You are on page 1of 7

Bonfring International Journal of Advances in Image Processing, Vol. 6, No.

4, November 2016 28

Application of Taguchi Method for Optimizing the


Parameters of Thermal Properties of Al/SiC-Gr
Hybrid Composites
Yanamandala Raghuram Chowdary and C. Yuvaraj

Abstract--- Using the Taguchi method, this study presents determining optimal process parameters of various
a systematic optimization approach for thermal properties of manufacturing processes. Der Ho and Mao Sheng (2004) used
Al/SiC-Gr composites which were synthesized by using Taguchi method to develop a robust design for the magnesium
metallurgy technique. The thermal conductivity, diffusivity and alloy die casting process.
specific heat capacity measurements were performed on a
The present investigation is on the optimization of
thermal physical testing instrument using a laser flash method. parameters such as wt. % SiC, wt. % of Gr and heat treatment
The controllable factors such as wt. % of SiC, wt.% of Gr and duration for maximum thermal properties using Taguchi
heat treatment duration have been optimized. The percentage
method.
contribution of each controllable factor was also determined.
The heat treatment is the most influential factor and its value
II. EXPERIMENTAL STUDIES
of percentage contribution is up to 99.57%, 99.46% and
97.74% for thermal conductivity, thermal diffusivity and The matrix used in the present study is aluminium-alloy
specific heat respectively. (6061), vowing to its properties like fluidity, high fatigue
strength, corrosion resistance, better mechanical properties
Keywords--- Hybrid Composites, Thermal Conductivity, and are heat treatable alloys. The reinforcements used were
Design of Expert, ANOVA. short SiCs & Graphite particles which is easily available &
econographitel. The properties are clearly mentioned in the
I. INTRODUCTION literature.

S ELECTION of an adequate method of composite


production as well as the type, size and distribution of
reinforcement has a direct influence on the composites
Table 1: Chemical Composition of Aluminium Alloy 6061
Al6061 alloy
Mg Si Fe Cu Ti Cr Zn Mn Be V Al
structure, which in turn determinates the properties of the 0.92 0.76 0.28 0.22 0.10 0.07 0.06 0.04 0.003 0.01 Bal.
finished product [1-3]. Among the industrial methods of The microstructure of any material is a complex function
aluminum metal matrix composites (MMCs) production, as of the casting process, subsequent cooling rates, heat
much as a 10% share belongs to the mechanical stirring treatments and aging processes. Therefore composites
method [4]. In other hand continuing miniaturization of fabrication is one of the most challenging and difficult tasks.
electronic devices has led to a significant increase in power Compo casting technique is used to prepare Al6061 MMCs in
density, resulting in an exponential growth of the heat flux on the present study.
computer chips in the current decade. In order to dissipate the Three ageing conditions were selected for thermal testing.
heat generated in electronic packages effectively, suitable The heat treatment used to achieve these conditions was a
materials must be selected as heat spreaders and heat sinks. solution treatment of 10 hours at a temperature 532 C
Overviews and selection criteria of this field are given by S.
followed by water quenching on hot water at 80 C aged at
K. Choudary and other authors [56].
room temperature for 12 hrs followed by ageing at 175 C for
Many process factors affect the effectiveness of the 0,2,4,6 and 8 hours.
mechanical process. The main purpose of this article is to
For characterization of thermal properties of the Al hybrid
establish the significant factors that influence the thermal
composites both steady state and transient methods can be
properties Al and its composites. This can be determined
used. Under steady state condition only one parameter i.e.
through a series of experiments. However, such experiments
thermal conductivity k (W/mK) is sufficient and is measured
will be expensive and time-consuming. Design of experiment
directly. For transient methods the thermal diffusivity of
(DOE) techniques like the Taguchi method. Can optimize
hybrid composites is obtained by the laser-flash method.
process parameters with minimum experimental runs.
Enormous amount of research has been conducted for

Yanamandala Raghuram Chowdary, Dept. of Aeronautical Engg,


Dayananda Sagar College of Engg, Bangalore, Karnataka, India.
C. Yuvaraj, Dept. of Mechanical Engg, Madanapalle Institute of
Technology & Science, A.P, India.
DOI:10.9756/BIJAIP.10479

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 29

III. RESULTS AND DISCUSSION

Fig. 1: A Typical Graph of Measured Values of Thermal Conductivity, Specific Heat and
Thermal Diffusivity of Al Hybrid Composites
The thermal diffusivity of Al hybrid composites was five levels on the thermal properties of Al hybrid composites
measured in longitudinal and transverse direction of samples. was studied using Taguchis method. The used level setting
The measured thermal conductivity, specific heat and thermal values of the main factor (A-C) and the OA 25 matrix
diffusivity are presented in Fig.1. Thermal conductivity employed to assign the considered factor are shown in the
initially increases then it decreases with respect to Table 2.
temperature. The specific heat linearly increases with respect
Table 2: Selected Fermentation Factors and their Assigned
to temperature, this is due to increasing Cp of constituents with
Levels
increasing temperature and higher values of Cp for both
graphite and SiC than of the aluminium. Symbol Factors Unit Levels
1 2 3 4 5
Taguchi Design A SiC Wt.% 0 2.5 5 7.5 10
Taguchi's OA analysis is used to produce the best B Gr Wt.% 0 2.5 5 7.5 10
parameters for the optimum design process, with the least C Heat treatment hr 0 2 4 6 8
number of experiments. In this work, the effect of three The thermal properties experiments were conducted as per
important factors including: wt. % of SiC, wt. % of graphite the parameters and level settings as given in the L25 as shown
and heat treatment duration, were selected and each factor at in Fig. 3.

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 30

Table 3: L25(35)OA which Parameters Levels and Responses of Al Hybrid Composites


Trail Wt. of SiC Wt. of Gr Heat treatment duration, hr Thermal conductivity Specific Heat Thermal diffusivity, mm2/s
No. W/mK J/gK
01 0 0 0 269.87 2.63 57.51
02 0 2.5 2 331.56 2.65 90.55
03 0 5 4 390.33 3.17 99.61
04 0 7.5 6 448.44 3.69 112.98
05 0 10 8 412.66 3.43 102.02
06 2.5 0 2 320.22 2.64 82.11
07 2.5 2.5 4 392.98 3.15 102.44
08 2.5 5 6 449.30 3.66 113.19
09 2.5 7.5 8 414.71 3.40 102.06
10 2.5 10 0 258.44 2.11 64.97
11 5 0 4 401.47 3.13 107.55
12 5 2.5 6 450.36 3.64 112.83
13 5 5 8 416.36 3.38 103.03
14 5 7.5 0 259.48 2.09 64.88
15 5 10 2 336.42 2.60 88.69
16 7.5 0 6 459.06 3.62 118.79
17 7.5 2.5 8 417.26 3.36 102.55
18 7.5 5 0 260.59 2.07 65.50
19 7.5 7.5 2 338.21 2.57 88.62
20 7.5 10 4 395.59 3.08 100.62
21 10 0 8 419.98 3.37 107.34
22 10 2.5 0 259.77 2.08 65.11
23 10 5 2 337.56 2.58 89.10
24 10 7.5 4 395.40 3.09 100.24
25 10 10 6 453.70 3.60 114.01
The predicted values of the main and interaction effects are indicates significant at the rate of 95% confidence level.
shown in Fig. 2. These obtained values are shown in random According to the reduced models which include only the terms
order of importance in the form of Pareto charts. This helps in which were found as significant at 5%, the relationship
testing the statistical importance of the effect of different between the responses (thermal conductivity, specific heat and
parameters. The probability plots generally used to verify the thermal diffusivity) and parameters were obtained with coded
importance, the significance of the parameters and their variables as follows.
interactions. Each of the parameters is estimated effect in
The Table 4 shows response table for mean of mean values
random order of magnitude. The length of each coloured bar is of mechanical properties and their ranks. The shaded cells are
proportional to the standard effect divided by its standard
optimum level for each factor and their respective properties.
error. The each bars which extending beyond five percentages The lost row shows the preference rank for each property.
Table 4: Response Table for Mean of Mean of Thermal Properties and their Ranks
Level Thermal conductivity Specific heat Thermal diffusivity
SiC G HT SiC G HT SiC G HT
1 370.6 374.1 361.6 3.114 3.078 2.195 92.53 94.66 63.59
2 367.1 370.4 332.8 2.993 2.978 2.610 92.96 94.69 87.81
3 372.8 370.8 395.2 2.967 2.974 3.125 95.39 94.09 102.09
4 374.1 371.2 452.2 2.941 2.966 3.643 95.22 93.75 114.36
5 373.3 371.4 416.2 2.945 2.964 3.388 95.16 94.06 103.40
7.0 3.7 190.5 0.173 0.113 1.448 2.86 0.94 50.77
Rank 02 03 01 02 03 01 02 03 01
Fig. 2 shows the combined effect of wt.% of SiC and wt.% heat treatment. The colour band shows the thermal properties
of graphite on thermal properties of Al/Al2O3 MMCs . The decreases after this line. This is due to agglorimation of
maximum thermal properties were observed on particular line particles.
between 8 wt% of SiC and 8 wt. % of graphite at as cast and

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 31

Contour Plot of Thermal conducti vs wt.% of SiC, Wt.% of Graphite


10
Thermal
conductiv ity ,
w/mk
8 < 300
300 350
350 400
400 450
> 450

wt.% of SiC
6

0
0 2 4 6 8 10
Wt.% of Graphite

Contour Plot of Specific Heat, j/gk vs wt.% of SiC, Wt.% of Graphite


10
Specific
Heat, j/gk
< 2.1
2.1 2.4
8
2.4 2.7
2.7 3.0
3.0 3.3
3.3 3.6
wt.% of SiC

6
> 3.6

0
0 2 4 6 8 10
Wt.% of Graphite

Contour Plot of Thermal diffusit vs wt.% of SiC, Wt.% of Graphite


10
Thermal
diffusitiv ity ,
mm2/s
< 60
8
60 70
70 80
80 90
90 100
wt.% of SiC

6
100 110
> 110

0
0 2 4 6 8 10
Wt.% of Graphite

Fig. 2: Contour Plots or Two Factors (wt.% of SiC and wt.% of graphite) Effect on Mean Values of Thermal
Conductivity, Specific Heat and Thermal Diffusivity Responses of Al Hybrid Composites

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 32

Contour Plot of Thermal conducti vs wt.% of SiC, Heat treatment,


10
Thermal
conductivity,
w/mk
< 300
8
300 350
350 400
400 450
> 450

wt.% of SiC
6

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Contour Plot of Specific Heat, j/gk vs wt.% of SiC, Heat treatment, hr


10
Specific
Heat, j/gk
< 2.1
2.1 2.4
8
2.4 2.7
2.7 3.0
3.0 3.3
3.3 3.6
wt.% of SiC

6
> 3.6

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Contour Plot of Thermal diffusit vs wt.% of SiC, Heat treatment,


10
Thermal
diffusitivity,
mm2/s
< 60
8
60 70
70 80
80 90
90 100
wt.% of SiC

6
100 110
> 110

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Fig. 3: Contour Plots or Two Factors (wt.% of SiC and Heat Treatment) Effect on Mean Values of
Thermal Conductivity, Specific Heat and Thermal Diffusivity Responses of Al Hybrid Composites

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 33

Contour Plot of Thermal conducti vs Wt.% of Graphite, Heat treatment,


10
Thermal
conductiv ity ,
w/mk
< 300
8
300 350
350 400

Wt.% of Graphite
400 450
> 450
6

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Contour Plot of Specific Heat, j vs Wt.% of Graphite, Heat treatment,


10
Specific
Heat, j/gk
< 2.1
2.1 2.4
8
2.4 2.7
2.7 3.0
Wt.% of Graphite

3.0 3.3
3.3 3.6
6
> 3.6

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Contour Plot of Thermal diffusit vs Wt.% of Graphite, Heat treatment,


10
Thermal
diffusitiv ity ,
mm2/s
< 60
8
60 70
70 80
Wt.% of Graphite

80 90
90 100
6
100 110
> 110

0
0 1 2 3 4 5 6 7 8
Heat treatment, hr

Fig. 4: Contour Plots or Two Factors (wt.% of Gr and wt.% of graphite) Effect on Mean Values of Thermal
Conductivity, Specific Heat and Thermal Diffusivity Responses of Al hybrid Composites

Analysis of variance (ANOVA) method is used to predict obtained or analyze the significance of factors. The
the parameters and experimentally found statistically significance of each factor judged using F or P obtained by
significant. MINITAB was a powerful statistical tool to ANOVA technique.

ISSN 2277-503X | 2016 Bonfring


Bonfring International Journal of Advances in Image Processing, Vol. 6, No. 4, November 2016 34

Table 5 Shows the ANOVA of thermal properties a) diffusivity respectively. In all cases only heat treatment is the
thermal conductivity, b) specific heat and c) thermal highest significant factor effect on thermal performance.
Table 5: Analysis of Variance of Thermal Properties
a) Thermal conductivity
Source DF Seq SS Adj SS M SS F P%
SiC content, wt.% 04 159.0 159.0 29.8 1.68 0.14%
Gr content wt.% 04 42.9 42.9 10.7 0.45 0.03%
Heat treatment, h 04 113172.9 113172.9 28293.2 1197 99.57%
Error 12 283.6 283.6 23.6 0.25%
Total 24 113658.5
b) Specific heat
SiC content, wt.% 04 0.10150 0.10150 0.02538 2.90 1.42%
Gr content wt.% 04 0.04648 0.04648 0.01162 1.33 0.65%
Heat treatment, h 04 6.89966 6.89966 1.72492 197.45 96.46%
Error 12 0.10483 0.10483 0.00874 1.46%
Total 24 7.15248
c) Thermal diffusivity
SiC content, wt.% 04 38.48 38.48 9.62 0.86 0.5%
Gr content wt.% 04 3.36 3.36 0.84 0.07 0.04%
Heat treatment, h 04 7654.42 7654.42 1913.60 170.25 97.74%
Error 12 134.88 134.42 11.24 1.72%
Total 24 7831.13

IV. CONCLUSION
The thermal conductivity, diffusivity and specific heat
capacity measurements were performed on a thermal physical
testing instrument using a laser flash method. The
controllable factors such as wt. % of SiC, wt.% of Gr and heat
treatment duration have been optimized. The percentage
contribution of each controllable factor was also determined.
The heat treatment is the most influential factor and its value
of percentage contribution is up to 99.57%, 99.46% and
97.74% for thermal conductivity, thermal diffusivity and
specific heat respectively.

REFERENCES
[1] S.K. Chaudhury, A.K. Singh, C.S. Sivaramakrishnan and S.C. Panigrahi,
Effect of processing parameters on physical properties of spray formed
and stir cast Al2MgTiO2 composites, Materials Science and
Engineering:A, Vol. 393, No. 1, Pp. 196203, 2005.
[2] M. Ferry and P.R. Munroe, Recrystallization kinetics and final grain
size in a cold rolled particulate reinforced Al-based MMC, Composites
Part A: Applied Science and Manufacturing, Vol. 35, No.9,
Pp. 10171025, 2004.
[3] Y.R.R. Chowdary, C. Yuvaraj, K. Prahlada Rao and B. Durgaprasad,
Neural Network for Prediction of EDM of Al/Sic-Graphite Particulate
Reinforced Hybrid Composites, International Journal of Emerging
Technology and Advanced Engineering, Vol. 2, Pp.730-739, 2012.
[4] S.K. Chaudhury and D. Apelian, Effects of rapid heating on
solutionizing characteristics of AlSiMg alloys using a fluidized bed,
Metallurgical and Materials Transactions, Vol. 37, No. 3, Pp. 763778,
2006.
[5] N.P. Hung, L.J. Yang and K.W. Leong, Electrical discharge machining
of cast metal matrix composites, Journal of Materials processing
Technology, Vol. 44, No. 3, Pp. 229-236, 1994.
[6] C. Fenggou and Y. Dayong, The study of high efficiency and
intelligent optimization system in EDM sinking process, International
Journal of Materials Processing Technology, Vol. 149, Pp. 83-87, 2004.

ISSN 2277-503X | 2016 Bonfring