Al-Nahrain Journal for Engineering Sciences NJES24(1)16-25, 2021

https://doi.org/10.29194/NJES.24010016

Study the Effect of Inductor and Pole Geometry on the

Surface Roughness and Material Removal Weight Using Magnetic
Abrasive Finishing Method
Basma Luay Mahdi1, Ali H. Kadhum2*

Abstract
Author’s affiliations: The traditional finishing method cannot keep up with recent labor
1) Al-Khwarizmi College of market requirements, solve the problem of increasing production, improve
Engineering, University of the surface roughness and accuracy of workpiece. While the unconventional
Baghdad, Iraq. magnetic abrasive finishing (MAF) method has shown as a promising
[email protected] technique that can be used to finish complicated surfaces. MAF finishes
u.iq metals, alloy, ceramic, and other materials that are difficult to finish by other
processes. In another word, MAF improves the quality of surfaces with low
2*) Al-Khwarizmi College of cost.
Engineering, University of This paper focuses on optimize and study the effect of inductor and
Baghdad, Iraq. pole geometry (radius of hole, angle of core, angle of pole, radius of pole), on
[email protected] (surface roughness (Ra) and material removal weight (W)) and fined the
optimum values that increase the efficiency of MAF method. Taguchi method
Paper History: employed to study the influence of geometry parameters and find the
optimum values using orthogonal array L9. The results conclude that the most
Received: 18th July 2020
significant factor that effects change in surface roughness (ΔRa) and material
removal weight(ΔW) are radius of the hole (R) and angle of core (α),
Revised: 3rd Dec. 2020
respectively.
Accepted: 23rd Jan. 2021 Keywords: Magnetic Abrasive Finishing, Surface Roughness, Material
Removal Rate, Geometry of Inductor and Pole.

1. Introduction: finish it. The MAF process removes a very small

Magnetic abrasive finishing is a new finishing amount of material by indentation and rotation of
technique, used for polishing flat, cylindrical, and other magnetic abrasive particles in the circular tracks [4, 5].
complicated internal and external shapes that are The principle of the MAF method is a ferromagnetic
difficult to polish by using a conventional machining abrasive powder densified by magnetic field energy
process. Magnetic abrasive finishing removes which acts as a surface finishing brush, as shown in
microchips and enhance the quality of the surface[1]. Figure (1.
MAF was initially developed as a machining process in The development of MAF in various materials and
the US in the 1930s, who have the first patent there shapes was a concerning issue for researchers. Yang et
[2]. In the MAF process, the magnetic field controls al. [6], implements an optimizing for MAF using
and manages the cutting force. Thus, finishing Taguchi's parameter design, on AISI304 stainless steel.
pressure is applied electromagnetically that have They analyze the magnetic field characteristics for
produced a magnetic field [3]. The cutting tool three different poles designs solid cylindrical pole,
significantly affects the quality and accuracy of hollow pole, and hollow cylindrical grooved
finishing processes. In the MAF process, the magnetic pole. Among these designs the hollow cylindrical
abrasive powder (MAP) plays the role of the cutting grooved pole shows a better surface roughness quality.
tool. MAP occupies the working gap between the pole In terms of parameters influence, the magnetic field,
and the workpiece. Accordingly, magnetic flux density pole rotational speed, feed rate, working gap, abrasive,
passes through the coil to create a magnetic field which and lubrication are the effective parameters relating to
magnetized MAP and formulates a magnetic abrasive the process. Their results showed that the MAF
flexible brush (MAFB), MAFB acting as a multipoint technique leads to considerable surface quality.
cutting tool. MAP gains an energy from the magnetic Deepak et al. [7], to reach a higher surface quality (like-
field, this gained energy is used to generate cutting and mirror surfaces) by improving MRR and Ra using
finishing forces to process the surface of machined stainless steel SS-409 as a workpiece. In this paper, it
parts. The workpiece is also magnetized. The magnetic showed that MAF achieves a higher surface quality
force acts between the workpiece and the brush to than the conventional finishing processes like lapping,

NJES is an open access Journal with ISSN 2521-9154 and eISSN 2521-9162
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

16
 NJES24(1)16-25, 2021
Mahdi & Kadhum

honing, etc. Yamaguchi et al. [8], an attempt to in this work is the magnetic abrasive particles (brush).
maintain the property of surface while enhance Ra The results show an improvement in Ra by (50-60%)
with minimum MRR. The most important parameter and tool life by (150%).

(a) (b)
Figure (1): Principle of MAF process. Where working gap is the space between the tool and the workpiece

Mousa [9], study the optimization and influence of value at the optimal finishing conditions. By using the
parameters (amplitude of pole geometry, number of S/N ratio and ANOVA analysis [12]. The MAF
cycles of pole geometry, working gap, finishing time, process is also effective on the internal surface such as
cutting speed, and current) on the quality of the aluminum pipes, by using adhesive magnetic abrasive
surface of MAF for stainless steel 316 particulate. The input parameters used in this process
workpiece. They Utilize Taguchi method for finding are (magnetic flux density, speed, abrasive grit size, and
the prediction models using the signal-to-noise ratio quantity of MAPs). The output characteristics are
with the MINITAB 17 software. The results show that roughness improvement rate (RIR) and material
the surface roughness was decreased from 1.030µm to removal rate (MRR), where Response Surface
0.370µm, that means a higher level of surface Methodology (RSM) method was utilized for
roughness reached to 0.760 µm. The effect of the main analyzing. It concludes that the maximum value of RIR
process parameters (rotational speed of magnet, and MRR that can be achieved at the optimum
working gap, and mixing ratio) was considered on condition are (81.49%, 2.74mg/min), respectively [13].
surface roughness (Ra) during the finishing for the flat Yet, measuring the impact of system geometry of
surface of Inconel 718 alloy. aimed to find the inductors (pole and core) in the MAF system is still an
optimum process parameter that achieves minimum open topic. In this paper, the optimum system
surface roughness value of Inconel 718 workpiece. geometry that achieves the best quality of surfaces
The effect of this parameter on ΔRa has been studied according to the selected criteria (surface roughness
using L9 orthogonal array and analysis of variance and material removal weight) was set.
ANOVA, it concludes that the minimum surface The objective of this paper is to study the effect of
roughness achieved up to 0.0316 microns. While, the varying the MAF system geometry on the physical
current and working gap is the most significant process properties of the finished surfaces. Taking into
parameters [10]. Furthermore, an attempt focused on consideration two criteria: surface roughness, and
developing the of finishing tool by using multiple pole- material removal weight, for Optimizing and
tip to achieve a higher Ra, their results shown 94% improving the quality and the properties of flat surface
improvement in Ra with higher surface quality [11]. (brass workpiece material) by using Minitab software
Besides, many researchers utilize the MAF process program that found the best geometry for magnetic
for polishing the external and internal surfaces. The inductor. Taguchi orthogonal array (OA) were utilized
external sphere surface such as (ball-bearing) made to design the experiments and to find the prediction
from AISI 316L stainless steel that has been polishing model for each criterion. According to the criteria's
by the MAF process. It has input parameters are response of changes in surface quality, Taguchi
(electromagnetic speed, current and the direct voltage recommends two routes for analyzing and optimizing
induced, magnetic flux density, the quantity of abrasive results: signal-to-noise ratio (SNR)and analysis of the
particle size, working environment, and workpieces variance (ANOVA) using Minitab 18 software.
materials), and the output criteria are (material removal Describe the overall procedure steps of MAF for
rate and surface roughness). It concludes that the best the nine experiments as shown in Figure 2.
surface finish (Ra) can be obtained with 0.0766 μm

17
 NJES24(1)16-25, 2021
Mahdi & Kadhum

Figure (2): Flowchart of Experimental Working Procedure.

1.1 Selection the levels and their parameter of Table (1): The values and levels of input (variable)
MAF parameters.
The objective of this work is studying the influence Input parameters code Level1 Level2 Level3
of magnetic inductor and pole geometry on the Radius of hole (mm) R 0 4.5 9
selected criteria (Ra, W). To formulate the problem, Angle of core (degree) α 82° 90° 98°
Angle of pole (degree) β 60° 90° 120°
four input variable parameters have been chosen
Radius of pole (mm) r -18 0 18
(radius of hole (R mm), angle of core (α deg), inner
Table (2): The values of constant variables.
radius of the pole (β mm) and angle of the pole (r deg)). No. Constant parameters Value
While fixing the remaining parameters (electric source, 1 Electric source current (A) 0.85
magnetic inductor velocity (m/s), finishing time (min), 2 Electric source voltage (V) 55
working gap (mm), and ferromagnetic abrasive 3 Magnetic inductor velocity(m/s) 550
powder doze (cc)). The ferromagnetic abrasive powder 4 Finishing time (min) 10
is a mixture of 50g of Tungsten Carbide and 100g of 5 Working gap (mm) 1.5
free Iron (ratio: 2 Iron to 1 Tungsten Carbide), with Doze of ferromagnetic abrasive powder
6 10
mesh (320µm). Measuring the effect of the input (cc)
variable parameters on surface roughness and material
removal rate in the magnetic abrasive finishing
process. The values and levels of input parameters
(variable parameters) and the values of constant
variables are listed in Table (1): The values and levels
of input (variable) parameters.
and
Table (2): The values of constant variables.
, respectively. The geometry parameters of
electromagnetic inductor and pole and their
dimensions are shown in Figure .

18
 NJES24(1)16-25, 2021
Mahdi & Kadhum

4 4.5 82 90 18
5 4.5 90 120 -18
6 4.5 98 60 0
7 9 82 120 0
8 9 90 60 18
9 9 98 90 -18

According to Taguchi technique matrix, nine are

proposed shape for inductor and pole geometry, as
shown in Error! Reference source not found..
After the fabrication of core and pole according to
the scheduled shapes from the Taguchi array, the
turning process started. With a copper wire of 0.5mm
diameter, in which each core is covered by (5000 turns)
Figure (3): The parameters of electromagnetic to form an inductor. The finishing process was
inductor and pole. confined to brass plates as workpiece. It has
dimensions of 100mm length with 50mm width and
1.2 Selection of orthogonal array (OA) for 3mm thickness. Figure 4 depicts the magnetic abrasive
nine experiments. brush that begins to form after the electric current
The input parameters have been applied to a set of passed through the inductor. One of the most
experiments according to an orthogonal array (OA). important factors that affect the shape of the brush is
Orthogonal array is a technique that minimizes the the geometry dimensions of the magnetic inductor and
number of experiments by taking the most effective pole.
experiments. An array of L9 (3^4) gives (81) The magnetic abrasive finishing machine is
independent experiments. Utilizing OA reduces the illustrated in Figure 5. The criteria that affected by a
number to only (9) experiments[6], [14]–[16]. change in surface characteristics are surface roughness
(Ra) and material removal weight (W). Inside the
Table (3): The experimental design of magnetic machining zone, three fixed points 1,2, and 3 as shown
inductor based on L9. in Figure 6, were selected as test points and measured
for each experiment individually, before calculating the
Experiment (R) (α) (β) (r)
average. Then, the results before and after machining
1 0 82 60 -18 are been subtracted to find the change in (ΔRa and
2 0 90 90 0 ΔW).
3 0 98 120 18

Figure (4): Brush formed in MAF process, a) Photograph b) Drawing

Table (4): Shapes of inductor and pole.

19
NJES24(1)16-25, 2021
Mahdi & Kadhum

Shape No.1 Shape No.2 Shape No.3

Shape No.4 Shape No.5 Shape No.6

Shape No.7 Shape No.8 Shape No.9

20
 NJES24(1)16-25, 2021
Mahdi & Kadhum

Figure (5): Magnetic abrasive finishing machine.

Figure (6): Spots of measured the brass plate

1.3 Measuring the output parameters of MAF and signal-to-noise S/N ratio are calculated by using
The nine workpieces are subjected to two different MINITAB 18 software [8 , 9].
tests, before MAF and after the MAF method
1.3.1 Surface Roughness (Ra) 𝟏 𝟏
S/N = −log10 ( ∑𝐧𝐢=𝟏 ( 𝟐 ) ) ….(3)
Surface roughness is measured for 9 experiments 𝐧 𝐲𝐢
by taking three values on the finishing zone, then
calculate the average of the three reads. We repeat the MSD =
𝟏 𝟏
∑𝒏𝒊=𝟏 ( 𝟐 ) ….(4)
measurement procedure before and after the process, 𝒏 𝐲𝐢
for every single workpiece. SRT-6210 was used as a Where n number of experiment (input), and yi
surface roughness measurement tool. number of measurement (output).
to calculate the change in surface Roughness (ΔRa).
𝜟𝑹𝒂 = 𝑹𝒂𝒃𝒆𝒇𝒐𝒓𝒆 − 𝑹𝒂𝒂𝒇𝒕𝒆𝒓 ….(1) 2.2 Analysis of the surface roughness (ΔRa).
Magnetic abrasive finishing operation were used
1.3.2 Material removal weight (W): for enhancing the performance and quality of surface;
The brass plate workpiece weight measured before analyzing the result using S/N ratio method to find the
MAF and after, using the delicate balance for optimal level for each experiment by taking the large
measuring weight. To calculate ΔW the change in number for S/N ratio. Selected this term according to
material removal equation (2) is applied. the standards of Taguchi statistical package in Minitab
𝜟𝑾 = 𝑾𝒃𝒆𝒇𝒐𝒓𝒆 − 𝑾𝒂𝒇𝒕𝒆𝒓 ….(2) software. In Minitab software there are three main sets
of quality characteristics: “larger is better, nominal is
better, smaller is better”. In this research, SNR has
2 Results and Discussion been chosen according to the criterion (it is better to
2.1 Signal to noise S/N ratio have larger) to have the maximum reaction. “SNR”
The optimal level of each parameter is selected and “mean squared deviation (MSD). The mean and
according to the standard (large is better) value of the S/N ratio results and the response for ΔRa are
S/N ratio to have a maximum reaction affected the illustrated below in Table 5 and Table 6 (a,b),
brass workpiece to obtain the refinement of (ΔRa) and respectively.
(ΔW). The results of mean square diversion (MSD)

Table 5. The Results of the mean and S/N ratio of ΔRa.

Radius of Angle of Angle of Radius
Experiment hole core α pole β of pole ΔRa S/N ratio Mean
R(mm) (degree) (degree) r (mm)
NUM. 𝑹 Α β r ΔRa SNRA1 MEAN1
1 0.0 82 60 -18 0.12100 -18.3443 0.12100

21
NJES24(1)16-25, 2021
Mahdi & Kadhum

2 0.0 90 90 0 0.16400 -15.7031 0.16400

3 0.0 98 120 18 0.17314 -15.2321 0.17314
4 4.5 82 90 18 0.11520 -18.7710 0.11520
5 4.5 90 120 -18 0.12085 -18.3551 0.12085
6 4.5 98 60 0 0.11634 -18.6854 0.11634
7 9.0 82 120 0 0.09590 -20.3636 0.09590
8 9.0 90 60 18 0.04930 -26.1431 0.04930
9 9.0 98 90 -18 0.04500 -26.9357 0.04500
Con.test 0 82 120 0

Table 6.a Response table for signal to noise 3 0.06340 0.11149 0.12996 0.11255
ratio for ΔRa. Delta 0.08931 0.00079 0.03442 0.02980
Larger is better Rank 1 4 2 3
Level R α β r
1 -16.43 -19.16 -21.06 -21.21 From Table 6.a and b, the value of S/N ratio and
2 -18.60 -20.07 -20.47 -18.25 mean for angle of core α (degree) are not compatible.
3 -24.48 -20.28 -17.98 -20.05 The main effect plot of S/N ratio and mean is shown
Delta 8.05 1.12 3.07 2.96 in Figure 6.a and b. The angle of core(α) according to
Rank 1 4 2 3 the large number is shown in Figure 6 (a,b) is 98° and
82°, respectively. Therefore, the prediction should
Table 6.b Response table for mean for ΔRa. have been done, according to a large number of S/N
Level R α Β r ratio for the reading that adopted in the present work,
1 0.15271 0.11070 0.09555 0.09562 as shown in Table 7. The angle of core (82°).
2 0.11746 0.11138 0.10807 0.12541

Figure 6.a main effect plot of the mean for (ΔRa) with respect to radius of the hole (R = {0, 4.5, 9}), angle of
core (α= {82, 90, 98}), angle of the pole(β= {60, 90, 120}), and radius of the pole(r= {−18, 0, 18}).

Figure 6.b main effect plot of S/N ratio for (ΔRa) with respect to radius of the hole (R = {0, 4.5, 9}), angle of core
(α= {82, 90, 98}), angle of the pole(β= {60, 90, 120}), and radius of the pole(r= {−18, 0, 18}).

22
 NJES24(1)16-25, 2021
Mahdi & Kadhum

Table 7. The production of the mean and S/N ratio of

ΔRa.
Mean S/N ratio

R α β r R α β r
0 98° 120° 0 0 82° 120° 0
Mean S/N ratio Mean S/N ratio
0.186007 -13.4341 0.185213 -12.3093

The best geometry according to the reading adopted in the present work to reach the minimum surface roughness
according to the boundary condition of the present research is (radius of hole R (0 mm) angle of core ∝1 (82°), angle
of pole β3 (120°), radius of pole r2 (0mm), which is (R1, ∝1, β3 , r2 ). While the most significant factor that has an
effect on ΔRa in MAF process by calculating the difference between the bigger and smaller value of level (Delta) are
radius of core (R) followed by angle of pole (β), radius of pole (r) lastly, angle of core (α) as shown in
Table 6.a Response table for signal to
noise ratio for ΔRa. 2.1.1Analysis of Material removal weight
Larger is better (ΔW).
.a and b, which is shown in the line with (Rank). The second criterion that responds to describe the
The best design level (R1, ∝1, β3 , r2 ) is not among performance and quality of MAF is the (ΔW),
these nine experiments from scheduled by the analyzing the result using the same step previously.
orthogonal array. Therefore, ΔRa can be determined at The result of mean and S/N ratio as shown in Table
8. and the response table for both mean and S/N ratio
the optimum condition ( R1, ∝1, β3 , r2 ). The basic
are shown in Table 9.a and b.
dimension of the inductor and pole to reach the
minimum of surface roughness are (ΔRa 𝑜𝑝 ) as shown
in Figure 8.

Figure (8): The optimum geometry dimension of the core and pole for the optimum shape (ΔRa 𝑜𝑝 ), depending on
values of parameters that adopted in the present work.

Table (8): The Results of the mean and S/N ratio of ΔW.
The Radius of The Angle of The Angle of The Radius of Change in
Experiment the hole a core the pole a pole material S/N ratio Mean
R(mm) α (degree) β (degree) r (mm) removal
NUM. 𝑹 α β r ΔW SNRA3 MEAN3
1 0.0 82 60 -18 0.0199100 -34.0186 0.0199100
2 0.0 90 90 0 0.0189650 -34.4409 0.0189650
3 0.0 98 120 18 0.0105600 -39.5267 0.0105600
4 4.5 82 90 18 0.0223100 -33.0300 0.0223100
5 4.5 90 120 -18 0.0051040 -45.8418 0.0051040
6 4.5 98 60 0 0.0069160 -43.2029 0.0069160
7 9.0 82 120 0 0.0101830 -39.8425 0.0101830
8 9.0 90 60 18 0.0110800 -39.1092 0.0110800
9 9.0 98 90 -18 0.0015500 -56.1934 0.0015500
Con.test 0 82 60 18

23
 NJES24(1)16-25, 2021
Mahdi & Kadhum

Table 9.a :response table for signal to noise ratio

for ΔW.
Level R α Β r
1 -36.00 -35.63 -38.78 -45.35
2 -40.69 -39.80 -41.22 -39.16
3 -45.05 -46.31 -41.74 -37.22
Delta 9.05 10.68 2.96 8.13
Rank 2 1 4 3

Table 9.b: response table for mean for ΔW.

Larger is better
Level R α Β r
1 0.016478 0.017468 0.012635 0.008855
2 0.011443 0.011716 0.014275 0.012021
3 0.007604 0.006342 0.008616 0.014650
Delta 0.008874 0.011126 0.005659 0.005795
Rank 2 1 4 3
From Table 9.a and b (angle of pole β) for both mean and S/N ratio are not matched according to the largest
number of S/N ratio (large is better), as shown in Table 10. The main effect plot of S/N ratio and mean is shown in
Figure 9(a,b). The angle of pole (β) according to the large number in Figure 9.a and b is 90° and 60°, respectively.
Therefore, the prediction must be adopted for solving that mismatch.

Figure 9.a main effect plot of Mean for (ΔW) with

respect to radius of the hole (R = {0, 4.5, 9}), angle
of core (α= {82, 90, 98}), angle of the pole(β=
{60, 90, 120}), and radius of the pole(r=
{−18, 0, 18}).

Figure 9.b main effect plot of S/N ratio for (ΔW)

with respect to radius of the hole (R = {0, 4.5, 9}),
angle of core (α= {82, 90, 98}), angle of the pole(β=
{60, 90, 120}), and radius of the pole(r=
{−18, 0, 18}).

Based on the higher number of signals to noise ratio (-25.8893) in Error! Reference source not found., the
optimal number is found at (raduis of hole R (0mm), angle of core ∝1 (82°), angle of pole β1 (60°),and radius of
pole r3 (18mm). Therefore, the optimum level is (R1, ∝1, β1 , r3 ) for optimum (ΔW). In the same way, according to
the line of (Rank) shown in

24
 NJES24(1)16-25, 2021
Mahdi & Kadhum

Table 9.a :response table for signal to noise is evaluated at these optimum condition values
ratio for ΔW. (R1, ∝1, β1 , r3 ) and is also not found among these nine
(a,b), the value of these optimum levels is not experiments that are shown in
found among the nine experiments. Therefore, (ΔW)
parameter with the highest impact on the ΔW followed
. The basic dimension and fabrication of the by the radius of the hole, followed by the radius of a
inductor and pole to reach the optimum of material pole. While the angle of the pole shows the lowest
removal rate (ΔW𝑂𝑃 ).as shown in Figure 10. impact on ΔW. as shown in Table 12.
The regression model for material removal rate has
2.2 ANOVA Technique been obtained below
2.2.1 ANOVA Technique for (ΔRa) 𝛥𝑊 = 0.0849 − 0.000986𝑅 − 0.000695 𝛼
Analysis of variance (ANOVA) test is a statistical − 0.000067 𝛽 + 0.000161 𝑟
processing which is used for determining the P% R-sq=93.21%
(percentage of contribution) for each parameter
(radius of the hole, angle of the core, angle of the pole,
the radius of a pole). ANOVA has been applied for the
first criteria (ΔRa) as shown in Table 11.
It can be noticed according to Table 11, that the
large parameter effect on surface roughness is the
radius of hole followed by the angle of pole followed
by the radius of a pole which means that these
parameters have a large effect on surface roughness
compared to the angle of the core. The regression
model for surface roughness has been obtained below
𝛥𝑅𝑎 = 0.0998 − 0.00992𝑅 + 0.000050 𝛼
+ 0.000574 𝛽 + 0.000470 𝑟
MINITAB18 is used to find R-seq value that is
clearly assigned the effectiveness of method.
R-sq=91.38% Figure (10): The optimum geometry dimensions
2.2.2 ANOVA Technique for (ΔW) for the inductor and pole for the optimum shape
Using the same way that used previously, ANOVA (ΔW𝑂𝑃 ). depending on values of parameters that
is performed on the second criterion (material removal adopted in the present work
rate ΔW). It has found that the angle of the core is the

Table (10): Analysis of variance (ANOVA) for surface roughness

Source DF Adj SS Adj MS F-Value P-Value Contribution
test
Regression 4 0.014173 0.003543 12.53 0.016
R 1 0.011965 0.011965 42.31 0.003 Significant 84.42108 (1)
Α 1 0.000001 0.000001 0.00 0.957 Not- Significant 0.007056 (4)
Β 1 0.001777 0.001777 6.28 0.066 Not- Significant 12.53792 (2)
R 1 0.000430 0.000430 1.52 0.285 Not- Significant 3.033938 (3)
Error 4 0.001131 0.000283
Total 8 0.015304

Table (11): Analysis of variance (ANOVA) for material removal.

Source DF Adj SS Adj MS F-Value P-Value Contribution value%
Regression 4 0.000378 0.000095 13.73 0.013
R 1 0.000118 0.000118 17.14 0.014 Significant 31.21693 (2)
Α 1 0.000186 0.000186 26.94 0.007 Significant 49.20635 (1)
Β 1 0.000024 0.000024 3.52 0.134 Not- Significant 6.349206 (4)
R 1 0.000050 0.000050 7.31 0.054 Significant 13.22751 (3)
Error 4 0.000028 0.000007
Total 8 0.000406

3. Conclusions (2) The most significant factor that affects the

According to the achieved results in this work, the change in surface roughness (ΔR) is the radius of hole
following conclusions have been obtained. R by (84%).
(1) Successful implementation and design of (3) The optimal parameter for achieving minimum
magnetic inductor and pole that used for flat surfaces surface roughness are radius of hole R (0mm), angle of
such as brass workpiece.

25
 NJES24(1)16-25, 2021
Mahdi & Kadhum

core α1 (82° ), angle of pole β3 (120°), radius of pole [14] Y. M. Baron, S. L. Ko, and J. I. Park,
r2 (0mm), which is (R1, ∝1, β3 , r2 ). “Characterization of the magnetic abrasive
(4) The change in material removal weight (ΔW) has finishing method and its application to deburring,”
many factors that have a significant effect on it is the Key Eng. Mater., vol. 291–292, pp. 291–296, 2005.
angle of core α by (49%) following by radius of hole R [15] F. F. Mustafa, A. H. Kadhym, and H. H. Yahya,
by (31%) and the radius of the pole by (13%). “Tool geometries optimization for friction stir
(5) The optimum parameters that give larger change welding of AA6061-T6 aluminum alloy T-joint
in (ΔW) are radius of hole 𝑅1 (0mm), angle of coreα1 using taguchi method to improve the mechanical
(82° ), angle of pole β1 (60°) and radius of pole behavior,” J. Manuf. Sci. Eng. Trans. ASME, vol.
137, no. 3, pp. 1–8, 2015.
r3 (18mm), which is (R1, ∝1, β1 , r3 ).
[16] D. K. Singh, V. K. Jain, and V. Raghuram,
“Parametric study of magnetic abrasive finishing
4. References process,” J. Mater. Process. Technol., vol. 149, no.
[1] V. Jain, Advanced machining processes. 2009. 1–3, pp. 22–29, 2004.
[2] D. Hayes and D. Wolford, “Art of reducing
magnetostrictive effects in magnetic materials,”
Pat. Trademark
Off., no. U.S. Patent No 2,234,968, 1941.
[3] R. K. Jain and V. K. Jain, “Abrasive Fine Finishing
Processes,” J. Manuf. Sci. Prod., vol. 2, no. 1, pp.
55–68, 2011.
[4] A. a Moosa, “Utilizing a Magnetic Abrasive
Finishing Technique ( MAF ) Via Adaptive Nero
Fuzzy ( ANFIS ),” Am. J. Mater. Eng. Technol.,
vol. 1, no. 3, pp. 49–53, 2013.
[5] M. G. Patil, K. Chandra, and P. S. Misra, “Magnetic
abrasive finishing - A review,” Adv. Mater. Res.,
vol. 418–420, pp. 1577–1581, 2012.
[6] L. D. Yang, C. T. Lin, and H. M. Chow,
“Optimization in MAF operations using Taguchi
parameter design for AISI304 stainless steel,” Int.
J. Adv. Manuf. Technol., vol. 42, no. 5–6, pp. 595–
605, May 2009.
[7] B. Deepak, R. Walia, N. S.-I. J. S. E. M. Technol,
and U. 2012, “Effect of rotational motion on the
flat work piece magnetic abrasive finishing,” vol. 2,
pp. 50–54, 2012.
[8] H. Yamaguchi, A. Srivastava, … M. T.-C. J. of, and
U. 2014, “Magnetic Abrasive Finishing of cutting
tools for high-speed machining of titanium alloys,”
Elsevier, vol. 7, no. 4, pp. 299–304, 2014.
[9] S. M. Mousa, “Optimization the Parameters of
Magnetic Abrasive Process Using Taguchi Method
to Improve the Surface Roughness,” Al-
Khwarizmi Eng. J., vol. 13, no. 2, pp. 11–19, 2017.
[10] A. Chaurasia and V. Wankhede, “Magnetic
Abrasive Finishing of Inconel 718 Super Alloy
Using Permanent Magnet Ankur,” Int. Res. J. Eng.
Technol., vol. 05, no. 05, pp. 1165–1168, 2018.
[11] Y. Tian, C. Shi, Z. Fan, and Q. Zhou,
“Experimental investigations on magnetic abrasive
finishing of Ti-6Al-4V using a multiple pole-tip
finishing tool,” Int. J. Adv. Manuf. Technol., vol.
106, no. 7–8, pp. 3071–3080, 2020.
[12] P. Munyensanga, P. Paryanto, and M. N. A. Aziz,
“Application of Polishing AISI 316L Stainless
Steel Ball Bearing with A Magnetic Abrasive
Finishing Process: A Review,” Rotasi, vol. 20, no.
4, p. 249, 2019.
[13] P. Singh, L. Singh, and A. Kaushik, “Parametric
optimization of magnetic abrasive finishing using
adhesive magnetic abrasive particles,” Int. J. Surf.
Eng. Interdiscip. Mater. Sci., vol. 7, no. 2, pp. 34–
47, 2019.

26