Reduction of Casting Defects Using Taguchi Method: Arun Basil Jacob Arunkumar O.N
Reduction of Casting Defects Using Taguchi Method: Arun Basil Jacob Arunkumar O.N
Abstract: This paper aims at the reduction of casting defects of a plate cylinder which is used in an offset
printing machine to transfer ink from the inking unit. Four variables are considered which directly affect the
casting of the cylinder. Taguchi method was selected for the optimization of these selected four parameters.
Taguchi method is based on performing evaluation or experiments to test the sensitivity of a set of response
variables to a set of control parameters (or independent variables) by considering experiments in “orthogonal
array” with an aim to attain the optimum setting of the control parameters. Four control parameters were used
to reduce the defects that occur on the plate cylinder.Moisture Content, green strength, sand particle size and
mould hardness were selected as the control parameters. The analysis showed a reduction in the number of
defects when optimum input factor values were selected.
Keywords: Green strength, mould, orthogonal, taguchi
I. Introduction
Metal casting process begins by creating a mold, which is the negative shape of the part required for
manufacturing. The mold is made from a refractory material, for example, sand. The metal is heated in an oven
until it melts, and the molten metal is poured into the mould cavity. The liquid takes the shape of cavity, which
is the shape of the part. It is cooled until it solidifies. Finally, the solidified metal part is removed from the
mould. A large number of metal components are made by casting. One of the most important metal casting
method is the sand casting method.A casting defect is an irregularity in the metal casting process that is
undesired. Some defects can be tolerated while others can be repaired, otherwise they must be eliminated. They
are broken down into mainly into five categories viz gas porosity, shrinkage defects, mold material
defects, pouring metal defects, and metallurgical defects. The removal of all the casting defects from a casting
process is not possible as there are a number of factors affecting it but the optimization of the control parameters
can lead us to achieve better results. This paper tries to reduce the defects in a casting process by optimizing its
control parameters.
II. Objectives
The paper aims at finding a method and using it which can be applied in industry to reduce the defects
that occur during casting process. The optimization is done with the help of taguchi method and usind
MINITAB software.
these defects. So it preferably necessary to reduce it as much as possible by appropriate analysis of the defects
which includes the root cause analysis so that actual reasons behind occurring the defects can be find out to
make the corrective action. In this paper use six sigma technique and Shainin tool for identify and analysis
casting defect. Shainin tool works on elimination principle. Final result of this work has to be reducing slag and
porosity defect by taking corrective action. Tool should be identifying the sources of variation clearly. [1]
Saurabh Gupta et al focus on minimizing the defects in Al-Mg alloy castings in green sand casting
process by optimizing the casting process parameters. Several process parameters contribute to these casting
defects. Literature review reveals that moisture content, binder percentage and pouring temperature are among
the most influencing parameters which contribute to the casting defects like sand drop, blow holes, scabs, and
pinholes. In this paper these three process parameters are optimized by using the Taguchi’s design of experiment
method. The Taguchi approach is used to capture the effect of signal-to-noise ratios of the experiments based on
the orthogonal array. Robust design factor values were estimated from the signal-to-noise calculations. [2]
Uday A. Dabade et al combined and used the design of experiments and computer assisted casting
simulation techniques to analyze the sand related and method related defects in green sand casting. An attempt
has been made to obtain the optimal settings of the moulding sand and mould related process parameters of
green sand casting process of the selected ductile iron cast component. The green sand related process
parameters considered are, moisture content, green compression strength, and permeability of moulding sand
and mould hardness (in horizontal direction). In first part of this paper Taguchi based L18 orthogonal array was
used for the experimental purpose and analysis was carried out using Minitab software for analysis of variance
(ANOVA) and analysis of mean (AOM) plot. ANOVA results indicate that the selected process parameters
significantly affect the casting defects and rejection percentage. In the second part, shrinkage porosity analysis is
performed using casting simulation technique by introduction of a new gating system designed, solid model
developed for four cavities mould. Number of iterations using casting simulation software was performed for
mould filling and solidification analysis to reduce the level and intensities of shrinkage porosities in cast
component. With new gating and feeding system design reduction in shrinkage porosity (about 15%) and
improvement in yield (about 5%) is observed. [3]
Udhaya Chandran. R. M. mainly focus on to minimize the casting defects such as, sand drop, sand
blow holes, scabs, pinholes. The optimization technique used for this purpose is taguchi method. The parameters
considered are moisture content (%), green strength(g/cm2 ), mould hardness, sand practical size (AFS).The
Taguchi approach is used to capture the effect of signal to noise ratio of the experiments based on the orthogonal
array used due to optimum conditions are found. The outcome of this paper is that the selected process
parameters continuously affect the casting defects in foundry. The improvement expected in reduction of casting
defects is found to be 47.66 %. [4]
B.R. Jadhav et al puts forward in their paper that it is almost impossible to produce defect free castings.
Occurrence of the defect may involve single or multiple causes. These causes can be minimized through
systematic procedure. The paper represents a procedure to analyze and minimize casting defect, cold shut in
automobile cylinder block of grey cast iron Grade FG150. The paper finds that gaiting systems are not always
responsible for the defect occurrence and the defect reduction by controlling alloy composition and pouring
temperature is done. The seven quality control methodology is used to analyze and reduce defects which
includes check sheet, pareto analysis, cause effect diagram, flow chart, scatter diagram, histogram and control
chart. [5]
The plate cylinders are positioned in such a way that it can quickly and accurately locate the printing
plate, by means of front and rear clamps which grip two ends of the plates and provide tension around the
cylinder. Offset or blanket cylinders are manufactured with considerable care and at first sight appear to be an
exclusive item. Many printers engaged is general printing use blankets in the medium range of hardness for all
purpose quite successfully.During the run of most jobs, the blanket will be cleaned periodically, and this should
be carried out in a mechanized manner. As there is almost certainly will be an accumulation of flat form the
paper on blanket of coating form such stoke. This must be removed by first washing the blanket with a swab
soaked in water, and then followed by a through washing with swab soaked in a good quality blanket wash the
procedure must to wash an area and immediately dry it with another swab. This prevents the blanket wash form
evaporating too quickly and leaving behind residue of ink in a thin film. This thin film can quickly develop to
form glaze over whole surface of the blanket which results in a poorly printed sheet, as the image is formed on a
glazed surface instead of the rubber and is not picked up and transferred properly .The periodic cleaning of the
blanket with methyl ethyl ketone will help in removing glaze, which is also caused by the natural oxidation of
the rubber. The impression cylinder carries the sheet through the nip, where the ink is finally transferred under
the pressure to the printing stock. This is the moment of truth where the smallest dots, the finest lines, the
lightest tints and heaviest solid are laid upon the paper. The slightest variation in pressure, packing, inking,
clamping will be recorded at this juncture.
In this paper the casting defects of the plate cylinder is minimized by the taguchi method.
V. Taguchi Method
Taguchi method is based on performing evaluation or experiments to test the sensitivity of a set of
response variables to a set of control parameters (or independent variables) by considering experiments in
orthogonal array with an aim to attain the optimum setting of the control parameters. Orthogonal arrays provide
a best set of well balanced (minimum) experiments. Table 4.1 shows eighteen standard orthogonal arrays along
with the number of columns at different levels for these arrays. An array name indicates the number of rows
and columns it has, and also the number of levels in each of the columns. For example array L 4(23) has four
rows and three “2 level” columns. Similarly the array L18(2137) has 18 rows; one “2 level” column; and seven
“3 level” columns. Thus, there are eight columns in the array L18. The number of rows of an orthogonal array
represents the requisite number of experiments. The number of rows must be at least equal to the degrees of the
freedom associated with the factors i.e. the control variables. In general, the number of degrees of freedom
associated with a factor (control variable) is equal to the number of levels for that factor minus one. For
example, a case study has one factor (A) with “2 levels” (A), and five factors (B, C, D, E, F) each with “3
level”. Table 1 depicts the degrees of freedom calculated for this case. The number of columns of an array
represents the maximum number of factors that can be studied using that array. [7]
Table 1:- Standard orthogonal arrays
There are three forms of signal to noise (S/N) ratio that are of common interest for optimization of
static problems.
This case arises when a specified value is the most desired, meaning that neither a smaller nor a larger
value is desired. From the above cases it is clear that we have to take the smaller the better approach to reduce
the defects.
Steps employed for taguchi method
1) Select the design matrix and perform the experiments
2) Calculation of factor effects
3) Selecting optimum factor levels
4) Developing the additive model for factor effects
5) Analysis of Variance (ANOVA)
6) Interpretation of ANOVA table.
7) Prediction of η under optimum conditions
Sand Particle 50 52 54
size(AFS)
The study is associated with four factors with each at three levels. Table 1 indicates that the best
suitable orthogonal array is L9.Table 3 shows the design matrix for L9 array for four factors A, B, C and D. Next
conduct all the nine experiments and observe the surface defect counts per unit area. The arrangement of the
factors in L9 array in MINITAB software is given in Table 4.
Table 3:- L9 array arrangement
The arrangement of the factors in arrays is represented in the table below. The software used for this
purpose is the MINITAB statistical tool.
For the given arrangement of the percentage rejection of parts due to defects were taken into account.
Here we have to minimize the number of rejections due to defects, so “smaller the better” approach is used here.
The observed rejections are shown in table 5.
The S/N ratio found using the smaller the best approach is shown in table 6.
Table 6:- S/N ratio
The effect of a factor level is defined as the deviation it causes from the overall mean. Hence as a first
step, calculate the overall mean value of η for the experimental region defined by the factor levels.The overall
mean 𝑚 is given by
1 9 1
𝑚= 𝑖=1 𝜂𝑖 = (𝜂1 + 𝜂2 + ⋯ + 𝜂9 )(1)
9 9
From the equation given𝑚 = -14.416
The effect of the moisture content at level A (at experiments 1, 2 and 3) is calculated as the difference
1
of the average S/N ratio for these experiments (m ) and the overall mean. The same is given as
A1
The effect of moisture content at :
1
LevelA = 𝑚𝐴1 − 𝑚 = 3 𝜂1 + 𝜂2 + 𝜂3 − 𝑚(2)
1
1
Level A = 𝑚𝐴2 − 𝑚 = 3 𝜂3 + 𝜂4 + 𝜂5 − 𝑚(3)
2
1
Level A = 𝑚𝐴3 − 𝑚 = 3 𝜂6 + 𝜂7 + 𝜂8 − 𝑚(4)
3
Using the S/N ratio data available in Table 6the average of each level of the four factors is calculated
and listed in Table 7. These average values are shown in Figure 3. They are separate effect of each factor and
are commonly called main effects.
Once the average η for different factor levels are found the S/N vs. factor level was plotted (Table 8).
Our goal in this experiment is to minimize the surface defect counts to improve the quality of the
silicon wafers produced through the chemical vapor deposition process. Since –log depicts a monotonic
decreasing function [equation 1] we should maximize η. Hence the optimum level for a factor is the level that
gives the highest value of η in the experimental region. From Figure 3and the Table 7, it is observed that the
optimum settings for moisture content, green strength, sand particle size and mold hardness are A3 B1 C2 D1.
Hence we can conclude that the settings A3 B1 C2 D1 can give the highest η or the lowest surface defect count.
Different factors affect the surface defects formation to a different degree. The relative magnitude of the
factor effects are listed in Table 7. A better feel for the relative effect of the different factors is obtained by the
decomposition of variance, which is commonly called as analysis of variance (ANOVA). This is obtained first by
computing the sum of squares
Total sum of squares = 9𝑖=1(𝜂𝑖 − 𝑚)2 = 45.24𝑑𝐵2 (5)
Sum of squares due to factor A
2
= [(𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑥𝑝𝑒𝑟𝑖𝑚𝑒𝑛𝑡𝑠 𝑎𝑡 𝐿𝑒𝑣𝑒𝑙 𝐴1) × 𝑚𝐴1 − 𝑚 ] + [(𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑥𝑝𝑒𝑟𝑖𝑚𝑒𝑛𝑡𝑠 𝑎𝑡 𝐿𝑒𝑣𝑒𝑙 𝐴2) ×
2 2
𝑚𝐴2 − 𝑚 ] + [(𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑥𝑝𝑒𝑟𝑖𝑚𝑒𝑛𝑡𝑠 𝑎𝑡 𝐿𝑒𝑣𝑒𝑙 𝐴3) × 𝑚𝐴3 − 𝑚 ](6)
From the equation 6 the sum of squares of each factor is calculated and found out to be:
1) Moisture content (%) = 6.121
2) Green strength (g/𝑐𝑚2 ) = 0.331
3) Sand particle size (AFS) = 35.026
4) Mould hardness = 3.075
The ANOVA table is tabulated and the F values are found out
Table 8:- Anova table
A. Moisture content
2 6.121 3.121 17.142
Total 8 45.245
In the present case-study, the degrees of freedom for the error will be zero. Hence an approximate
estimate of the error sum of squares is obtained by pooling the sum of squares corresponding to the factors
having the lowest mean square. As a rule of thumb, the sum of squares corresponding to the bottom half of the
factors (as defined by lower mean square) are used to estimate the error sum of squares. In the present example,
the factors D and B are used to estimate the error sum of squares. Together they account for four degrees of
freedom and their sum of squares is 4.095.
The major inferences from the ANOVA table are given in this section. Referring to the sum of squares
in Table 8, the factor C makes the largest contribution to the total sum of squares [(35.026/45.245) x 100 =
77%]. The factor A makes the next largest contribution (13.53 %) to the total sum of squares, whereas the
factors B and D together make only 9.657 % contribution. The larger the contribution of a particular factor to
the total sum of squares, the larger the ability is of that factor to influence η. Moreover, the larger the F-value,
the larger will be the factor effect in comparison to the error mean square or the error variance.
In the present case, the optimum condition or the optimum level of factors is A3 B1 C2 D1. The value of η under
the optimum condition is predicted using the additive model as
𝜂𝑜𝑝𝑡 = 𝑚 + 𝑚𝐶2 − 𝑚 + 𝑚𝐴3 − 𝑚 = −10.633(7)
Since the sum of squares due to the factors B and D are small as well as used to estimate the error
variance, these terms are not included in equation 7. The mean square count at the optimum condition is
calculated as
−𝜂 𝑜𝑝𝑡
y = 10 10 = 101.0633 =11.569(8)
The corresponding root-mean square rejectioncount is
𝑦 = 3.40 %(9)
From the results it can be seen that by selecting the found out factor levels the rejection percentage can
be kept at 3.40%.
VII. Conclusions
Casting defects were studied to find out the factors causing it. The reduction of defects was done by
optimizing the control variables or factors affecting the output. Taguchi method was the best available method
to optimize the factors and together with the help of analysis of variance the optimum control factors were found
out. Sand particle size and moisture content account for the main reason for controlling defects. The percentage
rejection of the plate cylinders can be reduced to 3.40 % which is a significant improvement in this case.
Taguchi method is thus a cost effective method which can be used in industry to reduce the wastage of cast parts
with the existing industry resources.
References
[1]. R. B. Heddure and M. T. Telsang“Casting defect reduction using shainin tool in CI Foundry – a case study”, International Journal of
Mechanical And Production Engineering.(2014) ISSN: 2320-2092, Volume- 2, Issue- 6.
[2]. Saurabh Gupta and N.M. Suri “ Optimization of process parameters to minimize the defects in al-alloy casting ” International
Journal For Technological Research In Engineering (2015), ISSN : 2347 – 4718, Volume 2, Issue 9.
[3]. Uday A. Dabade and Rahul C. Bhedasgaonkar “Casting Defect Analysis using Design of Experiments (DoE) and Computer Aided
Casting Simulation Technique” Forty Sixth CIRP Conference on Manufacturing Systems (2013) Procedia, Elsevier CIRP 7, page
616 – 621
[4]. Udhaya Chandran. R. M. “Optimization of process parameters to minimize the casting defects” International Journal of Advances in
Engineering Science and Technology (2013), ISSN: 2319-1120.
[5]. B.R. Jadhav and Santosh J Jadhav “INVESTIGATION AND ANALYSIS OF COLD SHUT CASTING DEFECT AND DEFECT
REDUCTION BY USING 7 QUALITY CONTROL TOOLS”, International Journal of Advanced Engineering Resea rch and
Studies (2012), ISSN2249–8974
[6]. www.conquestgraphics.com/services/printing-offset
[7]. Prof Amitava De“Design for Reliability and Quality” Indian Institute of Technology (IIT), Module 5,Lecture4.