INTRODUCTION
Final finishing operations in manufacturing of precise parts are always of
concern owing to their most critical, labor intensive and least controllable
nature. In era of nanotechnology, a deterministic high precision finishing method
is of utmost importance and is prime need of present manufacturing scenario.
The need of high precision in manufacturing was felt by manufacturers worldwide
to improve interchangeability of components, improve quality control and longer
wear/fatigue life. We require extremely smooth surface at the mating faces of
moving parts having relative motion between them. These smooth surfaces help
to reduce friction it is not the only application related to engineering field
(Mckewon, 1987). Polishing process is one of the main
secondary manufacturing processes and its purpose is to reduce the surface roughness
to a desired amount. Industrial applications of the polishing process span from
large products such as automobile, aerospace, to small parts such as optics
(Xi and Zhou, 2005). Electrophoresis is the motion of
charged particles in suspension under the influence of an electric field. Deposition
is the coagulation of particles to dnese mass, when an electromagnetic field
is applied, the abrasives are attracted to an anode and deposit on it. We call
this phenomenon “electrophoretic deposition (EPD)” (Ikeno
et al., 1994). Electrophoretic deposition (EPD) is a powerful method
for the production of coatings and both thin and thick films. Oxide particles
usually have some electrical charge in water. In EPD, charged particles migrate
to an electrode of opposite charge under the influence of a dc electrical field.
The suspension must have a high stability which can be achieved electrostatically
or by adding charged polymers or surfactants (Ikeno et
al., 1991).
Though the electrophoresis is well known phenomenon for the other processes
but its application to achieve for the mirror finish surfaces has been initiated
in the early nineties. Till date, very few authors have conducted experimental
research for producing optically smooth surfaces and ultrafine finish surfaces
using the electrophoretic deposition principle. There is considerably insufficient
literature available in the area of polishing of stainless steel using electrophoresis
deposition technique. Moreover, the effect of process conditions on the polishing
mechanism is not fully understood. Also the correlation between process parameters
and the surface characteristics has hardly been reported. EPD is a combination
of two processes: electrophoresis and deposition.
There are only few articles that can be found from prior literature review
utilizing electrophoretic deposition assisted polishing process. Takahata
et al. (1996) used this technique to polish 3dimensional micro components
in MEMS sing 100 microns alumina abrasives. Ikeno et
al. (1991) introduced chipping free dicing technology for silicon wafer,
lime glass, lithiumniobate and MnZn ferrite applying electrophoretic deposition
of ultra fine abrasive. They developed highly homogeneous palettes for electrophoretic
deposition of ultrafine abrasives. Tani et al. (1998)
employed infeed grinding of silicon wafers applying electrophoretic deposition
of ultrafine abrasives for grinding wheels which produce a mirror surface finish
(R_{t}) of 23 nm. Yan et al. (2007) introduced
micro energy EDM followed by EPD polishing of SKD 61 Die steel. They compared
polished surface generated by micro energy EDM and EPD polishing. A better surface
qualities within short period of time was observed using EPD polishing. Tusi
et al. (2007) studied the electrophoretic deposition of SiC particles
to polish stainless steel. They found improvement in surface finish (R_{a})
from 0.50.02 μm in 8 min with mirror like surface.
MATERIALS AND METHODS
Design of experiments: Design of Experiment (DoE) is an efficient experiment
planning process that allows the data obtained to be analyzed, valid conclusions
to be drawn and objectives to be set. DoE is used to determine the appropriate
number of tests and the experimental conditions necessary to obtain desired
goal of analyzing which factors of the process influence the response variable.
The most common design consist of running the test with all possible combinations
of variables at each of two levels, thereby obtaining most information required
of multilevel experiment. The experiments were planned to find optimum polishing
performance of EPDAP on SS316L. These are planned according to full factorial
design array (2^{k} where k = 4). For this, four parameters each of
them with tow levels are selected. Hence, the total of 2^{4} experiments,
total no of experiments is (Montgomery, 1997).
Selection of process parameters and their levels for optimizing results:
The process parameters selected for this experimental investigation aregrit
size, axial load, rotational speed and the polishing time. The levels of process
parameters were selected from past literature and their values are shown in
the Table 1. Table 1 shows variable parameter
and their levels and Table 2 shows fixed parameter of EPDAP.
Table 1: 
Control parameters and their levels 

Table 2: 
Fixed parameters used in EPDAP 

Table 3: 
Standard run order for full factorial2^{4} 

The standard run order of full factorial method 2^{k} (where k = 4,
i.e., number of parameters) is shown in Table 3.
Work material: The electrophoretic deposition assisted polishing process
is used to polish metals using electrolyte solution containing distilled water,
NaOH and abrasive particles. The best results of polishing are obtained with
metals with fine grain boundaries that are free of non metallic inclusions and
seams. The metals having a high content of silicon, lead or sulphur are usually
troublesome. Stainless steels are more frequently polished alloys. The type
SS316L steel was used for experiment purpose (Table 4). This
type of steel widely used in applications requiring corrosion resistance superior
to type 304.
Tooling: Polishing tool for the experiment is fabricated from copper
material. One end of the tool is held fixed on the collet of CNC milling machine
spindle. On the other end, a non woven fabric is pasted using electrically conductive
glue which is used as polishing head. A nylon cap is placed on the collet end
of the tool which acts as a barrier for preventing current flow into machine
as shown in Fig. 1.
Experimental set up: The EPD polishing system was established by using
CNC Milling Machine. Figure 2 shows the experimental set up
designed for process under investigation. It has an integral fixture which facilitates
quick change of work piece. The electrolyte tank made of acrylic sheets as the
material is chemically inert and does not react with any of the possible electrolyte
solution which may be used.

Fig. 1: 
Photograph of EPDAP polishing tool 
Table 4: 
Chemical analysis of SS316L 

The anode is made of copper. On this anode, non woven polyester fibers are
pasted with electrically conductive glue.
Experimental procedure: Table 5 shows the experimental
parameters combination used for conducting the experiments and the values of
the response variables obtained for surface roughness and MRR. Figure
3 presents a flow chart that deficit the procedure followed during experiments.
The first step is to remove the surface contaminants from the test specimens
by degreasing and cleaning. After cleaning, measurements of all specimens were
made to obtain surface roughness value using surface roughness tester.

Fig. 2: 
Experimental set up, (1) Machine table, (2) Clamp (3) Electronic
balance (4) Base (5) Electrolyte tank (6) Electrolyte (7) Polishing tool
(8) Voltage supply and (9) Work piece 
Table 5: 
Response values for surface roughness and MRR along with
the experimental parameter combinations 


Fig. 3: 
Flow chart of the EPDAP process 
The surface roughness R_{a}, R_{z}, R_{q} was measured
after EPDA polishing. Initially the electrolyte is prepared with the suitable
composition described previously. It is continuously stirred using a magnetic
stirrer and is ready when pH reaches to 9.
The specimen I placed under the polishing tool and an auxiliary copper electrode
is placed near the inner wall of the tank, so that sufficient SiC particles
can be deposited on the polishing tool during polishing. Therefore, the end
of the polishing tool retains sufficient SiC particles during the polishing
of the specimen. The surface roughness of the ground surfaces were measured
by stylus based surface roughness tester MITUTOYO SJ301, Japan. With a resolution
of 0.001 μm. and 3D surface roughness by using Dektak 150 Surface profiler
(Veeco) which is available in NCL Pune. The surface roughness parameters considered
for the pre finished work piece are R_{a}, R_{z} and R_{q}.
A surface roughness was measured for the entire polished specimen at six locations
and averaged them to obtain R_{a}, R_{z} and R_{q} values
for the specimen.
RESULTS AND DISCUSSION
After carrying out the experiments it is essential now to analyze the data
using statistical methods in order to obtain the meaningful conclusion. The
experimental data collected during the experiments were analyzed using statistical
software. Initial part of this chapter discusses the results of experiment.
The effects of four process parameters on the surface roughness and MRR using
ANOVA and main effects plot as been discussed.
Quantitative analysis: Figure 4ab
shows 3D surface roughness R_{a} of ground surface and Fig.
4c shows microscope photograph of ground surface at 100X. it is observed
that EPDA polished surface shows lower surface roughness R_{a} = 45.57
nm when the parameter were: abrasive grit size 2000, axial load increases to
12 N, rotational speed increases to 1500 rpm and polishing time of 10 min. was
used.
The polishing time dose not significantly influence the surface roughness R_{a}.
Figure 4ab shows 3D surface roughness
R_{a} of after EPDAP, Fig. 4c shows microscope photograph
after EPDAP at 100X (Grit size = 2000, L = 12 N, S = 1500 rpm, t = 10 min).
the higher surface roughness R_{a} = 98.68 nm, is obtained when the
parameter were set as; grit size 1500, axial load increases to 20 N, rotational
speed increases to 800 rpm and polishing time 10 min.
Figure 5ab shows 3D surface roughness
R_{a} of after EPDAP and Fig. 5c shows microscope
photograph after EPDAP at 100X (grit size = 1500, L = 20 N, S = 800 rpm, t =
10 min).
Qualitative analysis: Surface reflectivity can be quantitatively assessed.
However, quantitative measurement of surface reflectivity could not be obtained
due to lack of facility. But the surface reflection can be observed qualitatively
using reflection of letters through visual observation.
After polishing of the work piece it is seen that there is a significant increase
in the reflection of letters. This could be due to removed surface irregularities
and improved surface finish during electrophoretic deposition assisted polishing
(Fig. 6).
ANOVA for surface roughness: The Table 6 Shows ANOVA
summary table of arithmetic average surface roughness R_{a}.
Regression analysis for R_{a}: The equation 1 describes the
relationship between response arithmetic average surface roughness R_{a}
and predictor variables grit size, axial load (L), rotational speed (S) and
polishing time (t).
The regression equation is:

Fig. 4(ab): 
(ab) 3D surface roughness R_{a} of ground surface
and (c) Electron microscope photograph at 100X of
ground surface 

Fig. 5(ab): 
(ab) 3D surface roughness R_{a} after EPDAP (c)
Electron Microscope photograph at 100X of EPDAP surface 

Fig. 6(ab): 
Reflection photographs of work piece (a) Before and (b) After
electrophoretic deposition assisted (EPDA) polishing 

Fig. 7: 
Normal probability plot of the residuals for response Ra 
The regression coefficient Table 7 shows the linear fit coefficients
for arithmetic average surface roughness R_{a} and the constant. It
is observed that predictor’s
constant, grit size, axial load and rotational speed are more significant at
95% CI as it has pvalue less than 0.005. The other predictor polishing time
is not statistically significant as it has pvalue more than 0.005. Correlation
coefficient of 0.887 implies that the model has good fit with the experimental
results.
It is observed from ANOVA table model comparison of arithmetic average surface
roughness R_{a} shows that the model has well linear fit with the predicted
values as the pvalue is less than 0.05. It is seen from the ANOVA that the
parameters grit size. Axial load and the polishing time have statistically significant
effects on the response variable surface roughness. Further the model fir was
checked using normal probability plot for the residuals of the response variable.
Table 7: 
Regression coefficient table for surface roughness 

S: 0.00574412, R^{2}: 88.7% and R^{2} adjusted:
84.6%, ^{*}Significant at 95% CI 
Table 8: 
Constant and coefficient for material removal rate (MRR) 

S: 0.0687485 R^{2}: 97.0% and R^{2} adjusted:
95.9% CI 
Normal probability plot of the residuals for response R_{a} (Fig.
7) shows that the errors are normally distributed as the points in the plot
will roughly form a straight line.
Regression analysis for MRR: The Eq. 2 describes the
relationship between response Material Removal Rate (MRR) and predictor variables
grit size, axial load (L), rotational speed (S) and polishing time (t).
The regression equation is:
The regression coefficient Table 8 shows the linear fit coefficients
for Material Removal Rate (MRR) and the constant. It is observed that predictor’s
constant, grit size, axial load; rotational speed and polishing time are more
significant at 95% CI as it has pvalue less than 0.05. Correlation coefficient
of 0.887 implies that the model has good fit with the experimental results.
Table 9: 
ANOVA for material removal rate (MRR) 


Fig. 8: 
Normal probability plot of the residual for response MRR 
It is observed from ANOVA Table 9 that model comparison of
MRR shows that the model has well linear fit with the predicted values as the
pvalue is less than 0.05. Normal probability plot of the residual for response
Ra (Fig. 8) shows that the errors are normally distributed
as the points in the plot will roughly form a straight line.
CONCLUSIONS
In this experiment, the surface of SS316L was polished by using Electrophoretic
deposition assisted polishing (EPDAP) method. The surface quality can be distinctly
improved in short period of time. The following conclusions can be drawn:
• 
In this experiment, the surface of SS316L was polished by
using Electrophoretic deposition assisted polishing (EPDAP) method. The
surface quality can be distinctly improved in short period of time. The
following conclusions can be drawn. The surface roughness obtained up to
0.04557 μm (45.57 nm) in these experiments whose surface roughness
before EPDAP method was 0.75 μm of stainless steel 316L. Hence, surface
roughness reduction is achieved 93.92% 
• 
The optimum polishing performance for the surface roughness is obtained
at abrasive grit size 2000, Axial load (L) = 12 N, rotational speed (S)
= 1500 rpm, polishing time (t) = 10 min. the surface roughness is decreases
with decreasing in abrasive particle size 
• 
EPDAP also improves in surface quality in terms of reflectivity 
• 
The MRR = 4.2 mg min^{1} is obtained maximum at abrasive grit
size 1500, Axial load (L) = 20 N, Rotational speed (S) = 1500 rpm, Polishing
time (t) = 15 min 
• 
The grit size, factors axial load and rotational speed are more significant
for electrophoretic deposition assisted polishing of SS316L 
Thus the EPDAP is a very effective polishing process to improve surface finish.
ACKNOWLEDGMENTS
The authors wish to acknowledge the support of Govt. of India for the TEQIP
funds provided to purchase CNC Milling machine on which the experimental set
up was developed. Further thanks are to B. Tech. students Sameer Paritkar and
Prasad Ingale for their assistance in development of this EPDAP set up in the
lab.