
2007. 212 pp.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
44
[12] S.S.Yakovlev, V.I.Tregubov, Theory and technology of manufacturing large-sized axisymmetric
critical parts of highly anisotropic materials, Tula: Tula State University Publishing House, 2011. 230
pp.
[13] S.S.Yakovlev, V.D.Kuhar, V.I.Tregubov, Theory and technology of forming anisotropic
materials, M .: Engineering, 2012. 400 pp.
[14] S.S.Yakovlev, V.I.Tregubov , S.P.Yakovlev, Rotating hood with wall thinning rotationally
symmetric parts of anisotropic round billets - specialized equipment, M .: Engineering, 2009. 265 pp.
[15] S.S.Yakovlev, V.I.Tregubov, Theory and technology of manufacturing large-sized axisymmetric
critical parts of highly anisotropic materials, Tula: Tula State University Publishing House, 2011. 230
pp.
[16] S.S.Yakovlev, V.D.Kuhar, V.I.Tregubov, Theory and technology of forming anisotropic
materials, M .: Engineering, 2012. 400 pp.
[17] S.S. Yakovlev, V.I.Tregubov, V.D .Kuhar, V.Y.Travin, Deep drawing of anisotropic materials,
Tula: Tula State University Publishing House, 2013. 225 pp.
[18] S.S.Yakovlev, V.I.Tregubov, G.A.Nuzhdin, Plastic flow of anisotropic hardening material with
strength differential effect, Tula: Tula State University Publishing House, 2014. 115 pp.
S.S. Yakovlev, K.S. Belts, V.I.Tregubov, Stability of anisotropic sheet metal and of pipe blanks
during plastic deformation, Tula: Tula State University Publishing House, RARAN, 2014. 222 pp.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
45
MHD Stagnation Point Flow in a Boundary Layer Of a Nano Fluid Over a
Stretching Sheet in the Presence of Viscous Dissipation and Chemical Reaction
Ch. Achi Reddy
1,a
& B. Shankar
2
1 M.L.R. Institute of Technology, Dundigal, Hyderabad, 500 043, India
2 Professor, Department of Mathematics, Osmania University, Hyderabad, 500 007, India
a achireddy.ch@gmail.com
DOI 10.13140/RG.2.1.2022.4881
Keywords: boundary layer flow, exponentially stretching sheet, chemical reaction and viscous dissipation.
ABSTRACT. The paper shows an attempt of numerical investigation on the effect of viscous dissipation and Chemical
reaction on a viscous, steady and incompressible fluid over an exponentially stretching sheet within a specified boundary
layer. As a formal approach, the model has been adopted with the governing equations and the simulation is carried out
with the Keller Box method. The pattern or the profiles of the skin friction coefficient and the heat and mass transfer rates
are achieved in execution of mathematical model have been presented in the paper. The enhancement in magnetic
parameter leads to a considerable reduction in velocity and Chemical reaction parameter is predominant in controlling
the profile of concentration. An increase in Eckert number is observed to cause the enhancement in the temperature profile
whereas it decreases the concentration profile. The results obtained in the simulation of Keller box method are in well
agreement with realistic situation of the scientific scenario.
1. Introduction. The importance of stagnation point flow has drawn the attention of many researchers
due to its growing application in industry. The fluid is said to have reached its stagnation point when
local Velocity of the fluid becomes zero. In some situations, flow is stagnated by a solid wall while
in others; there is a line interior to a homogeneous fluid domain or the interface between two
immiscible fluids [1-3]. A good amount of research is done drawing the attention of several
researchers [4-12].
In 1993, during an investigation of new coolants and cooling technologies at Argonne National
Laboratory in U.S. Chai invented a new type of fluid called Nano fluid [13]. Nano fluids are fluids
that contain small volumetric quantities of nanometre sized particle, Called nanoparticles. The
nanoparticles used in Nano fluids are typically made of metals, oxides, carbides, or carbon nanotubes.
Common base fluids include water, ethylene glycol and oil. Nano fluids commonly contain up to a
5% volume fraction of nanoparticles to see effective heat transfer enhancements [25-27].
Nano fluids are studied because of their heat transfer properties: they enhance the thermal
conductivity and convective properties over the properties of the base fluid. Typical thermal
conductivity enhancements are in the range of 15-40% over the base fluid and heat transfer coefficient
enhancements have been found up to 40%.Increasing the thermal conductivity of this magnitude
cannot be solely attributed to the higher thermal conductivity of the added nanoparticles, and there
must be other mechanisms attributed to the increase in performance [28-29].
Stagnation point flow appears in virtually all fields of science and engineering. A flow can be
stagnated by a solid wall or a free stagnation point or a line can exist in the interior of the fluid
Domain. The study of stagnation point flow as pioneered by Hiemenzin 1911 [15] who solved the
two dimensional stagnation point problem using a similarity transformation.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
46
Magyari and Keller [16] investigated the steady boundary layers flow on a stretching continuous
surface with exponential temperature distribution while Partha et al. [17] analysed the effects of
viscous dissipation on the mixed convection heat transfer from an exponentially stretching surface.
In the present study we have investigated the viscous dissipation for different values of velocity ratio
parameter and observed that the nanoparticle volume decreases with the increase of chemical reaction
parameter for
= 0.1 and 2.1.
2. Mathematical Formulation. Consider a steady, two-dimensional boundary layer stagnation-point
flow of a Nano fluid over an exponentially stretching sheet, the stretching and free stream velocities
are assumed to be of the forms
x/l
w
aexu )(
, and
x/l
bexu
)(
, respectively. Where a > 0 and b are
constants, x is the coordinate measure along the stretching surface and
l
is length of the sheet. A non-
uniform transverse magnetic field of strength
lx/
eBxB
2
0
)(
is imposed parallel to the y-axis, where
Bo is a uniform magnetic field strength. It is assumed that the induced magnetic field due to the
motion of an electrically conducting field is negligible. Further, it is also assumed that the external
electric field is zero and the electrical field due to polarization of charges is negligible [22].
Figure (1) shows that the temperature T and the Nano-particles fraction C take forms Tω(x) and
Cω(x), respectively whereas the ambient values of temperature T∞ and Nano-particle fraction C∞
are obtained when y tends to infinity.
Fig. 1. Physical flow model and coordinate system.
The governing boundary layer equations of the conservation Law of mass, momentum, energy and
concentration in the flow as follows:
0
y
v
x
u
(1)
,
)(
2
2
2
uu
xB
y
uu
dx
du
u
y
u
v
x
u
u
ff
(2)
2
2
2
2
)(
)(
1
y
u
c
y
T
T
D
y
T
y
C
D
y
q
c
y
T
y
T
v
x
T
u
fp
T
B
r
f
(3)
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
47
CC
y
T
T
D
y
C
D
y
C
v
x
C
u
T
B
2
2
2
2
(4)
Here u and v are the velocity components in the x and y directions respectively,
µ is the viscosity;
ρ is the density of the base field;
σ is the electrical conductivity.
f
c
,
where k is the thermal conductivity;
f
c
is the heat capacitance of the base fluid.
f
p
c
c
,
where
p
c
is the heat capacitance of the nanoparticle;
D
B
is the Brownian diusion coecient;
D
T
is the thermophoresis diusion coecient;
r
q
is the radiation flux.
The Rosseland approximation is defined as [23, 24];
y
T
k
q
r
4
*
*
3
4
(5)
where σ
is the Stefan-Boltzmann constant;
k
is the mean absorption coecient.
It is assumed that the temperature dierence between the free stream T
and local temperature T is
small enough expanding T
4
in a Taylor series about T
and neglecting higher order terms results.
434
34
TTTT
(6)
After substituting Eqs. (5) and (6) in Eq. (3), it will be reduces to
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
48
22
2
2
3
)(
)(*3
*16
y
u
cy
T
T
D
y
T
y
C
D
y
T
c
T
y
T
v
x
T
u
fp
T
B
f
(7)
The subjected boundary conditions are
yasCCTTvbexuu
yatxCCxTTvaexuu
x/l
x/l
,,,0,)(
0),(),(,0,)(
(8)
The prescribed temperature and concentration on the surface of the sheet is assumed to be
T
ω
(x) = T
+ T
0
e
x/2l
and C
ω
(x) = C
+ C
0
e
x/2l
where T
0
, C
0
are the reference temperature and
concentration respectively, now, the non-linear partial dierential equations for the purpose of a
stream function ψ = ψ(x,y) is defined as
x
v
y
u
,
, (9)
where the continuity Eq. (1) is satisfied identically. A similarity transformation is defined as [20]
follows
lx
lx
elay
CC
CC
TT
TT
feal
2/
2/
2/,)(
)(),(2
(10)
As such Eq. (10), Eqs. (2), (4) and (7) reduce to the following system of nonlinear ordinary dierential
equations.
0)(22
22
fMffff
(11)
0
22
fENtNbffp
C
N
r
(12)
0
LeNtfLeLef
b

(13)
where
ff
ap
Bl
M
p
vab
2
0
2
,,/
(14)
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
49
)(
CCD
B
Nb
,
)()(
,
2
)(
TTCp
u
ENt
f
C
vT
TTD
T
,
u
lk2
,
where, prime denote the dierentiation with respect to
B,
is the velocity ratio parameter;
v is the kinematic viscosity of the fluid;
P
r
is the Prandtl number;
Le is the Lewis number;
M is the magnetic parameter;
A is the heat source parameter,
S is the suction parameter.
N
P
P
r
rN
3
4
1
1
,
where
*
*4
3
kk
T
N
is the radiation parameter,
Nb
Nt
Nt
b
,
where Nb is the Brownian motion parameter and Nt is the thermophoresis parameter;
Ec Eckert number;
γ is chemical reaction parameter;
The corresponding boundary conditions Eq. (8) are transformed into
asf
atff
0)(,0)(,)(
,01)(,1)(,1)(,0)(
(15)
The parameters of practical interest in the formulated problem are velocity, heat and mass transfer
respectively, which are presented in terms of Skin friction C
f
, Nusselt number Nu and Sherwood
numbers Sh. Using the transformed variables (10), the non-dimensional expressions for the Skin
friction coecient
)0()0( fC
fx
, the reduced Nusselt number
)0(
and the reduced Sherwood
number
)0(
respectively are defined as;
Skin friction Coecient. The Skin friction coecient
f
C
is defined by
2
2
1
pU
C
f
,
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
50
where τ
ω
is the local wall Shear stress;
ρ is the fluid density.
)0(Re fC
fx
Nusselt Number. In heat transfer at a boundary within a fluid, the Nusselt number (Nu) is the ratio
of convection to conductive heat transfer across the boundary
x
x
ω
ω
x
Nu
)(θ
) -TK(T
xq
sfer heat tranconductive
sfer heat tranconvective
Nu
Re
0
(16)
Sherwood Number. The Sherwood number (Sh) is also called the mass transfer Nusselt number. It
represents the ratio of convective to diusive mass transport
)(φ
Sh
) -C(CD
xJ
cientfer coeffimass TransDiffusive
icientsfer coeff mass TranConvective
Sh
x
x
ωB
ω
x
0
Re
,
where
v
x
(x)u
ωx
Re
is the local Reynolds number based on the stretching velocity.
3. Numerical Procedure. The equations (11) - (14) subject to the boundary conditions (15) are solved
numerically using an implicit finite - dierence scheme known as Keller box method. The method
has the following four basic steps.
1. Reduce equations (11) - (14) to first order equations.
2. Write the dierence equations using central dierences.
3. Linearize the resulting algebraic equations by Newton’s method and write them in matrix -vector.
4. Use the Block - tridiagonal elimination technique to solve the linear system.
4. Results and Discussion.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
51
Table 1. Comparison of the reduced Nusselt number
0)0(
LeNtNbwhen
P
r
M
N
[19]
)0(
[20]
)0(
[21]
)0(
Present results
)0(
1
0
0
0.9548
0.9548
0.9548
0.9548
2
0
0
1.4714
1.4714
1.4714
1.4715
3
0
0
1.8691
1.8691
1.8691
1.8692
1
0
1.0
0.5315
0.5312
0.5312
0.5311
1
1.0
0
---
0.8611
0.8611
0.8611
1
1.0
1.0
---
0.4505
0.4505
0.4505
Table 2. Values of the reduced Nusselt number
)0(
, reduced Sherwood number
)0(
, and the
skin friction coefficient
)0(
fx
C
for various values of Eckert number.
Nb = Nt = M =
=
= 0.1
)0(
)0(
)0(
fx
C
E
c
Le
N
0.1
10
1.0
0.5176
3.6996
1.2856
0.2
10
1.0
0.4976
3.7107
1.2856
0.3
10
1.0
0.4776
3.7219
1.2856
0.4
10
1.0
0.4375
3.7442
1.2856
Table 3. Values of the reduced Nusselt number
)0(
, reduced Sherwood number
)0(
, and the
skin friction coefficient
)0(
fx
C
for various values of Chemical Reaction parameter.
Nb = Nt = M =
= E
C
= 0.1
)0(
)0(
)0(
fx
C
Pr
Le
N
0.1
1.00
10
1.0
0.5176
3.6996
1.2856
0.2
1.00
10
1.0
0.5175
3.8407
1.2856
0.3
1.00
10
1.0
0.5174
3.9757
1.2856
0.4
1.00
10
1.0
0.5172
4.1055
1.2856
The system of ordinary differential equations [11-13] has been solved numerically using Keller-box
method. From the numerical computation, the main physical quantities of interest namely the local
Skin friction coefficient, the local Nusselt number and the local Sherwood number are obtained and
the results are presented in Table 2 and Table 3.
From Table 2 it is observed that with the increase in Eckert number, there is a decrease in rate of heat
transfer and increase in mass transfer.
From Table 3 it is observed that with increase in chemical reaction parameter, there is no significant
change in rate of heat transfer but there is an increase in the rate of mass transfer.
Figure 2 shows the effects of the magnetic Parameter M on the flow field velocity
)(
f
for three
different values of the Velocity ratio parameter
,
=0.1, 1 and 2.1.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
52
Fig.2. Velocity profile against
for different values of M
When
=0.1 the velocity profile and the boundary layer thickness decrease with an increase in M.
When
=2.1, higher value of the Lorentz force further reduces the velocity and consequently the
thickness of boundary layer reduces. When
=1, there is no influence of magnetic field
)(
f
and
)(
f
attains a constant value of 1 for any value of
indicating that there is no boundary layer of
fluid, as shown by a dotted line in Fig 2. This means that in the case when the external stream velocity
becomes equal to the stretching velocity. The flow is not influenced by the different values of the
incorporated flow parameters. When
=2.1 the flow velocity increases indicating the decrease in
thickness of boundary layer with an increase in M. As compared to
=0.1case, the boundary layer
thickness decreases causing an inverted boundary structure.
Fig. 3. Temperature profile against
for different values of E
c
Figure 3 shows an increase in Nano fluid Temperature with increases in the viscous dissipation
parameter, Eckert number. Which can be attributed to the action of viscous heating. Concentration
increases with an increase in the viscous dissipation parameter, Eckert number as shown in Fig. 4
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
53
Fig. 4. Concentration profile against
for different values of E
c
Fig. 5. Concentration profile against
for different values of
The influence of chemical reaction parameter on concentration profile is shown in Fig. 5.
Concentration decreases with an increase in the chemical reaction parameter indicating that the
nanoparticle volume fraction decreases with the increase of chemical reaction parameter, while effect
chemical reaction parameter is not significant on the temperature profile.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
54
Fig. 6. Temperature profile profile against
for different values of N
Summary. A numerical study corresponding to the flow and heat transfer in a steady flow region of
Nano fluid over an exponential stretching surface and effects of Chemical reaction parameter and
Eckert number are examined and discussed in detail. The main observations of the present study are
as follows.
(I) An increase in the magnetic parameter is to reduce the velocity profile;
(II) An increase in the Eckert number increases the temperature profile while it reduces the
concentration profile;
(III)With increasing values of Chemical reaction parameter (
) the concentration profile decreases.
References
[1] K. Hiemenz. Die Grenzschicht an einem inden gleichf¨ormigen Flu¨ssigkeitsstrom eingetauchten
geraden Kreiszylinder. Dingler’s Polytech. J. 1911; 326, 321-4.
[2] T.R. Mahapatra and A.S. Gupta. Heat transfer in stagnation-point flow towards a stretching sheet.
Heat Mass Tran. 2002; 38, 517-21.
[3] T.R. Mahapatra and A.S. Gupta. Stagnation-point flow towards a stretching surface. Can. J. Chem.
Eng. 2003; 81, 258-63.
[4] R. Nazar, N. Amin, D. Filip and I. Pop. Stagnation point flow of a micropolar fluid towards a
stretching sheet. Int. J. Nonlinear Mech. 2004; 39, 1227-35.
[5] Y.Y. Lok, N. Amin and I. Pop. Non-orthogonal stagnation point flow towards a stretching sheet.
Int. J. Nonlinear Mech. 2006; 41, 622-7.
[6] S. Nadeem, A. Hussain and M. Khan. HAM solutions for boundary layer flow in the region of the
stagnation point towards a stretching sheet. Comm. Nonlinear Sci. Numer. Simulat. 2010; 15, 47581.
[7] F. Labropulu, D. Li and I. Pop. Non-orthogonal stagnation-point flow towards a stretching surface
in a non-Newtonian fluid with heat transfer. Int. J. Therm. Sci. 2010; 49, 1042-50.
[8] A. Ishak, Y.Y. Lok and I. Pop. Stagnation-point flow over a shrinking sheet in a micropolar fluid.
Chem. Eng. Comm. 2010; 197, 1417-27.
[9] N. Bachok, A. Ishak and I. Pop. Boundary-layer flow of nanofluid over a moving surface in a
flowing fluid. Int. J. Therm. Sci. 2010; 49, 1663-8.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
55
[10] M. Patel and M.G. Timol. Orthogonal stagnation point flow of a power law fluid towards a
stretching surface. Int. J. Appl. Math. Mech. 2011; 7, 31-7.
[11] N. Bachok, A. Ishak and I. Pop. On the stagnation point flow towards a stretching sheet with
homogeneous-heterogeneous reactions eects. Comm. Nonlinear Sci. Numer. Simulat. 2011; 16,
4296-302.
[12] S.U.S. Choi, Z-G. Zhang, W. Yu, F.E. Lockwood and E.A. Grullce, Appl. Phys. Lett. 79, 2252
(2001).
[13] E. Magyari and B. Keller, “Heat and mass transfer in the boundary layers on an exponentially
stretching continuous surface”, Journal of Physics D, Vol. 42, no. 5, pp. 577-585, 1999.
[14] M.K. Partha, P.V.S.N. Murthy, and G.P. Rajasekhar, “Eect of viscous dissipation on the mixed
convection heat transfer from an exponentially stretching surface”, Heat and Mass Transfer, Vol., 11,
no. 4, pp. 360-366, 2005.
[15] M. Shakhaoath Khan, I. Karim, and M. Sirajual Islam, “Possessions of chemical reaction on
MHD heat and mass transfor nanofluid flow on a continuously moving surface,” The American
Chemical Science Journal, Vol. 4, no. 3, pp. 401-415, 2014.
[16] B. Bidin and R. Nazar. Numerical solution of the boundary layer flow over an exponentially
stretching sheet with thermal radiation. Eur. J. Sci. Res. 2009; 33, 710-7.
[17] A. Ishak. MHD boundary layer flow due to an exponentially stretching sheet with radiation
eect. Sains Malays. 2011; 40, 391-5.
[18] C.H. Chen. Eects of magnetic field and suction or injection on convection heat transfer of
nonNewtonian power law fluids past a power law stretched sheet with surface seat slux. Int. J. Therm.
Sci. 2008; 47, 954-61.
[19] R.D. Cess. The interaction of thermal radiation with free convection heat transfer. Int. J. Heat
Mass Tran. 1966; 9, 1269-77.
[20] F.M. Hady, F.S. Ibrahim, S.M. Abdel-Gaied and M.R. Eid. Radiation eect on viscous flow of
a nanofluid and heat transfer over a nonlinearly stretching sheet. Nanoscale Res. Lett. 2012; 7, Article
ID 229.
[21] E.I. Aziz, M.A. 2009, “Viscous dissipation eect on mixed convection flow of a micropolar fluid
over an exponentially stretching sheet”, Con. J. Phys. 87: 359-368.
[22] Partha, M.K. Murthy, P.V.S.N. and Rajasekhar, G.P., 2005: “Eects of viscous dissipation on
the mixed convection heat transfer from an exponentially stretching surface”. Heat Mass Transfer 41:
369-366.
[23] M.S. Khan, I. Karim, L.E. Ali, and A. Ismail, “Unsteady MHD free convection boundary-layer
flow of a nanofluid along a stretching sheet with thermal radiation and viscous dissipation eect,”
International Nano Letters, vol. 2, article 24, 2012.
[24] M.A. Hamad, I. Pop, and A.I. Ismail, “Magnetic field eects on free convection flow of a
nanofluid past a vertical semi-infinite flat plate, ” Nonlinear Analysis. Real World Applications, Vol.
12, no. 3, pp. 1338-1346, 2011.
[25] K. Jafar, R. Nazar, A. Ishak, and I. Pop, “MHD flow and heat transfer over stretching/shrinking
sheets with external magnetic field, viscous dissipation and Joule eect,” The Canadian Journal of
Chemical Engineering, vol. 90, no. 5, pp. 1336-1346.
[26] M. Gnaneswara Reddy, “Influence of the thermal radiation, viscous dissipation and hall current
on MHD convection flow over a stretched vertical flat plate,” Ain Shams Engineering Journal, Vol.
4, no. 1, pp. 169-175, 2011.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
56
[27] K. Vajravelu and A. Hadjinicolaou, “Heat transfer in a viscous fluid over a stretching sheet with
viscous dissipation and internal heat generation”, International Communications in Heat and Mass
Transfer, vol. 20, no. 3, pp. 417-430, 1993.
[28] F.H. Hady, E.S. Ibrahim, S.M. Abdel-Gaied, and M.R. Eid, “Radiation eect on viscous flow of
a nanofluid and heat transfer over a nonlinearly stretching sheet, “Nanoscale Research Letters, vo. 7,
article 229, 2012.
A. Postelnicu, “Influence of chemical reaction on heat and mass transfer by natural convection from
vertical surfaces in porous media considering Soret and Dufour eeects,” Heat and Mass Transfer,
vol. 43, no.6, pp. 595-602, 2007.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
57
Quality Characteristics of Cutting Surfaces in the Milling of the Titanium Alloy
Ti10V2Fe3Al
Michael Storchak
1
, Lucas Saxarra
1
, Like Jiang
2
, Yiping Xu
2
, Xun Li
3
1 Institute for Machine Tools, University of Stuttgart, Holzgartenstraße 17, 70174 Stuttgart, Germany
2 Changhe Aircraft Industries Group Limited Corporation, Jingdezhen, 333002, China
3 School of Mechanical Engineering and Automation, Beihang University, Beijing 100191, China
DOI 10.13140/RG.2.1.4655.1925
Keywords: Machining of titanium alloy, surface roughness, microhardness, residual stress, simulation.
ABSTRACT. As titanium alloys have unique mechanical propertie -titanium alloy Ti10V2Fe3Al (Ti-1023) is
widely used by the aerospace industry, among other things, when producing critical components such as parts of the
fuselage and the wings as well as various rotating components due to its extremely high ratio of strength to density, its
great resistance to fatigue, its excellent resistance to corrosion and fracture toughness. Within the group of titanium
-phase are among the materials which are most difficult to machine. In particular, this
concerns milling processes widely used in the production of various complicated components. In order to be able to
successfully apply the machining process of the titanium alloy Ti-1023, optimum cutting parameters of the tool have to
be used, guaranteeing a required machining quality. This paper presents the results of experimental tests into the formation
of quality characteristics such as roughness and microhardness as well as residual stresses and their simulation depending
on cutting parameters such as cutting speed, feed and radial depth of cut. To analyse more closely how the cutting
parameters affect the quality characteristics in milling, the individual dependences of the effects were described in
exponential equations. The exponents for the exponential equations were established according to the Gaussian
elimination method. The experimental results obtained and the developed FEM cutting models serve as a basis for further
optimising the machining processes of titanium alloys.
Introduction. -titanium alloy
Ti10V2Fe3Al (Ti-1023) is widely used by the aerospace industry, among other things, when
producing critical components such as parts of the fuselage and the wings as well as various rotating
components, due to its extremely high ratio of strength to density, its great resistance to fatigue, its
excellent resistance to corrosion and fracture toughness [1]. Thanks to its unique physical-mechanical
properties, this titanium alloy is gaining more and more importance besides the α+β-alloy Ti6AlV4,
which is commonly used in industry and commerce [2]. Within the group of titanium materials, the
alloys of the β-phase are among the materials which are most difficult to machine. Especially material
removal is characterised by low cutting speeds, small feeds and short tool lives, depending on the
material precipitation condition [3].
These difficulties in the machining of the titanium alloy Ti-1023 have a great effect on guaranteeing
the physical-mechanical [4], [5] and the geometrical quality characteristics [6] of the machined
components' surface layers. This concerns particularly milling processes commonly used when
producing various complicated components [7]. The cutting parameters have the greatest effect on
the formation of the quality characteristics in the machining of titanium alloys as well as other
materials. This paper presents the results of experimental tests into the formation of quality
characteristics such as roughness, microhardness and residual stresses as well their simulation
depending on cutting parameters such as cutting speed, feed and radial depth of cut of the milling
cutter.
Test set-up. Milling process. The experimental analyses of the milling process were conducted on
the machining centre Hermle UWF 1202 H. The test set-up for milling is shown in Fig. 1. The
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
58
workpiece made out of the titanium alloy Ti-1023 has a dimension of 20 mm 20 mm 60 mm and
is clamped in a six-component dynamometer, type 9129AA, by Kistler by means of a clamping
system. It was milled on a rectangular track along the cross-section of the workpiece in four steps
with varying real depth of cut a
e
of 0,5 mm, 1,0 mm, 1,5 mm and 2,0 mm (see Fig. 1). A cemented
carbide end milling cutter with five teeth, D20R3H30Q90L130 by Widia, was used as tool. The three
components of the resultant force F
x
, F
y
and F
z
as well as the milling moment M
z
around the rotational
axis of the milling cutter were measured during machining.
Fig. 1. Test set-up for analysing milling processes
Surface roughness. The roughness of machined surfaces was measured with a stylus profiler SV-
C3200 by Mitutoyo. The measurements were carried out according to the scheme presented in Fig.
2. For a statistical safeguarding, each milled surface was separately scanned several times lengthwise
and crosswise. Then the average of the obtained results was taken. The following parameters were
established in the roughness measurements [8]:
centre line average height R
a
;
average roughness height R
z
;
average depth of surface smoothness R
p
;
height root mean square R
q
;
roughness height R
t
;
average roughness width RS
m
;
material ratio Rmr, from which the material ratio curve or rather the Abbott curve was made
up to characterise the vertical material distribution of a surface.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
59
Fig. 2. Scheme of roughness measurement
Microhardness. The microhardness was established with the measuring system Picodentor HM500
by Fischer according to DIN EN ISO 14577 - Fig. 3.
a) b)
Fig. 3. Set-up (a) and scheme (b) for measuring microhardness
A standardised indenter here is pressed into the microstructure of the test specimen with a
predetermined test force [9]. During the entire process, the applied force is recorded with a load cell
and the penetration depth arising here is detected with a position-measuring system for each
individual test method - Fig. 4.
For guaranteeing the reproducible measuring results, it is necessary at the beginning to define the
optimum number of measuring points for each milled surface to be tested. After evaluating the
variation coefficient, it showed that an optimum result is achieved with ten measuring points. The
maximum test force was 500 mN, the load well time was 20 s and the delay under load was 5 s in the
measurements. Not only the Vickers hardness HV, but also the elastic penetration work W
e
, the plastic
penetration work W
p
and the total penetration work W
t
were established in the tests.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
60
Fig. 4. Curves of indenting load when measuring microhardness
Residual stress. The residual stresses were analysed at given positions of the specimens, which were
machined at varying cutting parameters and with or without supplying coolant. The cutting method
was used for this. Strain gauges of the FLA-3-11 type were centrally mounted onto the machined
surfaces, before the specimens were sawed up. The machined surfaces were cut at a distance of 1 mm
to the edge of the sample. Afterwards, the strains caused by cutting were determined with a measuring
bridge, type P3500 by Micro Measurements Group. The residual stress caused was calculated from
the strain difference to the initial condition using a Young’s modulus of E = 106 GPa and a Poisson’s
ratio of μ = 0,36.
Results of milling process analyses. The cutting parameters of cutting speed V
C
, feed f
z
and radial
depth of cut a
e
were varied in the experimental tests regarding the milling process of the titanium
alloy Ti-1023. The axial depth of cut was 7 mm in all analyses of the milling process.
The tests were carried out using factorial design of experiments. To be able to detect a definite
behaviour of quality characteristics such as the roughness and the microhardness of surface layers in
the machining of the titanium alloy Ti-1023 depending on the analysed cutting parameters, such
dependences are represented in the form of an exponential equation:
x y z
Ce
C V f a Q
, (1)
where Q is the corresponding quality characteristic;
C is the constant;
x, y and z are the exponents for the cutting parameters.
Based on the experimental tests of the quality characteristics, the exponents for the exponential
equations were established according to the Gaussian elimination method [10]. The analysis of the
exponents, presented below, guarantees that a clear effect of the respective cutting parameters on the
quality characteristics to be examined can be established.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
61
Kinetic characteristics. The three parameters of cutting speed, feed and radial depth of cut were
varied in the experimental analyses regarding the milling process of the titanium alloy Ti-1023.
Cutting speed was changed in five steps of 30 m/min, 45 m/min, 60 m/min, 90 m/min and 120 m/min.
Feed was varied in four steps of 0,06 mm/tooth, 0,08 mm/tooth, 0,1 mm/tooth and 0,12 mm/tooth,
and the radial depth of cut was changed in four steps of 0,5 mm, 1,0 mm, 1,5 mm and 2,0 mm.
As an example, Fig. 5 shows how the radial force F
x
(see Fig. 5, a) and the axial force F
z
(see Fig.
5, b) depend on the radial depth of cut of the milling cutter and the cutting speed at a feed f
z
of
0,12 mm/tooth. Both the radial force F
x
and the axial force F
z
increase with growing radial depth of
cut. The increase here is proportional to the growing material removal. As cutting speed increases, it
can partly be seen that the components of the resultant force rise. This increase does, however, not
apply to all analysed values of the cutting parameters and does not maintain its character in the entire
examined range.
a) b)
Fig. 5. Dependence of radial F
x
(a) and axial F
z
(b) forces on the radial depth of cut and the cutting
speed
Hence, on the basis of the performed tests, it cannot be clearly found that changing the cutting
speed affects the kinetic machining characteristics in milling.
Surface roughness. As examples, the presented results showed how roughness, namely the height
parameters R
a
and R
z
as well as the step parameter RS
m
of the surfaces machined crosswise, depends
on the cutting parameters.
Based on all measurements, the exponential equation (2) was made up, enabling an analysis of how
the cutting parameters affect the centre line average height R
a
of the machined surfaces:
0,045 1,641 0,038
35,8
a C Z e
R V f a
[µm]. (2)
According to Equation (2), feed f
Z
has the greatest effect on the centre line average height R
a
. As feed
increases, R
a
rises here by an exponent of 1,641. The cutting speed and the radial depth of cut of the
milling cutter have a considerably smaller effect. Here R
a
decreases slightly with increasing cutting
speed and rises slightly with growing radial depth of cut. Fig. 6 illustrates this regularity with a few
exceptions by showing how the centre line average height R
a
depends on cutting speed and feed at a
radial depth of cut of 2 mm (Fig. 6, a) as well as on cutting speed and the radial depth of cut at a feed
of 0,1 mm/tooth (Fig. 6, b).
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
62
a) b)
Fig. 6. Centre line average height R
a
depending on cutting speed and feed (a) as well as on cutting
speed and radial depth of cut (b)
How the cutting parameters affect the average roughness height R
z
was analysed analogously to the
previous case, using equation (3):
0,042 1,895 0,052
303,4
z C Z e
R V f a
[µm]. (3)
Here feed also affects R
z
much more than cutting speed and the radial depth of cut of the tool, and
this character is maintained like in the previous case. This effect is confirmed by Fig. 7, which
illustrates how the average roughness height R
z
depends on cutting speed and feed at a radial depth
of cut of 2 mm.
Fig. 7. Average roughness height Rz depending on cutting speed and feed
Analysing the measured dependences of the average roughness width RS
m
on the varied cutting
parameters resulted in the following exponential equation:
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
63
0,124 1,345 0,012
1,25 04
m C Z e
RS V f a e
[µm]. (4)
It follows from this equation that feed f
Z
has the greatest effect on RS
m
with an exponent of 1,345.
The average roughness width here increases with growing feed. RS
m
also increases slightly with rising
radial depth of cut and even more with growing cutting speed. Fig. 8 illustrates this dependence, using
the variation in cutting speed and feed at a radial depth of cut of 2 mm as an example.
Fig. 8. Average roughness width RS
m
depending on cutting speed and feed
The dependences for other values of the radial depth of cut show a similar behaviour, which
corresponds to the established exponential equation (4). It can be noticed here that the values
calculated with the exponential equations (2) (4) are only valid for the analysed range of changes
in the cutting parameters. The character of how the cutting parameters affect the roughness of the
machined surfaces should, however, be maintained outside the examined range as well.
Microhardness. How the analysed cutting parameters affect the microhardness of the machined
boundary layers is described with the exponential equations (5), (6) and (7), which were developed
owing to the evaluation of all experimental tests performed. Equation (5) here represents how
microhardness depends on the examined cutting parameters. Equation (6) describes the elastic
penetration work and equation (7) represents the plastic penetration work depending on the examined
cutting parameters.
0.066 0.039 0.061
345.1
C Z e
HV V f a
[MPa], (5)
0.029 0.021 0.012
0.0958
e C Z e
W V f a
[µJ], (6)
0.081 0.03 0.066
0.3981
p C Z e
W V f a

[µJ]. (7)
As examples, Fig. 9 illustrates how microhardness depends on cutting speed and the radial depth of
cut of the milling cutter at a feed of 0,1 mm/tooth (Fig. 9, a) as well as on cutting speed and feed at a
radial depth of cut of 1,5 mm (Fig. 9, b).
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
64
a) b)
Fig. 9. Microhardness depending on the radial depth of cut and feed (a) as well as on cutting speed
and feed (b)
The diagrams show that microhardness decreases as a function of the radial depth of cut and increases
slightly with growing cutting speed. The analysis of the exponential equations (5) (7) guarantees
further, more precise evaluations of the obtained results. It shows very clearly that microhardness
decreases with increasing radial depth of cut. This involves an increase in elastic and plastic
penetration work. Furthermore, it can be observed that there is a steady increase in microhardness
with growing cutting speed. The elastic penetration work continuously increases then, reaching its
maximum at a high cutting speed and a great feed. Although the elastic penetration work rises, the
plastic penetration work steadily decreases as cutting speed increases. The growing feed causes a
slight increase in microhardness and the plastic penetration work as well as a decrease in the elastic
penetration work. At last, it could be detected that the machining direction showed hardly any
influence on the surface quality or the microhardness of the boundary layers after a machining
process.
When analysing the experimentally determined dependence of the quality characteristics on the
cutting parameters, it showed that the character of the dependences reacts very sensitively to the
condition of the tool and is considerably influenced by this condition. Hence, these dependences are
very strongly influenced by tool wear. This guarantees that the machining quality varies in different
stages of the tool’s service life. Hence, the same tool differs, for example, in the roughness of the
machined workpiece surfaces, in the microhardness of the boundary layers, in residual stress,
resultant forces and cutting temperatures, etc. along the whole path length. Fig. 10 shows how the
radius of cutting edge rounding and the wear land on the flank face of the wedge vary along the whole
path length of the tool. It can be detected that the predominant kind of wear is represented by the
breakaways of the cutting edge. At the beginning, these breakaways make up a relatively small part
of the cutting edge. However, these breakaways and their proportion are later growing until they are
very large at the end of the tool life. Thus the radius of cutting edge rounding, the wear land on the
flank face of the wedge and the resultant forces increase considerably at this moment in the tool life.
This changes not only the absolute value of the quality characteristics such as roughness and
microhardness but also the character of their dependence on the cutting parameters. That is the reason
why the wear condition of the tool has to be taken into account in experimental analyses for
establishing dependences, so that the tool wear must remain constant during the tests. In the present
case, such tests were carried out up to a path length of the milling cutter between 3000 and 3200 mm.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
65
Fig. 10. Wedge edge radius and wear on the flank face depending on the cutting way
Residual stress. The residual stress in the boundary layers of the machined specimens was established
by experiment and numerically modelled with FEM. Afterwards, the results were compared.
Experiment. Fig. 11 illustrates how cutting speed affects dry and wet cutting by cutting feed
0,1 mm/tooth and radial depth of cut 1,5 mm. It can be observed that the residual stress in wet cutting
is in the negative range or rather proves itself as compressive stress. The only exception is the
machining process at a cutting speed of 120 m/min. That there are compressive stresses after wet
cutting can be attributed to the predominant hardening of the machined material. In the machining
process with a cutting speed of 120 m/min, the softening of the machined material predominates due
to the considerably greater effect of the cutting temperature. Thus, compressive stress is transformed
into tensile stress at this cutting speed.
Fig. 11. Residual stress depending on cutting speed in wet and dry cutting
Due to a considerably higher cutting temperature than in wet cutting, the machined material is
hardened at all cutting speeds during the machining without supplying coolant. This leads to tensile
stresses. Consequently, the residual stresses in dry machining are within the range of tensile stresses
for all analysed cutting speeds.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
66
Simulation. For the numerical modelling of the residual stresses, FEM models were created for the
orthogonal cutting and the milling of the titanium alloy Ti-1023. The commercially available FEA
software DEFORM 2D/3D
TM
was used for creating the cutting models [11]. The simulation is based
on the implicit Lagrange formulation of networks. It is assumed for both models that the material of
the workpiece is isotropic of the elastoplastic type and the material of the tool is absolutely rigid. The
material model of the workpiece is represented as a Johnson-Cook model, which is well-
acknowledged for the modelling of different cutting processes [12]. The material model of the tool is
determined by the model of the software’s internal database. The damage mechanism is reproduced
with the Cocroft & Latham model in the modelling of the cutting processes [13].
The conditions regarding the thermomechanical interactions of the workpiece with the tool and with
itself remained unaltered during the entire simulation in the modelling of all cutting processes. In the
modelling of all cutting processes, the friction model in the tool-workpiece contact is assumed to be
of the shear type with a constant value of 0,6 and as Coulomb model with a constant value of 0,15 in
the contact of the tool with itself.
Fig. 12, a shows the geometry of the FEM model for orthogonal cutting. The figure also presents the
geometry of the wedge, the relative positioning and movement of tool and workpiece as well as the
boundary conditions used. The boundary conditions are established by fixing workpiece and tool and
by giving the thermal conditions at the boundaries of the respective objects. The bottom of the
workpiece is rigidly fixed in the X- and Z-directions (see Fig. 12, a). The right-hand side of the tool
is rigidly fixed in the Z-direction. The conditions of the room temperature (RT) are given at the bottom
and the left-hand side of the workpiece as well as at the left-hand side and the top of the tool. The
absolute movement against the X-direction is given to the tool.
The material and fracture model parameters for workpiece were established due to the flow curves
obtained by the tensile and compression tests in the examinations with the specimens of the titanium
alloy Ti-1023 [14], [15]. In addition, the values for the cutting processes were stated more precisely
using the inverse method [16], [17]. For both cutting models, the following values were used as
coefficients of the Johnson-Cook model: A=976.9 MPa; B=502.3 MPa; C=0.028; n=0.22; m=1. It
was assumed here that the critical breaking stress according to the Cocroft&Latham model is
260 MPa. The softening or rather the load-carrying capacity of the elements here was 10% after
breaking.
Fig. 12, b presents the geometry of the FEM model for milling. The figure also shows the geometry
of the milling cutter, the relative positioning and movement of tool and workpiece as well as the
boundary conditions used. The workpiece is rigidly fixed analogously to the FEM model for oblique
cutting (see Fig. 12, b). The milling cutter is rigidly fixed in the Z-direction. The thermal conditions
of tool and workpiece are given. The absolute rotating motion at a rotational speed n and the
translational movement at a feed speed V
f
against the X-direction are given to the tool.
Fig. 13 depicts the simulation of the residual stresses, showing how stress is distributed in the
workpiece in the respective direction. Fig. 13, a presents the simulation of the orthogonal cutting
process at the cutting speed 32 m/min and cutting depth 0,3 mm, whereas Fig. 13, b illustrates the
simulation of the milling process at the cutting speed 90 m/min, cutting feed 0,1 mm/tooth and radial
depth of cut 1,5 mm. The height of the stresses is shown in the scale on the right-hand side.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
67
a) b)
Fig. 12. Initial geometry and boundary conditions of the orthogonal cutting model (a) and of the
milling model (b)
a) b)
Fig. 13. Simulation of stress distribution in orthogonal cutting (a) and milling (b)
Fig. 14 compares the experimental and the simulated residual stresses when milling at different
cutting speeds without supplying coolant. As can be inferred from the figure, the greatest difference
between simulated and experimental data was the relative value of 32%.
It will surely be possible to eliminate the error by modelling the thermomechanical interactions in the
shear zones as well as the fracture processes more precisely. Finally yet importantly, the mechanical
properties of the machined material that are suitable for the cutting process will have to be determined
more exactly here.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
68
Fig. 14. Comparison between experimental and simulated residual stress in milling
Summary. This paper presents the results of experimental tests into the formation of quality
characteristics such as roughness and microhardness as well as residual stresses and their simulation
depending on cutting parameters such as cutting speed, feed and radial depth of cut.
To analyse more closely how the cutting parameters affect the quality characteristics in the milling
of the titanium alloy Ti-1023, the individual dependences of the effects were described in exponential
equations. The exponents for the exponential equations were established according to the Gaussian
elimination method. The analysis of the exponents guarantees that a clear effect of the respective
cutting parameters on the quality characteristics to be examined can be established. The values
calculated with the exponential equations are only valid for the analysed range of changes in the
cutting parameters. The character of how the cutting parameters affect the quality characteristics of
the machined surfaces should, however, be maintained outside the examined range as well.
It was found out that feed f
Z
has the greatest effect on the centre line average height R
a
. The cutting
speed and the radial depth of cut of the milling cutter have a considerably smaller effect. Here R
a
decreases slightly with increasing cutting speed and rises slightly with growing radial depth of cut.
Feed also affects R
z
much more than cutting speed and the radial depth of cut of the tool, and this
character is maintained like in the dependences of R
a
on the cutting parameters. When taking the
dependences of the average roughness width RS
m
on the changed cutting parameters into account, it
could be found that feed f
Z
also has the greatest effect on RS
m
. The average roughness width here
increases with growing feed. RS
m
also increases slightly with rising radial depth of cut and even more
with growing cutting speed.
The microhardness in the boundary layers of milled specimens decreases with increasing radial depth
of cut. This involves an increase in elastic and plastic penetration work. Growing cutting speed leads
to a steady increase in microhardness. The elastic penetration work continuously increases then,
reaching its maximum at a high cutting speed and a great feed. Although the elastic penetration work
rises, the plastic penetration work steadily decreases as cutting speed increases. The growing feed
causes a slight increase in microhardness and the plastic penetration work as well as a decrease in the
elastic penetration work.
Experimental tests of the residual stress in wet cutting showed that it is in the negative range or rather
proves itself as compressive stress. The only exception is the machining process at a cutting speed of
120 m/min. Due to a considerably higher cutting temperature than in wet cutting, the machined
material is hardened during the machining without supplying coolant. This leads to tensile stresses at
all cutting speeds.
The developed FEM cutting models of orthogonal cutting and milling guarantee simulating the
residual stress in the machined workpiece. Comparing the simulated and the measured residual
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
69
stresses in the machining without supplying coolant shows a satisfactory agreement with a greatest
relative error of 32% and confirms that the created FEM cutting models are valid.
The experimental results obtained and the developed FEM cutting models serve as a basis for further
optimising the machining processes of titanium alloys.
Acknowledgements. The presented results were gained in the project 2013DFB70110, which was
funded by the International Science & Technology Cooperation Program of China. The authors would
like to thank the Government of the Republic of China for this support, which is highly appreciated.
References
[1] Boyer, R., & Briggs, R. (2005). The Use of β Titanium Alloys in the Aerospace Industry. Journal
of Materials Engineering and Performance, 14(6), 681-685. doi: 10.1361/105994905x75448.
[2] Biermann, D.; Machai, C. (2010). Die Alternative zu Ti-6Al-4V. Werkstatt und Betrieb, 143(4),
5658.
[3] Machai, C., & Biermann, D. (2011). Machining of β-titanium-alloy Ti10V2Fe3Al under
cryogenic conditions: Cooling with carbon dioxide snow. Journal of Materials Processing
Technology, 211(6), 1175-1183. doi:10.1016/j.jmatprotec.2011.01.022.
[4] Yang, H. C., Chen, Z. T., & Jiang, L. K. (2013). Experimental Study on Hardness of Titanium
Alloy Ti-1023 by Milling. AMR Advanced Materials Research, 690-693, 2446-2449.
doi:10.4028/www.scientific.net/amr.690-693.2446.
[5] Changfeng, Y., Daoxia, W., Liang, T., Junxue, R., Kaining, S., & Zhenchao, Y. (2013). Effects
of cutting parameters on surface residual stress and its mechanism in high-speed milling of TB6.
Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture,
227(4), 483-493. doi:10.1177/0954405413475771.
[6] Houchuan, Y., Zhitong, C., & Zitong, Z. (2014). Influence of cutting speed and tool wear on the
surface integrity of the titanium alloy Ti-1023 during milling. The International Journal of Advanced
Manufacturing Technology Int J Adv Manuf Technol, 78(5-8), 1113-1126. doi:10.1007/s00170-014-
6593-x.
[7] Wagner, V., & Duc, E. (2014). Study of Ti-1023 milling with toroidal tool. The International
Journal of Advanced Manufacturing Technology Int J Adv Manuf Technol, 75(9-12), 1473-1491.
doi:10.1007/s00170-014-6217-5.
[8] DIN EN ISO 4287. (1997). Geometrical Product Specifications (GPS) - Surface texture: Profile
method - Terms, definitions and surface texture parameters.
[9] DIN EN ISO 14577. (2015). Instrumented indentation test for hardness and materials parameters,
2015.
[10] Beutelspracher, A. Lineare Algebra. (2014). Eine Einführung in die Wissenschaft der Vektoren,
Abbildungen und Matrizen. Springer Spektrum, 386.
[11] Deform-User Manual SFTC-Deform V11.0.2. (2014). Columbus (OH), USA.
[12] Johnson, G. R.; Cook, W. H. (1983). A constitutive model and data for metals subjected to large
strains, high strain and high temperatures. Proc. 7th Int. Symp. On Ballistics, The Hague, Netherlands,
541547.
[13] Cockroft, M. G.; Latham, D. J. (1968). Ductility and workability of metals. Journal of the
Institute of Metals, 96, 3339.
[14] Robertson, D. G., & Mcshane, H. B. (1998). Analysis of high temperature flow stress of titanium
alloys IMI 550 and Ti-10V-2Fe-3AI during isothermal forging. Materials Science and Technology
Mats. Sci. Tech., 14(4), 339-345. doi:10.1179/026708398790301403.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
70
[15] Bao, R., Huang, X., & Cao, C. (2006). Deformation behavior and mechanisms of Ti-1023 alloy.
Transactions of Nonferrous Metals Society of China, 16(2), 274-280. doi:10.1016/s1003-
6326(06)60046-0.
[16] Shrot, A., & Bäker, M. (2012). Determination of JohnsonCook parameters from machining
simulations. Computational Materials Science, 52(1), 298-304.
doi:10.1016/j.commatsci.2011.07.035.
[17] Klocke, F., Lung, D., & Buchkremer, S. (2013). Inverse Identification of the Constitutive
Equation of Inconel 718 and AISI 1045 from FE Machining Simulations. Procedia CIRP, 8, 212-
217. doi:10.1016/j.procir.2013.06.091.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
71
A Comparison between Dual Phase Steel and Interstitial Free Steel Due To the
Springback Effect
E.A. Silva ¹, L.F.V.M. Fernandes ², N.A.S. Sampaio
3,4
, R.B. Ribeiro
2,5
, J.W.J. Silva
2,3
,
M.S. Pereira ¹
1 Universidade Estadual Paulista UNESP, Campus de Guaratinguetá, SP, Brazil
2 Faculdades Integradas Teresa D’Ávila – FATEA, Redes Salesianas, Lorena, SP, Brazil
3 Associação Educacional Dom Bosco – AEDB, Resende, RJ, Brazil
4 Universidade do Estado do Rio de Janeiro UERJ/FAT, Resende, RJ, Brazil
5 Faculdade de Tecnologia do Estado de São Paulo – FATEC, Cruzeiro, SP, Brazil
DOI 10.13140/RG.2.1.3749.7205
Keywords: springback, dual-phase steel, interstitial free steel, aspect ratio.
ABSTRACT. This is a study of the springback effect on two kinds of high strength steel, which are: dualphase and
interstitial free, currently used as feedstock in the production of vehicles. The mechanical characterization of the
springback effect was performed by means of a mechanical conformation test, called three-point air bending, performed
by adapting it to the unconstrained cylindrical bending test. It was also evaluated the mechanical properties of the material
defined by the tensile test in order to determine its tensile strength, yield strength, and elongation. Furthermore, it was
performed a microstructural characterization of advanced steels by identifying and quantifying the present phases in
coexistence by means of digital image processing. The results indicate that the springback effect in the dual-phase steel
has the highest springback rates due to its high mechanical strength, and it causes a decrease in the aspect ratio of the
grains that suffered mechanical conformation attempting to return it to its original form. On the other hand, the springback
effect has the lowest rates, and the change in aspect ratio depends only on the interstitial free steel elongation capacity
due to its lower mechanical strength.
1. Introduction. From the 1950’s, there was a concern of the automotive industry to produce steel
with high formability and low price, and this influenced in changes in vehicle models. In the 1970’s,
due to the oil crisis, many countries took serious measures to reduce and rationalize energy
consumption. The automobile industry then had to find solutions, bringing developments in
aerodynamics and reducing vehicles weight. In addition, in all countries, there can be no growth in
the automotive sector without giving due importance to the steel mills [1].
The automobile industry has taken steps such as reducing the size of vehicles, replacement of
materials, which are traditionally used for aluminum and plastic, and replacement of carbon steels.
As a result, steelmakers aimed primarily to promote the assembly of advanced materials with high
strength, ensuring an increased structural integrity, an increased resistance to shock, with a lower cost
to the final product [2].
A system sets the conventional high-strength steel (HSS) as those having yield strength between 210
and 550 MPa and tensile strength between 270 and 700 MPa, while the high strength advanced steel
(AHSS) has a yield strength that is greater than 550 MPa, and tensile strength greater than 700 MPa
[3].
The main difference between the HSS and the AHSS is their microstructure. HSS steels are ferritic-
pearlitic, of a single phase. AHSS are steels which mainly contain a microstructure with a different
phase than ferrite, for example, martensite, bainite, austenite and/or retained austenite in sufficient
quantities to produce their own mechanical properties. However, the widespread use of AHSS in the
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
72
automotive industry is limited due to challenges in formability, metal sheet, tool life, and to the
springback effect. It is a major problem that compromises the mass production of automotive
structural components with AHSS [4].
According Ramezani [5], it is a partially undesirable change occurring in steel sheets as a result of
the constraints removal after their conformation process. The two types of steels studied in this work
are among the major steels used by automobile industries nowadays because they are high strength
steels, being dual-phase steel (DP), which is an AHSS, and interstitial free steel (IF), which is a HSS.
2. Experimental procedure.
2.1. Metallography before the conformation. The following metallographic preparation processes
are standardized by ASTM ID: E 3-10 (2007). Test specimens were made from DP and IF steels as
delivered at the dimensions of 10 mm long, 10 mm wide, 0.8 mm thick.
After sectioning in the longitudinal direction of rolling, the specimens were subjected to hot
embedding with bakelite. During the grinding, the following abrasives were used: 220, 320, 400, 600,
1000, and 1200 mesh. The polishing was performed in an OP-U solution and distilled water,
subjecting the specimens to a rotation of 600 rpm. The chemical etching was done with a 2% Nital
solution to reveal the grain boundaries of ferrite and constituents [6].
The micrographs were obtained using the NIKON optical microscope, model EPIPHOT 200. The
image processing was done using the Image J 1.45 software. All images were standardized in the
same luminosity conditions and gray shades scale with the use of tools to enhance contrast, normalize
and equalize histogram.
2.2. Mechanical conformation tests. Test specimens were made from the same material as delivered
and sectioned at the following dimensions: 80 mm long, 30 mm wide, and 0.8 mm thick. Such
dimensions of the specimens were made according to the parameters defined for the unconstrained
cylindrical bending test presented at the Numisheet conference 2002 [7].
The specimens were subjected to a test called the three-point air bending. This experiment was made
in adapting the method to the unconstrained cylindrical bending test, in which the specimen is
subjected to a punch with the cylindrical body.
The punch had a 5 mm radius and the distance between the supports of the die was 13 mm, according
to the ASTM ID: E 290-09 standards to a sample thickness of about 1 mm. The three-point air bending
was performed in a universal Shimadzu testing machine, Autograph AG-X, model 50 kN.
The specimens were subjected to conformation until the internal angle of bending reached a
predetermined value. The values selected for the internal angle bending were: 30, 60, 90 and 120
degrees, respectively, for each bend, using three replicates for each angle in the same material. The
punch was removed from the material 20 seconds after reaching the bending angle, and then, the new
bend angle measurement was made to determine whether there was a springback effect or not. For
this measurement, it was used the Image J 1.45 software for processing images photographed on an
Olympus digital camera. Such measurements continue to be made for a period of 12 h, 24 h, 48 h and
72 h after conformation. Completed the 72 h after the mechanical bending, the resulting bending
angle was subtracted from the initial angle of bending which were of 30º, 60º, 90º or 120º,
respectively, and this subtraction resulted in a total springback angle (θ1 + θ2), as shown in Figure 1.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
73
Fig. 1. Definition of springback angle [8].
Fig. 2. Device mounted universal testing machine for testing three-point bending in air.
2.3. Metallography after the conformation. Test specimens were made from materials that were
subjected to a mechanical bending test after 72 hours of evaluation at the following dimensions: 10
mm length x 10 mm wide x 0.8 mm thick. The region of the steel plates chosen to obtain the specimens
was the one where a curvature was formed due to bending. The specimens were cut in a longitudinal
direction, i.e., in the direction of the steel plates lamination by dividing them in the middle. This
material was embedded in order to expose its inner surface in order to obtain samples of the region
that was deformed during the bending.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
74
The metallographic followed in the same manner as in the pre-conformation test, with the use of hot
embedding and polishing, and the chemical etching was made with a 2% Nital solution. The 30
analyzed pictures were for each treatment.
3. Results and discussion.
With respect to the mechanical properties, shown in Table 1, they were obtained by tensile tests,
extracting the specimens in a transverse direction at 45°, and in the rolling direction of the material.
In table I, the tensile strength was designed by RT in MPa, yield strength by LE in MPa, and
elongation by Elong in %.
Table 1. Mechanical properties of materials dual-phase steel (DP) and interstitial free steel (IF)
Material
RT (MPa)
LE (MPa)
Along (%)
DP
623.6 ± 2.9
407.3 ± 3.6
23.4 ± 1.4
IF
298.0 ± 2.1
147.9 ± 3.8
40.9 ± 1.9
As shown in Figure 3, the dual-phase steel showed the greatest springback angle (between 7 and 9
degrees), and the interstitial free steel had lower angles (between 4.5 and 6 degrees).
Thus, Figure 3 shows the results that can be compared with W.Gan’s work, whereby it was concluded
that materials with higher yield strength tend to have greater springback effect, as compared to other
materials with lower yield strength [9].
Moreover, as shown in Figure 2, the angular variation of the springback effect was increased from
120° to 30°. This means that as the extent of the internal angle bending was being reduced from 120°,
90°,60° to 30°, an increase in the springback effect occurred.
According to the dual-phase steel, a decrease in the internal angle bending causes a greater springback
effect.
Fig. 3. Angle variation of the springback effect for different bending angles after 72 hours
The ANOVA was used as a statistical tool for interpreting the results. It means a variance analysis,
and it is a test of treatment average comparison. It was used a two-factor ANOVA type with repetition,
submitted to an F test, at a significance level of 5%.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
75
According to Table II, the time and degree factors significantly influenced the results of springback
effect in the dual-phase steel. It can be verified that, according to F Test, - p value was lower than
0.05 which were different from each other in average. The same is not true for the interactions where
the p value is greater than 0.05, indicating that the treatment time and degree did not cause
interactions together with the results.
Table 2. Analysis of variance of springback effect to dual-phase steel
Source of variation
SQ
gl
MQ
F
Value-P
F-critic
Time
9.487148
4
2.371787
3.188721
0.023015831
2.605975
Degree
59. 80007
3
19.93336
26.79916
1.14078E-09
2.838745
Interactions
1.196571
12
0.099714
0.13406
0.999686828
2.003459
Inside
29. 75221
40
0.743805
Total
100.236
59
Examining Table III, where an analysis of variance for the aspect ratio is shown, - p value is less than
0.05, thus the variation in the aspect ratio for the dual-phase steel differentiates along the treatment
degrees at a significance level of 5%.
Table 3. Analysis of variance of the aspect ratio after 72 hours of the springback effect, considering
the four internal angles of bending applied to the dual-phase steel from 30 photos
ANOVA
Source of variation
SQ
gl
MQ
F
Value-P
F-critic
Among groups
0.03813
4
0.009533
2.631578
0.036708
2.434065
Within groups
0.525241
145
0.003622
Total
0.563372
149
An analysis of variance for the interstitial free steel subjected to the springback effect is shown in
Table IV which indicates that the difference between the means were significant only for the factor
degree, and this is the only one where the - p value is less than 0.05. The time factor and the
interactions between time and degree did not have an influence on the springback effect.
Table 4. Analysis of variance of springback effect to interstitial free steel.
Source of variation
SQ
gl
MQ
F
Value-P
F-critic
Time
3.204853
4
0.801213
0.535429
0.710460251
2.605975
Degree
24.25645
3
8.085484
5.403309
0.003232364
2.838745
Interactions
0.260574
12
0.021714
0.014511
0.999999999
2.003459
Inside
59.8558
40
1.496395
Total
87.57768
59
Table 5 has an analysis of variance to evaluate the influence of the springback effect in the aspect
ratio. It was concluded that this effect caused a significant change in aspect ratio, since the - p value
was less than 0.05.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
76
Table 5. Analysis of variance of the aspect ratio after 72 hours of the springback effect, considering
the four internal angles of bending applied to the interstitial free steel from 30 photos.
ANOVA
Source of variation
SQ
gl
MQ
F
Value-P
F-critic
Among groups
3.56368
4
0.89092
2.551933382
0.041591
2.434065
Within groups
50.62177
145
0.349116
Total
54.18545
149
As shown in Figure 4, the aspect ratio for the 180° angle refers to the material as delivered before
suffering mechanical bending. In this graph, it is seen that the treatment is the most different from the
others which was at 120°, thus obtaining a greater aspect ratio. As for the other treatments, the aspect
ratio values are closer to the material as delivered.
Fig. 4. Average aspect ratio values and their standard deviation of vertical bars 72 hours after
removal of constraints obtained from 30 pictures for each internal angle of bend of 30° to 120°, and
180° to material as received.
To view the mechanical influence of the springback effect on the conformation and aspect ratio, it is
mounted the graph of Figure 5 from which it is observed that the treatment at 30° was the only one
that had the highest mean values of the aspect ratio.
Mechanics, Materials Science & Engineering, May 2016 ISSN 2412-5954
MMSE Journal. Open Access www.mmse.xyz
77
Fig. 5. Average aspect ratio values and their standard deviation of vertical bars 72 hours after
removal of constraints obtained from 30 pictures for each internal angle of bend of 30° to 120°, and
180° to material as received.
Fig. 6. Samples of sheets, for illustration, already subject to three-point bending in air and subsequent
springback effect for 72 hours, where the numbers refer to the values of angles of bending applied,
with (1) was 120°, (2 ) was 90° (3) was 60° and (4) was 30° and the region conformed subjected to
the mounting respectively of 30° to 120°.
Figure 7 from (a) to (d) are images of the dual-phase steel microstructure obtained by an optical
microscopy. It is observed the presence of martensite microstructure (dark portion) in all images, like
islands immersed in the matrix of ferrite (light portion). Images (c) and (d) differ from each other
because it is observed the