Int J Adv Manuf Technol (2007) 34: 122–130

DOI 10.1007/s00170-006-0573-8

ORIGINA L ARTI CLE

Ching-Tien Lin . Lieh-Dai Yang . Han-Ming Chow

Study of magnetic abrasive finishing in free-form surface

operations using the Taguchi method

Received: 18 July 2005 / Accepted: 6 February 2006 / Published online: 9 May 2006
# Springer-Verlag London Limited 2006

Abstract This study employed magnetic abrasive finish- precise surfaces. Hence, numerous finishing techniques
ing (MAF) to conduct free-form surface abrasion of have been applied for finishing parts to obtain parts with
stainless SUS304 material operations. The operations high quality. These techniques include chemical mechan-
were demonstrated using a permanent magnetic finishing ical polishing (CMP), electrical polishing (EP), and many
mechanism installed at the CNC machining center. The others. However, both CMP and EP suffer from the
operations were performed using the Taguchi experimental formation of pollutants during its operations, and also yield
design, considering the effects of magnetic field, spindle surfaces with limited quality. Consequently, researchers in
revolution, feed rate, working gap, abrasive, and lubricant. the industry and academics have attempted to develop a
Furthermore, the experimental data was collected using the better means of obtaining a high-precision surface, with
Taguchi experimental design. The optimal parametric low cost, high efficiency, ease of operation and limited
conditions for processing stainless SUS304 material were environmental pollution.
applied in a two-stage process comprised of rough Following recent technological developments, stainless-
finishing that involved MAF followed by a precise steel materials with characteristics of anti-oxidizing, anti-
finishing of the surface. Prior to rough finishing, the corrosive, and shiny surface have been applied in electronic,
Rmax value was 2.670 μm; after rough finishing, the value biochemical and medical instrumentation equipments. The
was 0.158 μm. Precise finishing yields an even lower value surface of stainless-steel parts must be extremely smooth to
of 0.102 μm similar to that of the mirror surface. Therefore, prevent pollution. Optimally, the surface finish can reach a
the results revealed that MAF provides a highly efficient level where it looks like a mirror. A smooth stainless-steel
way of obtaining surface finish. surface not only improves part quality but it also prevents
rusting and staining of the part surface. Finished parts can
Keywords Magnetic abrasive finishing . prevent the occurrence of the following situations:
Taguchi experimental design . Surface finish
1. Chemical liquid or powder particles remaining on the
part surface, therefore, affecting product quality.
2. Contact between parts and the stainless-steel surface
1 Introduction
can cause particles to remain on the surface.
3. The rough surfaces that reside with oil dusk or food
The rapid development of the semiconductor, biotechnol-
particles can cause products to go bad.
ogy, and optical electronic industries has increased the
4. Stainless-steel burr of processed parts falling off when two
importance of geometrical precision and part surface
parts contact each other and cause products to go bad.
quality. Finishing is regularly applied to parts to obtain
Stainless steel is a soft, tough, and difficult finishing
C.-T. Lin . H.-M. Chow material. Thin-plate stainless steel that uses traditional
Department of Mechanical Engineering, processes is not easy to achieve a good surface finish.
Nan Kai Institute of Technology, Hence, manual finishing was usually applied to achieve a
Nan Tou, Taiwan, Republic of China
surface finish that looks like a mirror. However, it is very
L.-D. Yang (*) time consuming and inefficient to achieve a good surface
Department of Industrial Engineering and Management, finish using manual finishing techniques for stainless
Nan Kai Institute of Technology, container steel surfaces.
Nan Tou, Taiwan, Republic of China
e-mail: [email protected] To resolve these problems, magnetic abrasive finishing
Tel.: +886-49-2563489 (MAF) was recently created. MAF involves the use of a
Fax: +886-49-2554412 permanent magnet or an electronic magnet to generate a
 123

magnetic field, and the magnetic abrasives are formed as a Rotation

Magnetic pole
flexible magnetic brush for pressing the workpiece [1, 2].
Thus, the magnetic brush becomes a finishing tool, and the
magnetic abrasives of the magnetic brush stick to the
workpiece during the finishing. Moreover, the frictional Grooves
force generated by the abrasive finishing can remove Magnetic Abrasives Line of magnetic force
particles of free-form surface. The procedure is repeated
until a desired surface finish is attained. Magnetic equipotential
When a permanent magnet was installed on the topside Workpiece
line
of the workpiece, any uneven or concave areas on the part N pole i
could be finished [3–5]. Moreover, when the magnetic pole Fy
Fx
was installed inside or outside of the part, the internal and F
external pipes could also be finished [6]. Therefore, MAF
is a multi-function precise finishing method. Workpiece Workpiece
S pole P
materials can be magnetic (such as steel) or non-magnetic holder
(such as ceramic), and the material removal rate can also be Fig. 2.1 Flexible magnetic brush and dispersion of magnetic force
adjusted based on the size of the magnetic abrasives. The
finishing pressure is controlled via the magnetic field, so
MAF is used for micro-pressure finishing [7, 8]. Thus, the 2.2 Dispersion of magnetic force
MAF method achieved a highly efficient way of obtaining
a good surface finish. Figure 2.1 shows the free-form magnetic field distribution
This study first attempts to develop a surface finishing in the working zone and the magnetic force (F) acting on a
technique for stainless steel, with the aim of optimizing ferromagnetic particle. In the free-form MAF process, a
design parameters and thus reducing operating costs. sufficient quantity of magnetic abrasives is filled in the
Secondly, this investigation seeks to enhance the geometric working zone. Moreover, the inner magnetic field strength
precision and surface finish of parts, to meet the customer is larger than the outer. Therefore, a non-uniformly
requirements. concentrated magnetic field distribution is formed between
the magnetic pole and the workpiece holder.
The magnetic force (F) is applied to the magnetic
2 Magnetic abrasive finishing (MAF) abrasives at position “ i ”, which is situated at the outside of
the working zone to concentrate and pack them towards the
2.1 Fundamental principle working zone. The magnetic force (F) that acts on the
single volume of the magnetic abrasives can be expressed
Magnetic abrasive finishing of free-form surfaces involves by this formula Eq. (2.1) [9]:
filling the gap between the circular magnetic pole and the

workpiece with the magnetic abrasives. The magnetic F ¼ V0 BrB μ (2.1)
abrasives consist of sintered pure iron powder (99.9% Fe)
and Al2O3. The end face of the magnetic pole absorbs the
magnetic abrasives and forms a closed-loop magnetic field Where B and ∇ B are magnetic induction and its gradient
with the workpiece holder. The magnetic abrasives are in “ i ” point of the single volume in the working zone and
generated in a non-uniformly magnetic field in which the μ is magnetic permeability.
abrasives will join each other and follow the direction of Since the magnetic permeability of the magnetic abrasive
the magnetic force to form a flexible magnetic brush. Refer is larger than permeability of vacuum (μ>>μo). The magnetic
to Fig. 2.1 to see how the magnetic brush acts on the free- force in Eq. (2.1) can be expressed as a two-dimensional
form surface. The magnetic force lines generated power to magnetic field distribution following Eq. (2.2) [10]:
apply pressure from the magnetic abrasives to the work-
piece, and the magnetic brush became a tool for finishing @H
Fx ¼ V0 Xm μ0 H
the workpiece. Moreover, the magnetic abrasives in the @x
(2.2)
magnetic brush stick to the workpiece. When the magnetic @H
pole rotates and moves with the workpiece relatively, the Fy ¼ V0 Xm μ0 H
@y
frictional force generated from MAF causes the abrasives
to finish the particles of uneven or free-form surfaces until
it becomes smooth. Moreover, the magnetic brush Where Fx is the x component of the magnetic force, Fy is
continues to move on the x-y-z direction of the CNC the y component of the magnetic force, x is the direction of
machine, brushing the workpiece until it meets the the line of magnetic force, y is the direction of the magnetic
customer’s requirements. equipotential line, V0 is the volume of the magnetic
particle, Xm is magnetic susceptibility of the particle, μ0 is
permeability of vacuum, H is the magnetic field strength at
124
Table 3.1 Magnetic properties of Nd-Fe-B magnet
Characteristics grade Residual induction Br Coercive force bHc Intrinsic coercive force iHc Max. energy product (BH) max
T KGs KA/m KOe KA/m kOe KJ/m3 MGOe

N35 1.17–1.21 11.7–12.1 860–907 10.8–11.4 955 12.0 263–279 33–35

point “ i ”, and ∂H/∂x and ∂H/∂y are gradients of magnetic when the high magnetic field is applied and the finishing
field strength in the x and y directions, respectively. can be processed smoothly.
From Eq. (2.2), the magnetic force Fx and Fy are
proportional to the magnetic particle volume, susceptibility
of the magnetic particle, the magnetic field strength and its 3 Experimental design and setup
gradients. The magnetic forces Fx and Fy are also capable
of preventing the splashing of the magnetic abrasives 3.1 MAF mechanism
caused by the high-speed rotation of the magnetic pole.
During the finishing process, the congregated magnetic This investigation involved the MAF mechanism illus-
abrasives form a magnetic brush along the line of magnetic trated in Fig. 3.1. A permanent magnet generated the
force within the working zone, which causes the pressure P magnetic force; the magnetic field formed a closed loop
on the free-form surface and this pressure will act on the due to the interaction of the permanent magnet, magnetic
work surface. Equation 2.3 represents the pressure, P, as abrasives, workpiece, and workpiece holder (S10C steel).
follows [10]: The magnetic pole was made by a strong magnetic material
   (Nd-Fe-B) in which the magnetic properties are expressed
1 in Table 3.1. The magnetic flux density was close to 1.2
p ¼ μ0 H 1
2
2 (2.3)
μm Tesla in a 1.0-mm working gap (distance between magnetic
pole and workpiece holder). The S pole of the magnet was
where μm is the relative magnetic permeability of the established with a shank installed in the spindle of the CNC
magnetic abrasive particle. machine (Fig. 3.2). Meanwhile, the N pole of the magnet
The particles on the workpiece surface can be finished was designed to absorb the magnetic abrasives. The
out when the magnetic pole absorbs the magnetic abrasives magnetic pole had an external diameter of 20 mm and a
to rotate and move with the workpiece relatively. There- length of 40 mm. Furthermore, the N pole with a 10-mm
fore, an effective finishing surface can be achieved. From radius ball shape was processed into four grooves with
Eq. (2.3), P represents the magnetic pressure acting on the sizes of 1.5 mm width and 10 mm depth to reduce the ball
workpiece surface and when the magnetic pressure area of the magnetic pole and boost the magnetic field
increases so does the material removal rate. This also strength for achieving an efficient finishing.
creates a big finishing depth and affects the surface quality
directly. Simultaneously, the resistance force was generated
due to the abrasives finishing and the centrifugal force was 3.2 Magnetic abrasives
from a rotating magnetic pole, and their resultant force will
enforce the abrasives to splash out of the working zone. The magnetic abrasives must be able to be magnetized (Fe)
However, the splash out phenomenon can be prevented and have the ability of finishing (Al2O3). Generally, it is a

Spindle
Rotation

Magnetic pole
Workpiece
Magnetic Abrasives Workpiece
holder

Fig. 3.1 Schematic view of free-form MAF Fig. 3.2 Magnetic abrasive finishing mechanisms
 125

3.3 Taguchi experimental design

The Taguchi experimental design involved three stages.

First, a Taguchi orthogonal array L18 was used for
experiments to ensure consideration of the most significant
factors and levels, therefore, optimizing the surface finish
in MAF. This investigation considered five factors:
lubricant, revolution, feed rate, working gap, and abrasive.
Among these factors, the lubricant has two levels and the
other factors have three levels (Table 3.2). Secondly, after
the data collection, analysis of variance (ANOVA) was
analyzed using statistical software to identify the signifi-
cance of the factors considered in this study. Finally, the
optimal operation condition was generated and the
confirmatory tests were conducted.

4 Experimental results and discussion

Fig. 3.3 SEM (X 100) of magnetic abrasives
In the Taguchi experimental design, the signal-to-noise
complex material. The more easily magnetized materials ratio (S/N) is extremely important in measuring and
include iron, cobalt, nickel, etc. The materials most assessing part quality; the S/N ratio is always a “larger-
commonly used to finish the workpiece in MAF include: the-better” (LTB) quality characteristic. For example,
aluminum oxide, silicon carbide, boron nitrogen, boron signal transmission is an LTB characteristic for a television.
carbide, diamond powder, etc. After performing the experiments using an L18 array, a
In this study, the magnetic abrasives were typically surface roughness was collected.
mixed with 60 wt% iron powder and 40 wt% aluminum
oxide (5 μm). The abrasives were then compressed into a
cylindrical mode and sintered in a vacuum furnace at 4.1 Surface roughness analysis
1200°C. Following the sintering process, the magnetic
abrasives were crushed into small particles with diameters Surface roughness indicates a surface profile of a finished
of approximately 150 μm, and the magnetic abrasives part; Rmax (μm) is used to represent the surface roughness
became well mixed. During finishing, iron powder and in this study. The quality characteristic of Rmax is “smaller-
aluminum oxide were difficult to separate since they were the-better” (STB), and the η value of the S/N ratio is
cohered after sintering. Figure 3.3 shows the magnetic calculated according to Eq. (4.1) (refer to Table 4.1).
abrasives under a scanning electronic microscope (SEM, !
JSM-6360LV type). 1X n
η ¼
10 log y2
(4.1)
n i¼1 i

Table 3.2 Operation parameters of Taguchi experimental design

Observed values Control parameters Level
1 2 3

Surface roughness Rmax(μm) Lubricant Oil(A) Liquid(B)

Revolution (rpm)(C) 500 1,000 1,500
Feed rate (mm/min)(D) 10 20 40
Working gap (mm)(E) (magnetic flux density) 1.5 (1.0 Tesla) 2 (0.85 Tesla) 2.5 (0.7 Tesla)
Abrasive (gram)(F) 2 3 4
Note:
(A) Chinese Petroleum Corporation 31C cutting oil (ISO32)
(B) Mixture of 2–3% HD-233A finishing liquid and 97–98% water
(C) Revolution of magnetic pole
(D) Feed rate of magnetic pole
(E) The distance of magnetic pole and workpiece holder
(F) Abrasive mass of per finishing
126
Table 4.1 L18 array and S/N ratio for surface roughness
NO. Control parameter S/N ratio η (db)
Lubricant Revolutions Feed rate Working gap Abrasive Rmax

1 Oil 500 10 1.5 2 12.22

2 Oil 500 20 2 3 11.46
3 Oil 500 40 2.5 4 10.56
4 Oil 1,000 10 1.5 3 11.91
5 Oil 1,000 20 2 4 10.29
6 Oil 1,000 40 2.5 2 12.92
7 Oil 1,500 10 2 2 14.31
8 Oil 1,500 20 2.5 3 13.07
9 Oil 1,500 40 1.5 4 7.49
10 Liquid 500 10 2.5 4 14.26
11 Liquid 500 20 1.5 2 11.62
12 Liquid 500 40 2 3 9.81
13 Liquid 1,000 10 2 4 12.86
14 Liquid 1,000 20 2.5 2 15.3
15 Liquid 1,000 40 1.5 3 8.81
16 Liquid 1,500 10 2.5 3 14.05
17 Liquid 1,500 20 1.5 4 8.51
18 Liquid 1,500 40 2 2 11.09

Where n is the number of measurements made on the Table 4.3 Optimal experimental conditions
workpiece (in this study, n=4) and yi is the measured Rmax Control factor Notation Setting
value.
After the statistical analysis, analysis of variance Lubricant A2 Liquid (HD-233A)
(ANOVA) was generated to identify the significance of Spindle revolution B2 1,000 rpm
the factors considered in this study (refer to Table 4.2). Feed rate C1 10 mm/min
Table 4.2 shows that the working gap, feed rate, and Working gap D3 2.5 mm
abrasive significantly influence the surface roughness of Abrasive E1 2g
the workpiece since the statistical Ft values of working gap,
feed rate, and abrasive all exceeded the critical value of F2,
8, 0.05 = 4.46. The contributions of each factor were 40.35% of each factor, while the vertical axis represents their S/N
for working gap, 37.68% for feed rate, 19.46% for ratio. As mentioned above, the quality characteristic of the
abrasive, 1.36% for revolution rate, and 0.6% for lubricant, surface roughness is “smaller-the-better”, while the S/N
respectively. ratio is always “larger-the-better”. Accordingly, Fig. 4.1
After calculating the S/N ratio as listed in Table 4.1, the illustrates that the optimal quality levels of each factor are
response values for surface roughness were obtained, as as follows: lubricant at A2, spindle revolution at B2, feed
plotted in Fig. 4.1. The horizontal axis represents the level rate at C1, working gap at D3 and abrasive at E1,

Table 4.2 ANOVA and contribution of operation parameters

Parameter Degree Square sum Variance F-test F0.05 Contribution
(A) (fA) (SA) (VA) (Ft) (CA)(%)

Lubricant 1 0.24* 0.24 1.09 5.32 0.60

Revolution 2 1.08* 0.54 2.48 4.46 1.36
Feed rate 2 29.90 14.95 68.77 4.46 37.68
Working gap 2 32.01 16.01 73.65 4.46 40.35
Abrasive 2 15.44 7.72 35.52 4.46 19.46
Error 8 1.74* 0.22 0.55
Pooled error (11) (3.06) (0.2782)
Total 17 80.40 39.67 100.00
*Indicates the sum of squares added together to estimate the pooled error sum of squares in parentheses
 127
Fig. 4.1 Response values for 14.0
surface roughness S/N ratio
13.5

S/N Ratio of Surface Roughness ( db)

13.0

12.5

12.0

11.5

11.0

10.5

10.0

9.5

9.0
A1 A2 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3
Lubricant Revolution Feedrate Working gap Abrasive

respectively. The abbreviation of the optimal operation Where ηpre is the predicted S/N ratio at optimal finishing
combination is A2B2C1D3E1. Thus, the optimal operating conditions, η is the average S/N ratio of all control factors,
conditions applied in the final confirmatory experimental and Rmaxpre is predicted surface roughness.
runs are finishing liquid as lubricant, rate of revolution
1,000 rpm, feed rate 10 mm/min, working gap 2.5 mm and
mass of abrasive material 2 g (refer to Table 4.3).
From the ANOVA, the significant factors affecting 4.2 Confirmatory test
surface roughness were feed rate at C1, working gap at D3
and abrasive at E1, respectively. The prediction values of S/ The optimal operation condition was used to confirm the
N ratio
 and average for the surface roughness were (ηpre) experimental runs; usually if the response values from the
and Rmaxpre ; they could be obtained from Eqs. (4.2) and confirmation runs could achieve the 95% confidence
(4.3). The results are shown on Table 4.4. interval then the optimal operation condition will be
verified statistically. The collected Rmax of the six
ηpre ¼ η þ ðC1
ηÞ þ ðD3
ηÞ þ ðE1
ηÞ confirmation runs were (0.155, 0.156, 0.158, 0.159,
0.160, 0.160 μm). The average Rmax is equal to 0.158 μm
¼ C1 þD3 þ E1
2η and S/N ratio is equal to 16.026 db, respectively.
(4.2)
¼ 13:269 þ 13:359 þ 12:909
2 11:697 The following confidence interval (C.I.) statistical
formula for the optimal combination (A2 B2C1D3E1) will
¼ 16:143ðdbÞ
be applied to verify the confirmation runs. To simplify the
calculation of the formula, the researchers chose the most
significant factors (C1, D3, and E1) to calculate the
Rmaxpre ¼ 10
ηpre =20 ¼ 10
16:143=20 ¼ 0:156ðμmÞ (4.3)

Table 4.4 Compare confirmatory test results and predicted value

Optimal combination
Predicted values Confirmatory test results

Surface roughness Rmaxpre Surface roughness S/N ratio ηpre Surface roughness Rmaxver Surface roughness S/N ratio ηver
Level A2B2C1D3E1 A2B2C1D3E1
0.156 (μm) 16.143 (db) 0.158 (μm) 16.026 (db)
Rmax of the six confirmation runs were (0.155, 0.156, 0.158, 0.159, 0.160, 0.160 μm). The average Rmax is equal to 0.158 μm and S/N ratio
is equal to 16.026 db
The ηver value of the S/N ratio is calculated according to Eq. (4.1).
pre

ver ¼ 16:143
16:026 ¼ 0:117ðdbÞ
128

experimental effect; therefore, the following C.I. calcula- 4.4 The effects of the operation parameters
tion was obtained according to Eq. (4.4) [11].
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi From the analysis of ANOVA (Table 4.2), the significant
 ffi operation parameters of working gap, feed rate, and
1 1
CI ¼ F0:05;1;fe  Ve  þ abrasive will be discussed as follows:
neff r
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  (4.4) (1). Effects of working gap
7 1 The smaller the working gap between the magnetic
¼ 4:84  0:2782  þ ¼ 2:382ðdbÞ pole and the workpiece holder, the stronger the
18 6
magnetic field strength will be. At the same time, the
surface finish will be affected, therefore, it is important
Ve ¼ 0:24þ1:08þ1:74
11 ¼ 0:2782 (Refer to Table 4.2) to adjust the working gap properly in order to obtain a
god quality surface finish. In this study, when the
N 18 18 working gap was decreased (from 2.5 to 1.5 mm), and
neff ¼ ¼ ¼
1þn 1þ2þ2þ2 7 the magnetic field strength increased (from 0.7 to 1.0
Tesla). Simultaneously, increasing the density of the
magnetic abrasives will create a shorter and stiffer
Where fe is the degrees of freedom for pooled error (fe=
magnetic brush and a larger pressure on the workpiece.
11), Ve is the variance of the pooled error, N is the total trial
Then, the workpiece surface can be finished deeply due
number (N=18), n is the total main factor degrees of
to the softer and tougher SUS304 material (Vickers
freedom (n=6), r is the confirmatory test trial number (r=6).
hardness HV300), so the surface roughness becomes
And neff is the effective sample size.
worse in which the Rmax value is from Rmax=0.158 μm
After the calculation, the results showed ηpre
ηver ¼ (good quality) up to Rmax=0.232 μm (bad quality). On
0:117ðdbÞ (see Table 4.4), since 0.117 db is smaller than the contrary, a large working gap will generate
2.382 db, therefore, 95 % C.I. of confirmation test was insufficient abrasives, a dysfunction at magnetic brush,
verified. and obtain a bad workpiece surface. Thus, a working gap
(2.5 mm) of the Taguchi experimental design will yield
the optimum surface finish in this study.
(2). Effects of feed rate
4.3 The effects of finishing pass number on the surface
The lower the feed rate of the magnetic pole, the more
finish
the magnetic abrasives will pass through the working
zone at the same time, thus, improving the surface
The number of finishing passes must be considered to
finish. However, increasing the feed rate will enable
determine the differences between rough and precise
the magnetic abrasives to pass the working zone
finishing. Optimal operating conditions for performing
quickly. Thus, the material removal rate will be
rough and precise finishing (Table 4.5) obtained a surface
reduced relatively, and the previous finishing imprints
roughness Rmax of 0.158 μm for rough finishing and a
will be retained on the working surface. Therefore, to
surface roughness Rmax of 0.102 μm for precise finishing
attain an optimum surface finish we should choose the
as illustrated in Fig. 4.2. Although more finishing yields a
lower feed rate, which is 10 mm/min from the Taguchi
better surface roughness in this study, whenever the
experimental design in this study.
finishing number was over two times, the surface rough-
(3). Effects of the magnetic abrasives
ness remained unchanged, and the material removal rate
The mass of the abrasives does not have the charac-
declined as the number of finishing passes increased
teristic of larger-the-better because the extra abrasives
because of the abrasive material wearing off gradually
cannot produce the stirred function and decrease the
(refer to Fig. 4.3).
finishing function of the magnetic brush. Simulta-
Table 4.5 Optimal experimental conditions for two stage finishing neously, the abrasives become a condensed block
against the surface of the workpiece to perform a hard
Spindle revolution 1,000 rpm contact and retain large amounts of scratches, there-
Feed rate 10 mm/min fore, making the surface finish unacceptable.
Working gap 2.5 mm
Magnetic flux density 0.7 Tesla Meanwhile, a small mass of abrasives can cause an
Lubricant Liquid (HD-233A) insufficient filling in the working zone, and a dysfunction
Workpiece holder revolution 60 rpm
of the magnetic brush in which it cannot reach a good
Magnetic abrasive size Fe+Al2O3 sintering
surface finish. In this study, we should choose the right
Rough finishing 150 μm, 2 g
mass of abrasives (2 g) in order to boost a good surface
finish.
Precise finishing 75 μm, 2 g
In summary, to attain a highly efficient surface finish of
Pass number
SUS304, one should: enlarge the working gap, lower the
Rough finishing 1
feed speed, decrease the amount of abrasives, and choose
Precise finishing 2∼5
an adequate revolution in order to obtain an optimum
 129
Fig. 4.2 Before rough finishing
(Rmax =2.670 μm) (left) and
after precise finishing (Rmax =
0.102 μm) (right)

surface finish. This research used a permanent magnet light weight, and a high revolution with an easy tool
mounted on the spindle of the CNC machine as a finishing change. Furthermore, we have researched the magnetic
tool, in which it can attain the following advantages: a abrasive Al2O3 and ferrite powders and have concluded
magnetic pole with a permanent magnetism, strong mag- that it maintains high machining efficiency, is relatively
netic field intensity, low energy and cost, small size and cheap, and provides a splendid surface finish. The

Fig. 4.3 Effects of finishing 3 70

pass number on surface finish
2.75
Rmax MRQ
2.5 60

2.25
50 Material Removal Quantity(mg)
2
Surface Roughness Rmax (µm)

1.75
40
1.5

1.25
30
1

0.75
20
0.5

0.25 10
0

-0.25 0
0 1 2 3 4 5
Finishing pass number
130

application of 75-μm magnetic abrasives to the precision magnetic abrasive finishing for micro-parts may be
finishing surface of plane or curved plane can yield a value regarded as the response variables.
of Rmax 0.1 μm, which is similar to that of the surface of a
mirror. Therefore, the MAF technique can be applied in Acknowledgement The authors would like to thank the National
product precision finishing in the industry. Science Council of the Republic of China for financially supporting
this research under Contract No. NSC 92-2212-E-252-002.

5 Conclusions
References
This study reaches the following conclusions and makes the
1. Shinmura T, Takazawa K, Hatano E (1986) Study on magnetic
following suggestions regarding future research directions. abrasive finishing (1 Report) - On process principle and a few
(1). The ball-shaped magnetic pole has special grooves to finishing characteristics. J JSPE 52:851–857
2. Shinmura T, Takazawa K, Hatano E, Aizawa T (1985) Study on
form a flexible magnetic brush and increases a high magnetic abrasive process - process principle and finishing
finishing efficiency. This design will create a good possibility. J JSPE 19:54–55 Mar
surface finish for the soft, tough, and difficult finishing 3. Shinmura T, Hatano E, Takazawa K (1986) Development of
materials (e.g., non-ferromagnetic SUS304). plane magnetic abrasive finishing apparatus and its finishing
performance. J JSPE 52:1080–1086
(2). From the finishing operations, the researcher has found 4. Shinmura T (1989) Study on plane magnetic abrasive finishing
that the working gap has the largest impact on the (3 Report) - On the finishing characteristics of non-ferromag-
finishing quality. Accordingly, a proper working gap netic substances. J JSPE 55:1271–1276
(in this case, 2.5 mm) can reduce surface imprints and 5. Kim JD, Choi MS (1997) Study on magnetic of polishing free-
form surface. Int J Mach Tool Manuf 37:1179–1187
increase quality. 6. Shinmura T, Yamaguchi H (1993) Study on a new internal
(3). After analysis of the Taguchi method, the factors that finishing process by the application of magnetic abrasive
significantly affected the surface finish include the machining (internal finishing process of 7. Stainless-steel tube
working gap, feed rate, and the abrasive. The optimal and clean gas bomb). Trans Jpn Soc Mech Eng Part C 59:293–299
operation condition was a working gap of 2.5 mm, a 7. Shinmura T, Takazawa K, Hatano E (1987) Study on magnetic
abrasive finishing (3 Report) - Finishing characteristics non-
feed rate of 10 mm/min, and an abrasive mass of two ferromagnetic substances. J JSPE 53:1440–1446
grams. Even though the finishing lubricant and spindle 8. Shinmura T, Wang FH, Aizawa T (1993) Study on a new
speed were not significant factors affecting the surface finishing process of fine ceramics by magnetic abrasive
finish, the finishing lubricant (liquid, HD-233A) and machining- on the improving effect of finishing efficiency
obtained by mixing diamond magnetic abrasives with ferro-
spindle speed (1000 rpm) were applied to the magnetic particles. J JSPE 59:1251–1256
confirmation tests due to convenience and cost. 9. Baron YM (2000) Theory of magnetic abrasive finishing and
(4). In the confirmation runs using the optimal operation development of this method in Russia. Proceedings of 6th
condition to conduct six experiments, an average International Conference “Precision Surface Finishing and
Deburring Technology” pp 344–361
surface roughness of 0.158 μm is achieved and the S/N 10. Chang GW, Yan BH, Hsu RT (2002) Study on cylindrical
ratio is equal to 16.026 db, respectively. After magnetic abrasive finishing using unbounded magnetic abra-
statistical calculation, the 95% C.I. is 2.382 (db), sives. Int J Mach Tools Manuf 42:575–583
which is larger than 0.117 (db), therefore, the 11. Deng CS, Chin JH (2005) Hole roundness in deep-hole drilling
confirmation test was verified. as analyzed by Taguchi methods. Int J Adv Manuf Technol
25:420–426
(5). For further study, the researchers may suggest that
micro magnetic abrasives as a finishing tool and the