investigation on wear behaviour of lm13/sic …lm 13 aluminium alloy is used as metal matrix and its...
TRANSCRIPT
http://www.iaeme.com/IJMET/index.asp 530 [email protected]
International Journal of Mechanical Engineering and Technology (IJMET) Volume 8, Issue 6, June 2017, pp. 530–543, Article ID: IJMET_08_06_056
Available online at http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=8&IType=6
ISSN Print: 0976-6340 and ISSN Online: 0976-6359
© IAEME Publication Scopus Indexed
INVESTIGATION ON WEAR BEHAVIOUR OF
LM13/SiC ALUMINIUM METAL MATRIX
COMPOSITES BY RESPONSE SURFACE
METHODOLOGY
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
Department of Mechanical Engineering, Karunya University,
Coimbatore, Tamil Nadu, India
ABSTRACT
The present study focuses on the influence of Silicon carbide (SiC) particulates on
the wear behaviour aluminium alloy metal matrix composites using a statistical
technique called Response Surface Methodology (RSM). LM13 aluminium alloy is
reinforced with SiC particulates using stir casting (compocasting) technique. The
weight fraction of particles in the Aluminium Metal Matrix Composite (AMMCs) was
varied from 0 to 12 wt. % in steps of 3wt. %. The wear experiments were conducted
using four factors and five levels central composite rotatable design (CCD). Dry
sliding wear tests were conducted by pin on disc apparatus to study the influence of
sliding speed, sliding distance, normal load and reinforcement wt% on the wear rate
of composite specimens. The results were analyzed using analysis of variance
(ANOVA) for identifying the significant factors affecting the performance at 95%
confidence interval. The empirical quadratic model relationships were established
using RSM to predict the influence of wear parameters on the performance parameter
(wear rate) with reasonably good accuracy. RSM was used to optimize the wear
parameters for a minimum wear rate. Results showed that normal load is the most
influencing factor which increases the wear rate and sliding speed is a least factor
which affects the wear rate. The statistical analysis is carried out using Design Expert
10 software.
Key words: Aluminium matrix composites, silicon carbide, compocasting, wear,
response surface methodology, optimization.
Cite this Article: J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin
Prabhu. Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix
Composites by Response Surface Methodology. International Journal of Mechanical
Engineering and Technology, 8(6), 2017, pp. 530–543.
http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=8&IType=6
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 531 [email protected]
1. INTRODUCTION
Fuel economy is of great importance in the present automotive industry. Frictional losses in
heavy duty diesel engines (HDDE) lead to 2-2.5% of fuel consumption in the normal driving
condition. Half of these frictional losses are mainly due to the piston and piston rings [1].
Every automotive industry focuses on to manufacture parts with light weight, excellent
tribological properties and better performance. A special consideration is given during
manufacturing in parts such as pistons, piston rings, bearings, bushes and brake system
components where wear resistance is highly essential. AMMCs is considered as a potential
alternative material with conventional monolithic aluminium alloys in many applications
owing to its high specific strength and stiffness, low density, low thermal expansion
coefficient and high wear resistance. AMMCs are used in numerous industries that are not
limited to aerospace, automotive, defense, naval, electronic packaging, thermal and sports [2–
5]. AMMCs reinforced with SiC particulates exhibit higher modulus, strength and wear
resistance compared to conventional alloys. Among various AMMCs manufacturing
techniques liquid metallurgy is a commonly preferred method due to its simplicity, flexibility
and applicability for producing large quantities [6]. LM13 aluminium alloy containing 12%
silicon is widely used in manufacturing internal combustion engines parts such as pistons,
cylinder blocks, cylinder heads due to its high resistance to wear, corrosion & thermal
conductivities [7]. V.D. Londhe et al. have observed the wear increases with increase in
normal load at ambient temperature and the wear decreases with increase in normal load at
elevated temperature (125°C) of LM 13-SiC10% composite [8]. N.Radhika et al. have
investigated wear characteristics of LM 13/B4C Composites and concluded the wear increases
with increasing in normal load and decreases with increase in sliding velocity, sliding distance
and wt% of reinforcement [9]. S.Das has conducted a sliding wear, abrasive wear, erosion-
corrosion wear study on LM13-SiC composites and concluded the wear rate increases beyond
the seizure pressure and addition of SiC particles leads to improvement in seizure pressure
with that of unreinforced aluminium alloy. The wear rate of composites decreased with
increase in wt% of SiC particles. The alloy exhibited greater wear rate compared to the
composite in acidic medium and sand [10]. Garcia et al. have observed that the specific wear
rate of AA6061-SiC composite decreases with increase in volume fraction and size of
reinforcement [11]. Sahin et al. have observed a significant increase in wear resistance in
composites upto 10% addition of SiC and a saturated wear resistance beyond 10% to 55% SiC
addition produced by vacuum infiltration method. Among various other ceramics, Silicon
Carbide (SiC) which exhibits high elastic modulus (410 GPa), low density (3.2 g/cm3) and
high Vickers hardness (2600 HV) is very attractive for various automobile applications. SiC
is known to have better chemical compatibility with aluminium because it doesn’t forms any
inter-metallic phases during its interaction with the Al matrix, so it is a very common type of
reinforcement used in Al-MMCs [12]. The mechanical strength and wear resistance of
composites increases by adding SiC particulates to the matrix alloy. Prashant et al. conducted
experiments on Al 6061-SiC composite and concluded the hardness of metal matrix
composite increases with increase in reinforcement content and the wear rate of the Al6061-
SiC composite decreased with increasing SiC content [13]. Y.Sahin studied tribological
behaviour of 10 wt. % of SiCp with average particle size of 100 μm reinforced with Al 2014
alloy using orthogonal arrays and analysis of variance (ANOVA).The results indicated that,
the reinforcement exhibited greater effect on abrasive wear followed by applied load, whereas
sliding distance had the least effect [14]. Ravikiran et al. carried out the effect of sliding speed
on wear behaviour of A356 aluminium reinforced with 30 wt. % SiCp. The wear rate reduced
continuously with increase in speed [15]. The influence of input variables on responses in an
experiment is determined using statistical modelling technique. The response surface
methodology (RSM) is an empirical modeling approach for defining the relationship between
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 532 [email protected]
various process parameters and responses with the various desired criteria and searching the
significance of these process parameters on the coupled responses. It is a sequential
experimentation strategy for ramping up and optimizing the empirical model. Response
surface methodology (RSM) uses various statistical, graphical, and mathematical techniques
to acquire, improve, or optimize a process [16, 17]. RSM is being extensively used to analyze
engineering problems where input variables influences performance of response variables. It
provides quantitative data of possible interactions between input and response factors which
are difficult to obtain using other optimization techniques [18-23]. However various
tribological studies have been conducted on unlubricated metal matrix composites (MMC)
using RSM, parametric studies on tribological behaviour of piston alloy does not exist. An
attempt is made here to study the effect of SiC on the wear behaviour of LM13 aluminium
alloy (piston alloy) using statistical tools. The Design of experiments (DoE), analysis of
variance, and regression analysis, prediction and optimization are carried out through
Response Surface Methodology (RSM) using Design Expert 10 software.
1.1. Aim of Research
Improvement in the fuel economy and emission is a major challenge to the present automotive
industry. The vehicle weight and frictional losses are the two major factors which affects the
fuel economy. The friction encountered within the I.C Engine itself contributes about 40% of
power loss from the produced power as well wear of the piston rings and liner. Even a small
improvement in engine would reflect in the efficiency, fuel economy, emission levels which
can have a major effect on the worldwide fuel economy and the environment in the long-term.
One such attempt is carried out to find the wear behaviour of eutectic aluminium alloy (LM
13) as metal matrix reinforced with Silicon Carbide particulates.
2. EXPERIMENTAL SETUP
2.1. Development of Design Matrix
The most popular design in response surface methodology is central composite design (CCD).
The various factors and their levels which are used for analysis and construction of design
matrix are given in Table 1. The design matrix was developed using rotatable central
composite design with 31 set of trial runs, input and response results are shown in Table 2.
The literatures of the design matrix are available elsewhere [24, 25].
2.2. Production of Composite by Compocasting Process
Al–SiC composites were fabricated using compocasting process [26, 27]. LM 13 aluminium
alloy is used as metal matrix and its chemical composition is given in Table 3. The
reinforcement, SiC particulates details is given in Table 4. In the first step 1 kg of aluminium
alloy was measured and melted at 800◦C in a graphite crucible using a stir casting furnace.
According to the wt% SiC particulates were measured and preheated at a temperature of
400◦C for about 30 min in the preheating furnace. The preheating was done to remove the
surface impurities and reduce the oxide formation
Table 1 Factors and their levels in CCD experimental deign
S.No. Parameters Notations Unit Levels
-2 -1 0 1 2
1 Sliding Velocity V m/s 0.4 0.8 1.2 1.6 2
2 Sliding Distance D m 400 800 1200 1600 2000
3 Normal Load F N 10 20 30 40 50
4 Silicon Carbide S wt % 0 3 6 9 12
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 533 [email protected]
Table 2 Design matrix with its experimental results
Trial
Run
Sliding wear parameters Response
parameter
Sliding Velocity
(V)
Sliding
Distance
(D)
Normal
Load
(F)
Reinforcement
(SiC )
(S)
Wear Rate
(W)
X 10-5
mm3/m
01 0.8 800 20 3 848
02 1.6 800 20 3 732
03 0.8 1600 20 3 643
04 1.6 1600 20 3 542
05 0.8 800 40 3 1509
06 1.6 800 40 3 1216
07 0.8 1600 40 3 1218
08 1.6 1600 40 3 938
09 0.8 800 20 9 650
10 1.6 800 20 9 520
11 0.8 1600 20 9 461
12 1.6 1600 20 9 359
13 0.8 800 40 9 1295
14 1.6 800 40 9 1013
15 0.8 1600 40 9 985
16 1.6 1600 40 9 731
17 0.4 1200 30 6 856
18 2 1200 30 6 474
19 1.2 400 30 6 844
20 1.2 2000 30 6 455
21 1.2 1200 10 6 355
22 1.2 1200 50 6 1260
23 1.2 1200 30 0 1795
24 1.2 1200 30 12 366
25 1.2 1200 30 6 751
26 1.2 1200 30 6 706
27 1.2 1200 30 6 864
28 1.2 1200 30 6 810
29 1.2 1200 30 6 776
30 1.2 1200 30 6 750
31 1.2 1200 30 6 819
by absorption of gases. Once the metal is melt it was continuously stirred at 600–800 rpm
to create a vortex with the help of a mechanical stirrer for 10 mins during which a
hexochloroethane tablet (C2Cl6) was added to the melt to degas and liberate any unwanted
gases generated during the melting of the aluminium. The preheated SiC particles were added
slowly and continuously into the vortex of the molten metal. To improve wettability 1 wt. %
of Mg was added to the molten metal [28]. The stirrer was frequently moved vertically up and
down within the mixture to ensure uniform distribution of the added particles. After all the
SiC particulates were added into the molten metal the temperature of the furnace was set to
550◦C and the composite mixture was allowed to attain the solidus state in the crucible. In the
second step the slurry mixture was reheated and melted again at 750◦C. The molten metal was
again stirred at 300 rpm for about 2 mins. Finally it was cast into a 100 x 100 x 10 mm
preheated m.s mould. The composite was allowed to solidify in the atmospheric air and was
removed from the mould after solidification. The AMMCs with different weight percentage
(3, 6, 9 and 12 wt. %) of SiC particulates were produced by the same procedure.
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 534 [email protected]
Table 3 Chemical Composition of LM 13 aluminium alloy
Eleme
nt Al Si Fe Cu Mn Mg Ni Ga Zn Ti Pb Sn
Conten
t %
83.4
2
13.0
2
0.31
0
1.04
7
0.25
2
0.86
5
0.93
8
0.01
4
0.006
8
0.06
8
<0.00
5
<0.004
9
Table 4 Details of Reinforcements
Reinforcements Hardness (GPa) Grain size (µm) Density(g/cm3)
SiC 24.5-2 25 3.22
2.3. Micro-hardness Measurement
The micro-hardness (HV) of the composites was measured as per ASTM A-370 using Vickers
hardness tester (MITUTOYO-HM-114).The test was conducted on the polished specimens at
a load of 500 g applied for duration of 15 s at 5 different locations on all the specimens.
2.4. Macro-HARDNESS Measurement
The macro-hardness (HRC) of the composites was measured using Rockwell hardness tester
(model RAB-250). The test was conducted according to ASTM E-18 standard at a load of 100
kg applied for duration of 15 s at 5 different locations.
2.5. Wear Test
The wear test specimens 10 mm x 10 mm x 50 mm are obtained from the cast composites by
machining. The end surface of the specimens are cleaned and polished with 600 grade
followed by 1000 grade abrasive paper. The dry sliding wear test was conducted on pin-on-
disc wear apparatus (DUCOM TR20-LE) at room temperature according to ASTM G9905
standard. A grey cast iron disc with micro hardness (HV) and surface roughness 0.0001
microns was used to conduct the test. A computer aided data acquisition records the height
loss. The height loss is multiplied with the area of cross section of the composite pin to obtain
volume loss. Finally the volume loss is divided by sliding distance to obtain wear rate. The
experiments were carried out based on design matrix and the wear rate is computed for 31
trial runs which is presented in table 2. The wear surface of selected specimens was observed
using scanning electron microscope.
3. RESPONSE SURFACE MODEL FOR PREDICTION OF WEAR RATE
In the traditional experimental design one factor is varied at a time by keeping the other
factors constant to find the output response which results in more number of experiments for
various factors and levels. It is also very difficult to find the combined effect of the input
factors on the response in the experiment. Response surface methodology (RSM) is one of the
popular design which simplifies and reduces the number of experiments as well as it helps to
find the combined effect of input parameters on the output responses. In this study a RSM
with full factorial design of experiments consisting of four factors (sliding velocity, sliding
distance, normal load and reinforcement content) and five levels (-2,-1,0,+1,+2) was used. In
RSM the input process parameters are represented in the form of response in the quantitative
form as:
Y = f (X1, X2, X3, . . ., Xn) ± ε (1)
Where Y is output or response factor, f is response function and ε is experimentation error
respectively. X1, X2, X3. . . Xn are independent input process parameters.
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 535 [email protected]
The response function f is formulated by independent input process parameters by
applying a lower order polynomial equation. If the model is suitable the response of the model
can be represented by a linear function in terms of independent process parameters, then the
response represented in Eq.(1) can be written as
Y= C0 + C1X1 + C2X2 + . . .Cn Xn ± ε (2)
However, there are possibilities of curvature to be formed in the developed response
system. In this case a higher order polynomial, i.e., quadratic equation will be used to
represent the response equation and following equation may be used.
n n
Y = C0 + ∑ Ci Xn + ∑ di Xi2± ε (3)
i=1 i=1
As the wear rate of the composite is a function of sliding velocity, sliding distance, normal
load and wt % of SiC particulates, Eq.1 is rewritten as
W = f (V, D, F, S) (4)
The second order polynomial regression Eq.3 is rewritten with the four input factors is
expressed as:
W = C0+C1V + C2D C3F+ C4S + C12VD + C13VF + C14VS + C23DF + C24DS + C34FSC11V2
+ C22D2+C33F
2+C44S
2+ (5)
The experiment results were statistically analyzed by RSM technique using Design Expert
10 software which is widely used in many engineering research fields. Analysis of variance
(ANOVA) is performed to check the statistical significance of the quadratic model of wear
rate is presented in Table 5.
Table 5 ANOVA table for Wear Rate
Source Sum of
squares
Degree of
freedom Mean square F-value Prob.>F
Model 3.074 x 106 14 2.196 x 10
5 5.44 0.0009 Significant
V-Sliding velocity 2.838 x 106 1 2.838 x 10
5 7.03 0.0174
D-Sliding distance 3.680 x 105 1 3.680 x 105 9.11 0.0082
F-Normal load 1.340 x 106 1 1.340 x 10
5 33.20 <0.0001
S- Silicon carbide 7.357 x 106 1 7.357 x 10
5 18.22 0.0006
VD 1444 1 1444 0.036 0.8524
VF 21904 1 21904 0.54 0.4721
VS 132.25 1 132.25 3.275 x 10-3
0.9551
DF 7569 1 7569 0.19 0.6708
DS 132.25 1 132.25 3.275 x 10-3
0.9551
FS 1406.25 1 1406.25 0.035 0.8543
V2 618.79 1 618.79 0.015 0.9030
D2 2079.08 1 2079.08 0.051 0.8234
F2 27433.04 1 27433.04 0.68 0.4219
S2 2.815 x 106 1 2.815 x 106 6.97 0.0178
Residual 6.461 x 105 16 40381.77
Lack of fit 4.496 x 105 10 44959.95 1.37 0.3623 Not significant
Pure Error 1.965 x 105 6 32751.48
Cor. Total 3.720 x 106 30
Std. Dev. 200.95 R-Squared 0.9263
Mean 812.45 Adj R-
Squared 0.8744
The results are analyzed with confidence level 95% or p-value of 0.05. This implies that
any factor with p-value equal to or less than 0.05 is significant and greater than 0.05 is termed
non-significant. ANOVA shows the “Model” as “Significant” while the “Lack of fit” is “Not
significant”, which are desirable for a model. The “Prob. > F” column indicates the
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 536 [email protected]
significance and non-significance of the factors and its interactions. The Model F-value is
5.44 and Prob.>F value is 0.0009 implies the model is significant and is important. There is
just a 0.01% chance that an F-value could occur due to interference or noise. Values greater
than 0.1000 indicate the model terms are not significant. Table 5 also shows the other
adequacy measures of R2, Adj R
2, and Pred R
2 are nearer to 1 which means the regression
model indicates the goodness of fit between input variables (wear parameters) and output
response (wear rate). The regression equation is obtained from the Design Expert software in
terms of actual factors is used to predict the wear rate with a reasonable accuracy is given in
Eq.6.
Wear Rate (W) = 1083.81845 + 18.52678 * V - 0.07543 * D + 24.54940 * F - 175.53373
* S + 0.05937 * V * D - 9.25000 * V * F - 2.39583 * V * S - 5.43750 * 10-3
* D * F-2.39583
* 10-3
* D * S - 0.31249 * F * S - 29.07366 * V2
- 5.32924 * 10-5
D2
+ 0.30973 * F2
+11.02480 * S2
(6)
Figure 1 (a) Comparison of Predicted vs. Actual values of wear rate
Design-Expert® Softwarewear rate
Color points by value ofwear rate :
1795
328
Actual
Pre
dic
ted
Predicted vs. Actual
0
500
1000
1500
2000
0 500 1000 1500 2000
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 537 [email protected]
Figure 1 (b) Residuals vs. Predicted values of wear rate
Fig.1 (a) shows the predicted vs. actual values of wear rate which are very useful in
validating the experimental data. The values fall well near the straight line which indicates the
experimental data are acceptable. Fig.1 (b) shows the residuals vs. predicted values of wear
rate. There is no specific pattern formed in the plot and the residuals are falling in near the
straight line, which indicates that the errors are normally distributed. Hence the model formed
by RSM can be applied to predict the wear rate within the defined range of input parameters.
4. RESULTS AND DISCUSSION
4.1. Effect of Reinforcements on Micro and Macro Hardness
The results of the micro-hardness (HV) and macro hardness (BHN) tests conducted on the
specimens with different wt % of SiC composites is shown in Table.6. It is very evident as the
wt. % of SiC increases, the micro and macro hardness of the composite increases. The
increase in the hardness shows a linear trend. It is experimentally proven fact that whenever a
hard reinforcement is reinforced with a ductile matrix, the hardness of the matrix material is
enhanced. [29].
Table 6 Details of LM13-SiC composites
Design-Expert® Softwarewear rate
Color points by value ofwear rate :
1795
328
Predicted
Exte
rnally
Stu
dentized R
esid
uals
Residuals vs. Predicted
-6.00
-4.00
-2.00
0.00
2.00
4.00
200 400 600 800 1000 1200 1400 1600
3.83802
-3.83802
0
wt % of SiC 0 3 6 9 12
Micro Hardness (HV) 130 135.2 143 151.4 160
Macro Hardness (HRC) 105 109 115.2 122.4 130
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 538 [email protected]
4.2. Effect of Variables on Wear Rate
The change in wear rate with respect to sliding velocity, sliding distance normal load and
silicon carbide wt% is shown in Fig.2. It is very evident, as the sliding velocity and sliding
distance increases the wear rate of the composite decreases. The wear rate increases with
increase in normal load. This is due to the vertical force which is acting on the pin makes the
pin to rub heavily on the rotating disc results in excessive wear. As the wt% of silicon carbide
increases the wear rate decreases linearly. It can be seen the aluminium alloy undergoes a
severe wear with that of the unreinforced alloy. The interaction effects of the variables are
presented in Fig. (3-5).The same effects can be observed in the interaction plots. From Fig.3
the wear rate which was high at the lower sliding velocity and sliding distance tend to
decrease with increase sliding velocity and sliding distance. This may be due to the fact as the
sliding speed increases the temperature of the rubbing surface increases leads to oxidation
which makes the rubbing surface hard results in low wear of the composites. Also at high
sliding speeds the interaction time between the sliding surfaces decreased which results in low
wear. In addition to this the sliding of pin over long distances causes hardening of the surface
layer composition of the waste debris and reduces wear. The effect of sliding velocity and
normal load on wear rate is shown in Fig.4. The wear rate is less at low level sliding speed
and normal load. The wear rate shows a linear trend with increase in the sliding speed but
whereas during the increase in normal load the wear rate also increases. This is due to the fact
when the load increases the pressure on the sliding surfaces increases which resulted in
increased wear at high loads. Fig.5 shows the interaction effect of sliding speed and silicon
carbide wt% on wear rate. The wear rate is high at low sliding speed and unreinforced
aluminium alloy. As the sliding speed increases the wear rate of the composites decreases.
This is mainly due to increase in temperature which led to softening of the matrix and
composite pin surface. As the silicon carbide content in the aluminium alloy increases the
wear rate of the composite decreases. This is mainly due to the mechanically mixed layer
(MML) formed on the worn surface of the composite which controls the wear properties of
the composites.
Figure 2 Variation of wear rate
Design-Expert® SoftwareFactor Coding: Actualwear rate (mm3/m)
Actual FactorsA: Sliding Velocity = 1.2B: Sliding Distance = 1200C: Normal Load = 30D: Silicon Carbide = 6
-1.000 -0.500 0.000 0.500 1.000
-500
0
500
1000
1500
2000
A
A
B
BC
C
D
D
Perturbation
Deviation from Reference Point (Coded Units)
wear
rate
(m
m3/m
)
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 539 [email protected]
Figure 3 Effect of sliding velocity and sliding distance interaction on wear rate
Figure 4 Effect of sliding velocity and normal load interaction on wear rate
Design-Expert® SoftwareFactor Coding: Actualwear rate (mm3/m)
Design points above predicted valueDesign points below predicted value1795
328
X1 = A: Sliding VelocityX2 = B: Sliding Distance
Actual FactorsC: Normal Load = 30D: Silicon Carbide = 6
400
800
1200
1600
2000
0.4
0.8
1.2
1.6
2
0
500
1000
1500
2000
wear
rate
(m
m3/m
)
A: Sliding Velocity (m/s)B: Sliding Distance (m)
Design-Expert® SoftwareFactor Coding: Actualwear rate (mm3/m)
Design points above predicted valueDesign points below predicted value1795
328
X1 = A: Sliding VelocityX2 = C: Normal Load
Actual FactorsB: Sliding Distance = 1200D: Silicon Carbide = 6
10
20
30
40
50
0.4
0.8
1.2
1.6
2
0
500
1000
1500
2000
wear
rate
(m
m3/m
)
A: Sliding Velocity (m/s)C: Normal Load (N)
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 540 [email protected]
Figure 5 Effect of sliding velocity and silicon carbide (wt %) interaction on wear rate
4.3. Optimization of Wear Parameters
The objective of optimization is to provide the optimum wear parameters which give the
minimum wear rate. This analysis is based on “smaller is better” concept. It means low wear
rate is considered as optimum [30]. The input wear parameters were selected based on the
desirability values. The desirability values close to 1 unit were selected as the most effective
parameters value with respect to wear rate. The ramp function and the desirability bar graphs
are shown in Figs. 6 and 7, respectively. The optimal solution for the wear rate is shown in
Fig. 6. It is seen from Fig. 6 that the optimal values of input parameters are sliding velocity is
1.45938, sliding distance is 1674.11, normal load is 18.5766 and silicon carbide wt% is
7.72246. It is seen from Figs. 6 and 7 that approximately 100% of desirability is achieved for
the output response. The bar graphs show the overall desirability function of the responses.
Desirability varies from 0 to 1 depends on the nearness of the response toward the objective.
The bar graph clearly indicates how well the each variable satisfies the criterion, a value close
to one is considered to be proficient.
Figure 6 Ramp function graph of desirability of wear rate
Design-Expert® SoftwareFactor Coding: Actualwear rate (mm3/m)
Design points above predicted valueDesign points below predicted value1795
328
X1 = A: Sliding VelocityX2 = D: Silicon Carbide
Actual FactorsB: Sliding Distance = 1200C: Normal Load = 30
0
3
6
9
12
0.4
0.8
1.2
1.6
2
0
500
1000
1500
2000
wear ra
te (m
m3/m
)
A: Sliding Velocity (m/s)D: Silicon Carbide (Wt %)
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 541 [email protected]
Figure 7 Desirability bar graph of wear rate
4.4. Validity of the Wear Model and Confirmation of Experiments
The validity of the dry sliding wear model was evaluated by conducting dry sliding wear test
on composites at different values of the experimental factors such as sliding speed (V), sliding
distance (D), normal load (F) and silicon carbide wt% (S). As the equation of response for the
model is derived from quadratic regression, confirmation test must be conducted in order to
confirm their validity. The independent variable selected for the confirmation experiments
must lie within the ranges for which equations were derived. The data of the confirmation
experiments and their comparison with the predicted values for wear rate are listed in Table 6.
From the table it is inferred that the error between experimental and predicted values is within
±5% for the response. All the experimental values of each run are within the 95% prediction
interval. This shows the quadratic model obtained is accurate which confirms the
experimental conclusion.
Table 7 Confirmatory test and comparison of experimental and RSM model
Exp.No.
Sliding wear parameters Response parameter
Sliding
Velocity
(V)
Sliding
Distance
(D)
Normal
Load
(F)
Reinforcement (SiC )
(S)
Wear Rate (W)
X 10-5 mm3/m
Exp. RSM Error
(%)
1 1.5 1675 20 8 359 360.535 1.5
5. CONCLUSIONS
In the present experimental work LM13/SiC metal matrix composites were successfully
fabricated by compocasting process and dry sliding wear test was conducted on pin-on-disc
apparatus and then following conclusions are drawn.
1. The micro and macro hardness of composite increased when compared with base metal
matrix. The hardness of the composite increased with increase of SiC wt%.
Investigation on Wear Behaviour of LM13/SiC Aluminium Metal Matrix Composites by
Response Surface Methodology
http://www.iaeme.com/IJMET/index.asp 542 [email protected]
2. The wear resistance of the composite increased when compared to conventional metal
matrix. The wear resistance of the composite increased with the increase of hardness of the
composite.
3. The wear rate of composites decreased with increase in sliding speed, sliding distance and
reinforcement wt % and decreased with increase in normal load.
4. The wear resistance of developed composites was higher than that of cast metal matrix.
This is due to the formation of MML on the worn surface of the composite which played a
key role played in controlling the wear properties of the composites.
5. The ANOVA indicated that normal load is the most influential factor followed by
reinforcement wt %, sliding distance and load on the wear rate of composites.
6. Optimization by RSM method for a minimum wear rate of 360.535, the optimal value of
wear parameters sliding velocity is 1.45938, sliding distance is 1674.11, normal load is
18.5766 and silicon carbide wt % is 7.72246 respectively.
7. The confirmation experiments showed that the error between experimental and predicted
value of wear rate lies within the range of ±5% for the response.
REFERENCES
[1] Richardson, D.E. Review of power cylinder friction for diesel engines. J Eng Gas
Turbines Power, 122(4), 2000, pp. 506–19.
[2] Alaneme, K. K. and Sanusi, K. O. Microstructural characteristics, mechanical and wear
behaviour of aluminium matrix hybrid composites reinforced with alumina, rice husk ash
and graphite, Eng. Sci. Technol. Int. J, 18, 2015, pp. 416–422.
[3] Lara, R. D., Soltani, N., Bahrami, A., Castañeda, E. G., Sánchez, E. G. and Rodríguez, M.
A. L. H. Tribological characterization of Al7075–graphite composites fabricated by
mechanical alloying and hot extrusion, Mater. Des, 67, 2015, pp. 224–231.
[4] Liu, H., Hu, Y., Zhao, Y. and H. Fujii. Microstructure and mechanical properties of
friction stir welded AC4A + 30 vol.%SiCp composite, Mater. Des, 65, 2015, pp. 395–400.
[5] Shoba, C., Ramanaiah, N. and Nageswara Rao, D. Effect of reinforcement on the cutting
forces while machining metal matrix composites – An experimental approach, Eng. Sci.
Technol. Int. J,. 18, 2015, pp. 658–663.
[6] Kumar, A., Mahapatra, M. M. and Jha, P. K. Fabrication and Characterizations of
Mechanical Properties of Al-4.5%Cu/10TiC Composite by In-situ Method, Journal of
Minerals and Materials Characterization and Engineering, 11, 2012, pp. 1075–1080.
[7] Jayaprakash, R. H. Budan, A. and Hemanth, J. Preparation and Investigation on
Properties of Cryogenically Solidified Nano Metal Matrix Composites, International
Journal of Engineering Research and Applications, 4, 2014, pp. 74-79.
[8] Londhe, V. D., Mhaske, M. S. and Kapgate R. A. Evaluation of Tribological Behaviour of
LM13-Silicon Carbide Composite under Ambient & Elevated Temperature. International
Journal of Scientific Research Engineering & Technology, 3, 2014, pp. 334-340.
[9] Radhika , N. and Raghu, R. Evaluation of Dry Sliding Wear Characteristics of LM 13
Al/B4C Composites. Tribology in Industry, 37, 2015, pp. 20-28.
[10] Das, S. Development of Aluminium Alloy Composites For Engineering Applications.
Trans. Indian Inst. Met, 57, 2004, pp. 325-334
[11] Garcia-Cordovilla, C., Narciso, J. and Louis, E. Abrasive wear resistance of aluminium
alloy/ceramic particulate composites. Wear, 192,1996, pp. 170–7.
[12] Sahin, Y. and Acilar, M. Production and properties of SiCp-reinforced aluminium alloy
composites. Composites, 34 A, 2003, pp. 709–18.
[13] Prashant, S. N., Nagaral, M. and Auradi, V. Preparation and evaluation of mechanical and
wear properties of Al 6061reinforced with graphite and sic particulate metal matrix
composites. International Journal of Mechanical Engineering and Robotics Research, 3,
2012, pp. 106-12.
J. Eric David Praveen, D.S. Robinson Smart, R. Babu and A. Vijin Prabhu
http://www.iaeme.com/IJMET/index.asp 543 [email protected]
[14] Sahin, Y. Tribological behaviour of metal matrix and its composite, Materials and Design,
28, No. 4, 2007, pp. 1348-1352.
[15] Ravikiran, A. and Surappa, M. K. Effect of sliding speed on wear behaviour of A356 Al-
30 Wt. % SiCp MMC, Wear, 206, No. 1-2, 1997, pp. 33-38.
[16] Vettivel, S. C., Selvakumar, N. and Leema, N. Experimental and prediction of sintered
Cu-W composite by using artificial Neural Network. Mater Des, 45, 2013, pp. 323-35
[17] Vettivel, S. C., Selvakumar, N. and Haiter, A.L. Electrical resistivity, wear map and
modelling of extruded tungsten reinforced copper composite. Mater Des , 56, 2014, pp.
977-96
[18] Sentilkumar, C., Ganesan, G. and Karthikeyan, R. Study of electrochemical machining
characteristics of Al/SiCp Composites . Int J Adv manuf Technol, 43, 2009, pp. 256-63
[19] Oraon, B., Majumdar, G. and Ghosh, B. Parametric optimization and prediction of
electroless Ni-B deposition, Mater Des, 28, 2007, pp. 2138-47
[20] Huiping, L., Guoqun, Z., Shanting, N. and Yiguo, L. Technologic parameter optimization
of gas quenching process using response methods. Comput Mater Sci, 38, 2007, pp. 561-
70
[21] Oktem, H., Erzurumlu, T. and Kurtaran, H., Application of response surface methodology
in the optimization of cutting conditions for surface roughness. J mater Process Technol,
170, 2005, pp. 11-16
[22] Correia, D. S., Goncalves, C. V., Cunha, S. S. C. and Ferraresi, V. A. Comparison
between genetic algorithms and response surface methodology in GMAW welding
optimization. J Mater Process Technol, 160, 2005, pp. 70-76
[23] Vettivel, S. C., Selvakumar, N., Narayanasamy, R. and Leema, N. Numerical modelling
prediction of Cu-W nano powder composite in dry sliding wear condition using response
methodology. Mater Des 50, 2013, pp. 977-966
[24] Montgomery, D. G. Design and Analysis of Experiments, Wiley, Hoboken, 2001.
[25] Box, G. E. P., Hunter, W. H. and Hunter, J. S. Statistics for Experiments, Wiley, New
York, 1978.
[26] Khosravi, H., Bakhshi, H. and Salahinejad, E. Effects of compocasting process parameters
on microstructural characteristics and tensile properties of A356−SiCp composites,
Transactions of Nonferrous Metals Society of China, 24, 2014, pp. 2482−2488.
[27] Selvam, J. D. R., Smart, D. S. R. and Dinaharan, I. Microstructure and some mechanical
properties of fly ash particulate reinforced AA6061 aluminium alloy composites prepared
by compocasting, Materials and Design, 49, 2013, pp. 28–34.
[28] Liuy, Z., Zu, G., Luo, H., Liu, Y. and Guangchun Yao, Influence of Mg Addition on
Graphite Particle Distribution in the Al Matrix Composites, Journal of Materials Science
& Technology, 26, 2010, pp. 244-250.
[29] Ramesh, C. S., Pramod, S. and Keshavamurthy R. A study on microstructure and
mechanical properties of Al 6061–TiB2 in-situ composites, Materials Science and
Engineering, A 528, 2011, pp. 4125–4132.
[30] Vettivel, S.C., Selvakumar, Narayanasamy, P. and Leema N. Numerical modelling,
prediction of Cu-W nano powder composite in dry sliding wear condition using response
surface methodology, Materials and Design, 50, 2013, pp. 977-996.
[31] T. Madhusudhan and M. Senthil Kumar. Experimental Study on Wear Behaviour of SiC
Filled Hybrid Composites Using Taguchi Method. International Journal of Mechanical
Engineering and Technology, 8(2), 2017, pp. 271–277.
[32] Syed Khaja Naimuddin, Md Touseef Ahamad, B. Purna C handra Sekhar, P. Narendra
Mohan, Dr. Vidhu Kampurath and Dr. Yousuf Alif, Wear Behaviour of Welded and Non-
Welded Samples of Friction Stir Welding of Similar and Dissimilar Joints of Aluminium
Alloys (Aa6061-Aa6061), (Aa6082-Aa6082) & (Aa6061-Aa6082) Under T6 Condition.
International Journal of Mechanical Engineering and Technology, 7(4), 2016, pp. 289–
299.