Journal of Engg.

Research Online First Article

Multi-Response Optimization of cutting parameters in MQL assisted

Turning of Haynes 25 alloy with Taguchi based Grey Relational Analysis

DOI : 10.36909/jer.9877

Tushar Dhote1, Rahul A Mali1, Jitendra Katiyar2, T V K Gupta1*

1Department of Mechanical Engineering, Visvesvaraya National Institute of Technology, Nagpur, India.
2 Department of Mechanical Engineering, SRM Institute of Science & Technology, Kattankulathur, Tamil Nadu, India.
*Corresponding Author: [email protected]

ABSTRACT

Haynes 25 is a cobalt based superalloy gaining its importance in aerospace, heat

treatment applications, chemical handling equipment, commercial gas turbine engines, bearing
material, etc. This alloy is featured with low thermal conductivity, wear and corrosion resistance,
strength with good resistance to oxidation at high temperatures. In the present study,
optimization of process parameters in turning Haynes 25 alloy with uncoated and coated carbide
tools under the minimum quantity lubrication (MQL) using Taguchi based grey relational
analysis (GRA) method is attempted. The influence of cutting parameters and nano-particle
concentration on surface roughness, tool wear, cutting and thrust forces are analyzed to improve
the alloy machinability. The work also compares the responses obtained with uncoated and
coated tool inserts and analyzes the effect of nano-particle concentration. Further, the
experimental cutting and thrust forces are computed and validated using FE based DEFORM 3D
software. The results obtained through simulation are in good accordance with experimental data
within an average relative error of about 12 %.

Keywords: Haynes 25, Grey Relational Analysis, Minimum Quantity Lubrication,

DEFORM 3D.

1
Journal of Engg. Research Online First Article

INTRODUCTION

The recent advances in the manufacturing industry have posed variety of challenges to
deliver quality machined parts at reduced cost and increased machinability for difficult to cut
materials. The process efficiency can be improved by reducing the machining time which can
be achieved with high-speed machining. The phenomenon of thermal softening and chemical
stability of tool material occurs in machining hard to cut materials such as superalloys, due to
which the cutting speed is hampered. So, it’s important to understand and investigate the
machinability aspect of hard to cut alloys with appropriate machining parameters and converge
to attain a set of optimal parameters. A good machinability is defined as a combination of cutting
force, tool temperature; surface roughness, tool wear, power consumption, etc., are all to be
low. Optimization of machining parameters improves not only the machining economics but
also product quality. Superalloys are categorized according to the base elements like Nickel
(Ni), Iron (Fe) and Cobalt (Co), etc. and in particular Co-based alloys are gaining importance
due to their tendency to work-harden and maintain higher strength at elevated machining
temperatures. Some of the commonly used Co based superalloys are Stellite, Haynes 188,
Haynes 25, etc.

Machining Haynes 25 alloy with uncoated carbide tools (Sarikaya and Abdulkadir,
2014, 2016) reports that working with cutting speeds beyond 45 m/min decreases the surface
roughness; which is due to excessive tool wear. The alloy when machined under MQL
environment gave better machinability. The optimal cutting parameters were established using
Taguchi’s signal to noise (S/N) ratio analysis, which shows lesser tool wear at high cutting
speeds because of shortened contact duration between tool and work. The optimal parameters
with MQL cooling system (Sarikaya and Abdulkadir, 2015) for achieving flank and notch wear
with surface roughness are investigated at different lubrication flow rates. The effect of dry and
cryogenic cooling on tool wear in machining Co-based alloys is carried through ANOVA to
identify suitable contributing parameters (Sarikaya and Abdulkadir, 2015). The role of MQL
on cutting temperature, chip formation and product quality in turning AISI-1040 steel with
uncoated carbide tools at different cutting speeds and feed was attempted (Dhar et al., 2006).
The authors reported that MQL enables substantial reduction in the cutting temperature,
dimensional inaccuracy depending on the levels of cutting speed and feed rate. In this case, the
chip formation and chip-tool interaction becomes more favorable under MQL environment with
maximum flank wear in machining with cryogenic cooling was 60 % lower than in dry
machining. The cutting tool wear in machining Co-based composite materials (Prasanth et al.,
2018) was found to be optimum at 60 N load, sliding speed of 1 m/s and 1000 m sliding distance
2
Journal of Engg. Research Online First Article

while performing wear analysis. The reports concluded that sliding distance affects the wear
which contributes to 61.05 %. A detailed and cost-effective energy model for a sustainable
machining process in turning AISI52100 steel (Khan et al., 2020) has been developed with
nanofluid assistance after analyzing the viscosity, thermal conductivity and coefficient of
friction. The surface roughness, power and energy consumption, tool life and cost per part are
evaluated and later compared with the present MQL assisted machining. Multiple regression
models were developed (Sharma et al., 2019) in turning with MWCNT hybrid nanoparticle
MQL environment to measure the temperature and tool wear using RSM. The results evidence
an 11 % and 27 % reduction in tool wear and temperature than alumina based lubricants.

The cutting parameters are correlated with surface roughness and forces in machining
AISI 52100 steel with CBN tools using Response surface methodology (RSM) (Bouacha et al.,
2010). The results reported that surface roughness is being affected with feedrate and cutting
speed, while thrust force was sensitive to work hardness and the tool negative rake angle. The
effect of machining parameters and side cutting edge angle on surface roughness and tangential
force in dry turning of AISI 1045 steel with multi-layer tungsten carbide tools (Noordin et al.,
2004) concluded that feedrate has a significant role. The cutting speed doesn’t influence the
cutting forces in turning MDN250 steel (Lalwani et al., 2008), while the thrust forces are
affected by feedrate and depth of cut; also the surface roughness is highly affected by depth of
cut. Machining with CVD and PVD coated tools with TiCN coating to estimate the surface
roughness (Cakir et al., 2009) illustrates that better finish is achieved with PVD coated tools.
The surface roughness increases with feedrate and decreases with nose radius while machining
AISI 410 steel (Ashvin and Nanavati, 2013). A second order model developed to predict the
surface roughness parameters namely, Ra, Rq, and Rms using RSM (Routaraetal., 2012) has
reported that better finish is observed at higher cutting speeds and depth of cut, while the
roughness increases with feedrate.

Today, manufacturing sectors are looking for alternative tools to understand the metal
cutting process for improving quality and machining performance at minimum cost. One such
alternative is the finite element (FE) based machining simulation. These simulations are used
to estimate and predict the chip formation, tool wear, cutting forces, residual stresses,
temperature, etc. in machining. Lagrangian method (Rityuj et al., 2017) in DEFORM 3D
software to predict cutting forces and temperatures is used in turning AISI H13 with ceramic
tool that saved time and consumables without performing the actual experiments. The cutting
forces in turning TC4 alloy (Chen, et al., 2006) are estimated in DEFORM 3D. The

3
Journal of Engg. Research Online First Article

experimental results are validated with FE-based DEFORM 3D simulation in machining

Ti6Al4V alloy (Satyanarayana et al., 2013) and further performed multi-objective optimization
using Genetic Algorithms that have given a relative error of about 9.07 %. The tool geometry
is optimized to achieve minimum flank wear and surface roughness (Tamizharasan et al., 2012).
The FE-based simulation (Rahul Mali et al., 2019) was carried using DEFORM 3D in dry
turning of Aluminum 7075 alloy to validate the estimated forces with experiments.

Not much literature is available in any form to predict cutting forces and temperature in
machining Haynes 25 alloy. The present work is an attempt to study Haynes 25 machining with
minimum quality lubrication (MQL) that can improve the surface quality and tool life. The
Taguchi-Grey relational analysis is used to develop a mathematical model for parametric
optimization and DEFORM 3D for simulating the forces.

MATERIALS, MEASURMENTS, METHODOLOGY

A. MATERIALS
Haynes 25 alloy has good properties like high-temperature strength, resistance to
oxidation, etc. and has variety of applications in aerospace, high temperature applications, etc.
with chemical composition mentioned in Table 1. In the experimentation, a cylindrical rod of
25 mm dia. and 300 mm length is taken as the input raw material and a turning length of 5 mm
is considered for analysis. An MQL system (KENCO make) is used for experimentation. The
high temperatures inducted in hard material machining makes the work-tool material interface
reacts with each other causing high tool wear. The problem of chemical reaction can be avoided
by applying coatings like TiC, Al2O3, TiN, etc. on the insert surface creating a thermal barrier
that impedes the interface reaction which also increases the tool life. In the present work,
CCMT09T308MP, grade WS10PT, PVD coated inserts (WIDIA make) and CCMT09T308
uncoated inserts having two cutting edges are used for experimentation. Later confirmation tests
are performed with uncoated and coated inserts to estimate the improvement in performance.
Table 2 gives the details of cutting inserts that has a rake angle, major edge cutting angle, cutting
edge inclination angle and nose radius as 7o, 75o, 6oand 0.8 mm respectively.

Table 1. Chemical composition of Haynes 25 (Sarikaya, 2015)

Material C Cr Fe Ni P S Si W Co
%weight 0.10 20.30 1.58 10.20 <0.005 0.0008 0.01 14.7 Balance

4
Journal of Engg. Research Online First Article

Table 2. Cutting tool insert details

ISO Catalog Number: Insert Material: Tungsten
CCMT09T308MP Carbide
Insert Shape: Rhombic Coating: PVD
Chip Breaker: UF Grade: WS10PT
Insert Length: 9.70 mm Insert Thickness: 3.97 mm

B. CUTTING FLUID
Coconut oil was chosen as a cutting fluid with added graphite nanopowder in different
wt % (0.25, 0.75 and 1.25) respectively to improve the quality of cut. The particle size is 50-
100 nm (supplied by Nano Research Lab, Bangalore, India) and the homogeneous mixing of
coconut oil and the nanopowders are done by stirring process at 1000 RPM for 30 min. Figure
1 shows the mixture homogeneity as observed using transmission electron microscopy (TEM).

C. MINIMUM QUALITY LUBRICATION

The fabricated cutting fluid is further used in the machining interface by MQL system.
In this present study, the MQL system reservoir has a capacity of 5L from KENCO lubrication.
The input air pressure is maintained within 3 to 5 kg/cm2 with a relief valve and the discharge
of oil and air can be adjusted with a flow adjustment valve which is maintained at 60 ml/hr.

Figure 1. TEM image of coconut oil mixed with 0.75 wt% graphite nanopowder

D. MEASUREMENTS
The cutting forces are measured with Kistler 9257B, a 6-component dynamometer
having maximum measuring capacity of 10kN with 5 pC/N sensitivity. Mitutoyo Surftest SJ-
410 is used for roughness measurement at a speed of 0.5 mm/s. The tool wear is measured with
a USB digital microscope (Model No. UM5-GSAS 225) having a magnification of 1600-1200
pixels.

5
Journal of Engg. Research Online First Article

E. METHODOLOGY
A systematic experimental plan is designed based on Taguchi’s 4 parameters, 3 levels
with L9 orthogonal array and accordingly tests are carried. The responses are analyzed using
Grey relational analysis (GRA) method for process optimization. Speed, feed, depth of cut and
nano-particle concentration are the input parameters while surface roughness (Ra in µm), tool
wear (VBmax in mm), cutting force (Fc in N) and thrust force (Ft in N) are measured responses.
The confirmatory tests are performed with uncoated and coated inserts as per the optimal
parameters. The experimental forces are validated with FE based DEFORM 3D simulation
software.

EXPERIMENTATION, GREY RELATIONAL ANALYSIS, CONFIRMATORY TEST

A. EXPERIMENTATION
All the tests are performed on a conventional lathe machine (Model No. 141, PSG make)
having a maximum spindle speed of 1600 rpm. A total of 18 experiments (9 each for coated and
uncoated inserts) are performed with corresponding levels mentioned in Table 3. Figure 2 shows
the experimental setup while the measurements are summarized in Tables 4, 5.

Table 3. Process parameters and levels

Process parameters, Symbols Levels (1, 2, 3)
Speed (rpm) A 800, 1150, 1600
Feedrate (mm/rev) B 0.1, 0.16, 0.2
Depth of cut (mm) C 0.1, 0.25, 0.5
Nano Particle (h-BN) conc. (%) D 0.25, 0.75, 1.25

Dynamometer

Haynes 25 w/p

Figure 2. Experimental Setup

6
Journal of Engg. Research Online First Article

Table 4. Responses measured with uncoated tool

Test Speed Feedrate DOC Con. Ra VBmax
Fc (N) Ft (N)
No. (rpm) (mm/rev) (mm) (%) (µm) (mm)
1 800 0.10 0.10 0.25 1.738 0.196 125 33
2 800 0.16 0.25 0.75 0.886 0.167 256 65
3 800 0.20 0.50 1.25 1.492 0.190 390 92
4 1150 0.10 0.25 1.25 1.343 0.162 192 62
5 1150 0.16 0.50 0.25 1.120 0.178 434 140
6 1150 0.20 0.10 0.75 0.871 0.167 167 29
7 1600 0.10 0.50 0.75 0.954 0.186 266 88
8 1600 0.16 0.10 1.25 1.454 0.176 115 17
9 1600 0.20 0.25 0.25 0.436 0.194 261 69

Table 5. Responses measured with coated tool

Test Speed Feedrate DOC Con. Ra VBmax
Fc (N) Ft (N)
No. (rpm) (mm/rev) (mm) (%) (µm) (mm)
1 800 0.10 0.10 0.25 0.447 0.120 84 23
2 800 0.16 0.25 0.75 0.790 0.128 235 52
3 800 0.20 0.50 1.25 1.133 0.113 387 90
4 1150 0.10 0.25 1.25 0.367 0.140 168 42
5 1150 0.16 0.50 0.25 0.830 0.155 336 110
6 1150 0.20 0.10 0.75 1.185 0.110 144 22
7 1600 0.10 0.50 0.75 0.368 0.102 247 70
8 1600 0.16 0.10 1.25 0.699 0.091 125 21
9 1600 0.20 0.25 0.25 0.855 0.091 246 45

B. GREY RELATIONAL ANALYSIS

GRA method is used in the present analysis which doesn’t require an equation or a
model correlating input-output parameters to obtain optimum parameters. In GRA, the
responses are normalized between 0 and 1, known as grey relational generation from which the
Grey relational coefficients (GRCs) are calculated to express the relationship between desired
and actual responses. The grey relational grades (GRGs) are computed by averaging the GRCs
of each response. Evaluation of the multi-objective responses is through GRG which is finally
converted into a single grade. In the present study, lower the better criterion has been used to
generate the grades which are calculated as follows:

max yi (k )  yi (k ) Eq. 1
xi (k ) 
max ii (k )  min yi (k )
Where xi(k) is the value after the grey relational generation, min yi(k) is the smallest value of
yi(k) for kth response, and max yi(k) is the largest value of yi(k) for kth response. An ideal

7
 Journal of Engg. Research Online First Article

sequence is [x0(k), (k=1, 2, 3, 4)] for the responses. The GRC  i (k ) is calculated as:

 min   max
 i (k )  Eq. 2
 0i (k )   max

Where,  0i  x0 (k )  xi (k )  difference of the absolute value between x0(k)and xi(k).

The overall GRG is calculated by averaging GRCs corresponding to the selected responses,
with an objective to show the degree of relation between the sequences [x0(k) and xi(k), i = 1 to
9]. The GRG is calculated as

1 n
i    i (k )
n k 1
Eq. 3

Where n= number of responses.

Accordingly, GRCs, GRGs and the ranks are calculated that are mentioned in Tables 6
and 7 for uncoated and coated inserts. The main effect plots are shown in Fig. 3 and 4 as
obtained in MINTAB software. This represents the input-output correlations for uncoated and
coated inserts respectively. The ranks obtained based on GRGs shows that A2B3C1D2 and
A3B2C1D3 are optimum input parameters combination for uncoated and coated inserts.
Table 6. GRC, GRG and Rank (uncoated)
Responses Grey Relational Generation Grey Relation Coefficient
GRG Rank
Ra VBmax Fc Ft Ra VBmax Fc Ft Ra VBmax Fc Ft
1.738 0.196 125 33 0.000 0.000 0.969 0.870 0.333 0.333 0.941 0.794 0.600 6
0.886 0.167 256 65 0.654 0.853 0.558 0.610 0.591 0.773 0.531 0.562 0.614 4
1.492 0.190 390 92 0.189 0.176 0.138 0.390 0.381 0.378 0.367 0.451 0.394 9
1.343 0.162 192 62 0.303 1.000 0.759 0.634 0.418 1.000 0.674 0.577 0.667 3
1.120 0.178 434 140 0.475 0.529 0.000 0.000 0.488 0.515 0.333 0.333 0.417 8
0.871 0.167 167 29 0.666 0.853 0.837 0.902 0.599 0.773 0.754 0.837 0.741 1
0.954 0.186 266 88 0.602 0.294 0.527 0.423 0.557 0.415 0.514 0.464 0.487 7
1.454 0.176 115 17 0.218 0.588 1.000 1.000 0.390 0.548 1.000 1.000 0.735 2
0.436 0.194 261 69 1.000 0.059 0.542 0.577 1.000 0.347 0.522 0.542 0.603 5

8
 Journal of Engg. Research Online First Article

Figure 3. Main effects plot for GRG (uncoated)

Table 7. GRC, GRG and Rank (coated)

Responses Grey Relational Generation Grey Relation Coefficient
GRG Rank
Ra VBmax Fc Ft Ra VBmax Fc Ft Ra VBmax Fc Ft
0.447 0.120 84 23 0.902 0.547 1.000 0.978 0.836 0.525 1.000 0.957 0.829 2
0.790 0.128 235 52 0.483 0.422 0.502 0.652 0.492 0.464 0.501 0.589 0.511 7
1.133 0.113 387 90 0.064 0.656 0.000 0.225 0.348 0.593 0.333 0.392 0.417 8
0.367 0.140 168 42 1.000 0.234 0.723 0.764 1.000 0.395 0.643 0.679 0.679 3
0.830 0.155 336 110 0.434 0.000 0.168 0.000 0.469 0.333 0.375 0.333 0.378 9
1.185 0.110 144 22 0.000 0.703 0.802 0.989 0.333 0.627 0.716 0.978 0.664 5
0.368 0.102 247 70 0.999 0.828 0.462 0.449 0.998 0.744 0.482 0.476 0.675 4
0.699 0.091 125 21 0.594 1.000 0.865 1.000 0.552 1.000 0.787 1.000 0.835 1
0.855 0.091 246 45 0.403 1.000 0.465 0.730 0.456 1.000 0.483 0.650 0.647 6

Figure 4. Main effects plot for GRG (coated)

9
Journal of Engg. Research Online First Article

C. CONFIRMATORY TESTS
Experiments are performed for the optimum input parameters and corresponding
responses are measured for both the cases (uncoated, coated). On identification of the optimal
set, confirmatory tests are performed to predict and verify the performance improvement in the
responses. The estimated GRG, ˆ at optimum process parameters is calculated as:

ˆm i m Eq. 4

Where γm is the total mean GRG, γi is mean GRG at the optimum level. The responses (surface
roughness, flank wear, cutting and thrust forces) measured for the optimized experimental
conditions are mentioned in Table 8. The results show that the confirmation tests are in
agreement with the predicted values showing an improvement of 24.5 % in GRG. This
improvement in the experimental results over the initial design parameters confirms the validity
of Taguchi method coupled with GRA. Similarly, confirmation tests are performed with coated
inserts at optimal conditions to validate the experimental data and the responses are summarized
in Table 9 shows an improvement of 6.4 % in GRG.

Table 8. Result of the confirmation experiment (uncoated)

Initial design Optimal design parameters
Responses
parameters Predicted Experimental
Parameter Level A1B1C1D1 A2B2C1D2 A2B2C1D2
Surface roughness 1.738 0.796
Tool wear 0.196 0.170
Cutting force 125 151
Thrust force 33 28
GRG 0.600 0.751 0.747
Improvement in GRG 24.5 %

Table 9. Result of the confirmation experiment (coated)

Initial design Optimal design parameters
Responses
parameters Predicted Experimental
Parameter Level A1B1C1D1 A3B3C1D2 A3B3C1D2
Surface roughness 0.477 0.548
Tool wear 0.120 0.085
Cutting force 84 114
Thrust force 23 17
GRG 0.829 0.9212 0.882
Improvement in GRG 6.4 %

10
Journal of Engg. Research Online First Article

SIMULATION MODELING

The FE based simulation of cutting process is carried using DEFORM 3D machining

software to estimate cutting and thrust forces. The 3D model of the insert is generated in CATIA
V5 software with dimensioning module, shown in Fig. 5. The CAD model is imported to
DEFORMTM 3D in STEP format for further analysis.

Figure 5. CAD model of tool insert

The work material in metal cutting is subjected to high plastic strains as 1-5; strain rates
of 103 - 106 s-1 and temperatures of the order of 1500°C respectively based on tool and work
combination. The challenging task in metal cutting simulation is access to these parameters in
relation to strength of the deforming material by conventional material tests. The flow stress of
a material is represented with the Johnson-Cook material model in terms of strain, strain rate
and temperature, where the constants can be obtained from the tests carried with Split
Hopkinson Pressure Bar (SHPB) or ballistic impact tests. The constitutive material model given
by Eq (5) is used with Haynes 25 alloy constants mentioned in Table 10 is used as material
model input for simulation.

     T  Tr  
   A  B n  * 1  c ln  * 1   
     T T
m r 
0
Eq. 5
Where:

 : Flow stress  : Equivalent plastic strain

A: Yield stress constant  : Strain rate
B : Strain hardening coefficient  0 : Reference strain rate
n : Strain hardening exponent T : Temperature
C :Strain rate dependence coefficient Tr : Room Temperature
m :Temperature dependence coefficient Tm : Melting Temperature
Table 10. Johnson-Cook model constants for Haynes 25 (Hickman, et al., 2017)

A(MPa) B(MPa) C n m Tr Tm 0
455 2475 0.2375 0.9 0.725 20 1455 0.001

11
Journal of Engg. Research Online First Article

In metal cutting, friction plays an important role by affecting the chip geometry, built
edge formation, cutting temperature and tool wear. In the present work, the frictional force
defined by Coulomb’s law is given in Eqn. (6) and where a shear friction factor of 0.6 is
considered.

f s   t Eq. 6

Simulations are performed in DEFORM 3D with process parameters mentioned in Table

11 along with other essential input variables (speed, feedrate and depth of cut) as per Table 3.
The modulus of elasticity and thermal conductivity of Haynes 25 at different temperatures
mentioned in Table 12 are given as input for simulation. A size ratio of 4 for insert and 7 for
workpiece material are considered. Meshing is done with a tetrahedral element of 25000 mesh
size for both workpiece and insert. In the present simulation, the workpiece is treated as plastic
material and insert as rigid material. For a cutting (turning) length of 5 mm and ensuring proper
contact between the tool and workpiece, simulation steps of 500 are taken for processing. The
flow stresses for different temperatures, strain and strain rates of 0.00001 s-1, 0.001 s-1, 0.1 s-1
are considered separately.

Table 11. Process parameters

Environment temperature (oC) 30
Shear friction coefficient 0.6
Heat transfer coefficient (N/sec/mm/oC) 45
Convection coefficient 0.2
Table 12. Modulus of elasticity and Thermal conductivity for Haynes 25
Modulus of Thermal Conductivity
Temp. (oC)
Elasticity (GPa) (W/m oC)
25 225 10.5
100 222 12.0
200 214 14.0
300 204 15.9
400 197 17.7
500 188 19.5
600 181 21.2
700 174 22.9
800 163 24.5
900 154 26.0
1000 146 27.5

All the simulations of machining with uncoated inserts are carried as per the
experimental runs to analyze the variation in cutting and thrust forces with process parameters.
12
Journal of Engg. Research Online First Article

A sample image of the force simulation (test run 9, uncoated insert) is given in Fig. 6 while the
simulated cutting and thrust forces for test run 2 are plotted in Fig. 7.

Figure 6. Simulated Force from DEFORM 3D

(a) (b)
Figure 7. Simulated (a) Cutting force (b) Thrust force (800 rpm, 0.16 mm/rev, 0.25 mm)

(a) (b)
Figure 8. Comparison of simulated and experimental (a) Cutting force (b) Thrust force

The graphs plotted in Fig. 8 shows the comparison of experimental and simulated forces

13
Journal of Engg. Research Online First Article

and the results are summarized in Table 13 for uncoated inserts machining. An average relative
error of 12.40 % in cutting forces and 11.84 % in thrust forces are obtained.

Table 13. Experimental and simulation forces (uncoated)

Process Parameters Experimental Simulation Relative Error (%)
Exp. No
Speed Feedrate DOC Fc Ft Fc Ft Fc Ft
1 800 0.10 0.10 125 33 132 36 5.60 9.09
2 800 0.16 0.25 256 65 286 76 11.71 16.92
3 800 0.20 0.50 390 92 465 99 19.23 7.60
4 1150 0.10 0.25 192 62 210 72 9.37 16.12
5 1150 0.16 0.50 434 140 488 119 12.44 17.64
6 1150 0.20 0.10 167 29 191 33 14.37 13.79
7 1600 0.10 0.50 266 29 315 91 18.42 3.40
8 1600 0.16 0.10 115 17 117 15 1.73 13.33
9 1600 0.20 0.25 261 69 301 75 18.77 8.69
Average Relative Error 12.40 11.84

Similarly, the simulation studies are also carried for the tests performed with coated
inserts under the same experimental conditions. The simulated cutting and thrust forces for test
run 2 are shown in Fig. 9, while Fig. 10 shows the comparison of simulated and experimental
cutting and thrust forces. In this case, the average relative error is 12.19 % in cutting forces and
6.05 % in thrust forces is observed as mentioned in Table 14.

(a) (b)
Figure 9. Simulated (a) Cutting force (b) Thrust force (800 rpm, 0.16 mm/rev, 0.25 mm)

14
Journal of Engg. Research Online First Article

(a) (b)
Figure 10. Comparison of simulated and experimental (a) Cutting force (b) Thrust force

Table 14. Experimental and simulation forces (coated)

Process Parameters Experimental Simulation Relative Error (%)
Exp. No
Speed Feedrate DOC Fc Ft Fc Ft Fc Ft
1 800 0.10 0.10 84 23 86 26 2.38 13.04
2 800 0.16 0.25 235 52 256 54 8.92 3.84
3 800 0.20 0.50 387 90 460 101 18.86 12.22
4 1150 0.10 0.25 168 42 197 45 17.26 7.14
5 1150 0.16 0.50 336 110 397 111 18.15 0.91
6 1150 0.20 0.10 144 22 164 23 13.88 4.54
7 1600 0.10 0.50 247 70 275 71 11.33 1.42
8 1600 0.16 0.10 125 21 136 22 8.80 4.76
9 1600 0.20 0.25 246 45 271 48 10.16 6.66
Average Relative Error 12.19 6.05

CONCLUSIONS
In the present work, an optimal set of process parameters in turning Haynes 25 alloy
with uncoated and coated carbide inserts under MQL environment are obtained using Grey
relational analysis. Further, the experimental results are compared with the FE based simulated
forces using 3D DEFORM machining software. Some of the specific conclusions are:
 Turning with uncoated inserts gave the parametric combination A2B1C1D3 to be optimum
with 1150 rpm cutting speed, 0.1 mm/rev feed, 0.1 mm depth of cut and 1.25 % nano-
particle concentration. Similarly, in case of tests with coated inserts, the parametric
combination A3B3C1D2 is optimum with 1600 rpm cutting speed, 0.2 mm/rev feedrate, 0.1
mm depth of cut and 0.75 % nano-particle concentration.
 The depth of cut is observed to be a significant parameter for both coated and uncoated
15
Journal of Engg. Research Online First Article

experiments, and the results are satisfactorily validated through confirmation tests. The
experimental data is further computed and compared using DEFORM 3D simulation. These
results are found to be in good agreement with the experimental results with an average
relative error of 12.40 % and 11.84% for uncoated inserts and 12.19 % and 6.05 % for coated
inserts in cutting and thrust forces respectively.
 The other responses like cutting temperature, residual stresses, chip morphology, hardness,
etc., can be validated through simulations. The comparative study considering different
nano-particles (types, shape, blending, concentration, etc.) at varying coolant rate can be
attempted further.

REFERENCES

Sarikaya, M. & Güllü, A. 2014. The analysis of process parameters for turning cobalt-based
superalloy Haynes 25/L 605 using design of experiment. Solid State Phenomena, 220-221:
749-753.
Sarikaya, M., Yılmaz, V. & Güllü, A. 2016. Analysis of cutting parameters and
cooling/lubrication methods for sustainable machining in turning of Haynes 25 superalloy.
Journal of Cleaner Production. 133: 172-181.
Sarikaya, M. & Güllü, A. 2015. Multi-response optimization of minimum quantity lubrication
parameters using Taguchi-based grey relational analysis in turning of difficult-to-cut alloy
Haynes 25. Journal of Cleaner Production.91: 347-357.
Sarikaya, M. & Güllü, A. 2015. Examining of tool wear in cryogenic machining of cobalt-
based Haynes 25 superalloy. World Academy of Science, Engineering and Technology,
International Journal of Chemical, Molecular, Nuclear, Materials and Metallurgical
Engineering. 9(8):984-988.
Nikhil Ranjan Dhar, Islam, M. W., Samnun Islam & M.A.H. Mithu, 2006. The influence
of minimum quantity of lubrication (MQL) on cutting temperature, chip and dimensional
accuracy in turning AISI-1010 steel. Journal of Materials Processing Technology. 171(1):
93-99.
Achuthamenon Sylajakumari, P., Ramakrishnasamy, R. & Palaniappan, G. 2018. Taguchi
Grey Relational Analysis for Multi-Response Optimization of Wear in Co-Continuous
Composite. Materials, 44(3): 580-587.
Aqib Mashood Khan, Munish Kumar Gupta, Hussein Hegab, M Jamil, M Mia, Ning He,
Qinghua Song, Zhanqiang Liu, Caalin I Pruncu, 2020. Energy-based cost integrated
modeling and sustainability assessment of Al-GnP hybrid nanofluid assisted turning of
AISI52100 steel. Journal of Cleaner Production, 257(1): 1-17.
Anuj Kumar Sharma, Jitendra Kumar Katiyar, Shubrajit Bhaumik, Sandipan Roy, 2019.
Influence of alumina/MWCNT hybrid nanoparticle additives on tribological properties of
lubricants in turning operations. Friction, 7: 153-168.

16
Journal of Engg. Research Online First Article

Bouacha, K., Yallese, M. A., Mabrouki, T. & Rigal, J. F. 2010. Statistical analysis of surface
roughness and cutting forces using response surface methodology in hard turning of AISI
52100 bearing steel with CBN tool. International Journal of Refractory Metals and Hard
Materials, 28(3): 349-361.
Noordin, M. Y., Venkatesh, V. C., Sharif, S., Elting, S. & Abdullah, A. 2004. Application
of response surface methodology in describing the performance of coated carbide tools
when turning AISI 1045 steel. Journal of Materials Processing Technology. 145(1): 46-58.
Lalwani, D. I., Mehta, N. K. & Jain, P. K. 2008. Experimental investigations of cutting
parameters influence on cutting forces and surface roughness in finish hard turning of
MDN250 steel. Journal of Materials Processing Technology.206(1-3): 167-179.
Cakir, M. C., Ensarioglu, C. & Demirayak, I. 2009. Mathematical modeling of surface
roughness for evaluating the effects of cutting parameters and coating material. Journal of
Materials Processing Technology. 209(1): 102-109.
Ashvin J. Makadia & Nanavati, J. I. 2013. Optimization of machining parameters for turning
operations based on response surface methodology. Measurement, 46(4): 1521-1529.
Routara, B. C., Sahoo, A. K., Parida, A. K. & Padhi, P. C. 2012. Response surface
methodology and genetic algorithm used to optimize the cutting condition for surface
roughness parameters in CNC turning. Procedia engineering. 38: 1893-1904.
Rityuj Singh Parihar; Sahu, R. K. & Srinivasu, G. 2017. Finite Element Analysis of Cutting
Forces Generated in Turning Process using Deform 3D Software. Materials Today:
Proceedings. 4(8): 8432-8438.
Zhuo Chen, Lu-Fang Qin & Li-Juan Yang. 2006. Cutting Force Simulation of Titanium
based on DEFORM-3D. Proceedings of the 3rd International Conference on Material,
Mechanical and Manufacturing Engineering. 1846-1849.
Satyanarayana, K., Gopal, A. V. & Babu, P. B. 2013. Finite Element Simulation of Cutting
Forces in Turning Ti6Al4V Using DEFORM 3D. In ASME 2013 IMECHE.
V02AT02A083-V02AT02A083.
Tamizharasan, T. & Kumar, N. S. 2012. Optimization of cutting insert geometry using
DEFORM-3D: Numerical simulation and experimental validation. International Journal of
Simulation Modelling. 11(2): 65-77.
Slunder, C. J. 1963. Short time tensile properties of the Co-20Cr-15W-10Ni cobalt base alloy
(L 605). BATTELLE MEMORIAL INST COLUMBUS OH DEFENCE METALS
INFORMATION CENTER. 56-67.
Rahul A. Mali, Mata Dayal Agrahari & T. V. K. Gupta, 2019. FE based simulation and
experimental validation of forces in dry turning of aluminum 7075. Materials today:
Proceedings.
Hickman, R. J., Wise, J. L., Smith, J. A., Mersch, J. P., Robino, C. V. & Arguello, J. G.
2017. Observation and simulation of motion and deformation for impact loaded metal
cylinders. AIP Conference Proc. 1793, 100018: 1-6.

17