Effects of wire-EDM Parameters On The Surface Integrity and Mechanical Characteristics of Additively Manufactured Inconel 939
Effects of wire-EDM Parameters On The Surface Integrity and Mechanical Characteristics of Additively Manufactured Inconel 939
Effects of wire-EDM Parameters On The Surface Integrity and Mechanical Characteristics of Additively Manufactured Inconel 939
a r t i c l e i n f o a b s t r a c t
Article history: Inconel 939 is nickel-based superalloy which offers remarkable material characteristics at high temper-
Received 7 May 2020 ature applications. It is therefore mainly used in hot section of gas turbine engines parts that require high
Accepted 19 August 2020 temperature corrosion resistance and creep performance. In this research, the W-EDM parameters such
Available online 25 September 2020
as pulse off time and pulse on time are investigated to examine their effects on surface integrity and
mechanical characteristics of additively manufactured Inconel 939 with different building direction.
Keywords: Experimental study is performed to reveal the surface integrity properties including recast layer and sur-
Wire EDM
face roughness in addition to their impacts on the tensile properties of Inconel 939.
Selective laser melting
Surface integrity
Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the International Confer-
ence on Modern Trends in Manufacturing Technologies and Equipment 2020.
https://doi.org/10.1016/j.matpr.2020.08.486
2214-7853/Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the International Conference on Modern Trends in Manufacturing Technologies and Equipment
2020.
O.C. Ozaner et al. Materials Today: Proceedings 38 (2021) 1861–1865
Table 1
The Nominal Chemical Composition of Inconel 939 in wt%
Cr Co C Fe W Mo Nb Ta Ti Al Zr B Ni
22.5 19.0 0.15 – 2.0 – 1.0 1.4 3.7 1.9 0.1 0.01 Bal.
material removal rate, the pulse on and pulse off are considered dimension of 40 16 9 mm are produced and machined by W-
more significant parameters whereas the voltage and pulse on EDM. The experimental studies were performed on Agie Charmilles
have more effects on surface quality [9]. In literature, pulse on CUT300Sp Wire-EDM machine. Thermo Tex copper wire with
and pulse off have been investigated extensively since they are diameter of 0.30 mm was used in the experiments. Afterwards,
considered as substantial parameters affecting the discharged the specimens are etched with Schanz prior to the investigation
energy [4,5,9]. through optical microscope so as to reveal the effects of varying
To the best of author’s knowledge, no studies have been parameters on the thickness recast layer. The recast measurements
reported associated with surface integrity of Inconel 939 processed is conducted based on the continuous thickness identified on the
by means of W-EDM. Therefore, the current research attempts to surface of each sample at multiple locations and average value is
reveal the detailed analysis of surface properties of Inconel 939 calculated. Similarly, the final geometry of dog bone shaped tensile
specimens produced by Laser Bed Powder Fusion and Wire EDM. samples are extracted from 125 26 3 mm prism based on the
Furthermore, the tensile properties of test samples are revealed varying processing parameters and all surfaces are machined by
with respect to Wire EDM parameters. The design of experiment Wire EDM, as seen both in Fig. 1.
is generated to extract the influence of process parameters of Wire Based on the different number of passes, the experimental
EDM on the surface characteristics and the tensile strength of design and process parameters are chosen for sample processing
specimens. in a single pass for Inconel 939. In this way, different energy dis-
charges are achieved in a single pass with the parameters used.
The surface morphology of specimens was examined through Poly-
1.1. Materials and methods
tec TMS-500 profilometer by extracting the three dimensional
topography with a cut-off length of 0.8 mm at multiple locations.
The fabrication of Inconel 939 specimens was conducted using
The area used for 3D surface analysis on gauge length of tensile
EOS M290 Direct Metal Laser Sintering (DMLS) machine equipped
has dimensional area of 22 mm 6 mm. The surface analysis was
with 400 W Yb-fiber laser. The dog bone shaped tensile specimens
conducted prior to tensile tests in order to facilitate the under-
based on the ASTM E8 standard were designed in CAD model and
standing of results with respect to surface properties. Accordingly,
placed on the build plate in both parallel and perpendicular,
tensile tests were conducted using servo-hydraulic machine in
referred as XY and Z samples, respectively. The build platform tem-
ambient air.
perature was pre-heated 80 °C and fabrication was performed
under argon gas atmosphere in order to prevent oxidation. After-
wards, they were cut off the platform using Wire Electrical Dis-
charge Machining. Accordingly, the heat treatment of solution
annealing followed by aging were applied to samples, 4 h at
1160 °C and 16 h at 850 °C, respectively. The chemical composition
of Inconel 939 is demonstrated in Table 1 [10].
The design of experiments is carried out based on Taguchi L18
orthogonal array of W-EDM process parameters as detailed in
Table 2. Four factors with three levels of input parameters, namely
pulse on, pulse off and open voltage are correlated in experimental
design with two levels of building directions, and remaining Fig. 1. The recast measurement samples and tensile geometries processed by W-
parameters are kept constant. The rectangular prism samples with EDM.
Table 2
Experimental results.
Sample No Pulse on (ms) Pulse off (ms) Voltage (V) Recast layer (mm) Surface roughness Sa (mm) Tensile strength (MPa)
and direction
1 XY 0.1 3 80 3,88 2,72 1470
2 XY 0.1 6 120 2,54 3,62 1460
3 XY 0.1 12 200 2,58 3,60 1420
4 XY 0.4 3 80 4,03 4,11 1360
5 XY 0.4 6 120 2,55 4,43 1420
6 XY 0.4 12 200 3,34 3,65 1430
7 XY 0.6 3 120 5,34 8,12 1440
8 XY 0.6 6 200 5,00 4,75 1420
9 XY 0.6 12 80 3,00 4,29 1460
10 Z 0.1 3 200 3,27 4,22 1420
11 Z 0.1 6 80 2,55 2,50 1450
12 Z 0.1 12 120 1,67 2,90 1420
13 Z 0.4 3 120 2,64 3,54 1430
14 Z 0.4 6 200 3,87 4,46 1340
15 Z 0.4 12 80 4,56 3,28 1420
16 Z 0.6 3 200 3,38 5,07 1420
17 Z 0.6 6 80 3,33 4,03 1420
18 Z 0.6 12 120 4,25 4,18 1410
1862
O.C. Ozaner et al. Materials Today: Proceedings 38 (2021) 1861–1865
Fig. 2 shows the optical image of recast layers for some of spec-
imens processed with pre-defined parameters. Additionally, Table 2
presents that varying recast thickness values have been attained
based on the different combination of parameters. The results of
the current work indicate that the variation in recast thickness
on Inconel 939 specimens considerably depends on the machining
parameters. As it can be seen in Fig. 5, mean value of recast thick-
ness noticeably increases with higher pulse on and shows constant
growth in contrast with remaining parameters. Considering the
pulse on, similar results have been obtained previously in litera-
ture [5]. The increasing trend of recast layer with respect to pulse
on can be associated with excessive heat input and eventually the
increased discharge energy, which is calculated using Eq. (1) as
follows;
Z te
1
Ee ¼ V e ðt ÞIe ðt Þdt ¼ V e Ie t on ð1Þ
0 ton þ t off
1863
O.C. Ozaner et al. Materials Today: Proceedings 38 (2021) 1861–1865
Fig. 4. Pulse Off vs recast layer thickness. The ultimate tensile strength of samples are presented in
Table 2. Fig. 5c demonstrates the tensile strength behavior of
Inconel 939 specimens based on building directions along with dif-
and crater diameter can be explained as function of discharge cur-
rent and pulse off time. During EDM process discharge current can
be changed and this may lead unstable cases resulting in formation
of craters.
On the other hand, it has been detected that voltage effects on
recast formation has no consist change as the increasing voltage
leads a slight decrease in recast thickness with 120 V, and eventu-
ally shows an increase with 200 V. The current work utilizes open
voltage, which is a function of distance between the workpiece and
wire. As the open voltage value increases, the distance between
part and wire expands proportionally resulting in larger zone for
the removal of raw material. As a result, reduced recast layer thick-
ness is expected to be achieved with increasing voltage; however,
the recast growth also depends on interaction of parameters each
other. Even if the high voltage is provided for adequate distance,
insufficient pulse off time yields stacking of molten material
between workpiece and wire.
It is noteworthy that EDM has many inputs and the optimum
machining parameters should be determined by considering the
interaction of remaining parameters such as flushing, current and
distance between nozzles. Accordingly, the optimum values are
crucial for these parameters to achieve improved recast layer. As
a result, it is shown that pulse on is more influential input param-
eter as regards the recast thickness since the fluctuations are
noticeably more than that of pulse off and voltage.
1864
O.C. Ozaner et al. Materials Today: Proceedings 38 (2021) 1861–1865
ferent Wire EDM input parameters. Due to the fact that additive Different building direction of Inconel 939 specimens led varia-
manufactured parts indicate anisotropy in grain growth mecha- tion in tensile strength attributing the fact that elongated grains
nism in a dissimilar manner based on building direction, tensile generated during fabrication yields anisotropy in microstruc-
strength of specimens shows noticeable variation as the Z samples ture and mechanical characteristics.
have mean of 1414 MPa whilst the XY samples higher mean value
of 1431 MPa. This variance is primarily attributed to inconsistency
of microstructure generated by elongated grains as a result of dif- Declaration of Competing Interest
ferent building direction [12]. As for the pulse on and pulse off, no
consist change has been observed considering tensile strength of The authors declare that they have no known competing finan-
Inconel 939 specimens, however, the largest variation is noticed cial interests or personal relationships that could have appeared
in varying pulse on value since its range is more than remaining to influence the work reported in this paper.
parameters. On the other hand, increasing voltage up to 200 V
resulted in apparent decrease of mean tensile strength from Acknowledgement
1430 MPa to 1408 MPa. From the Fig. 5b, it is observed that higher
voltage leads the rougher surface texture and eventually resulted Authors would like to express sincere gratitude to TUSAS
in detrimental effect on tensile strength. In the meantime, the Engine Industry (TEI) for generous support to current study.
kinetic energy of ions and electrons, which are essential factors
of erosion between part and wire, increases as a result of more References
intensive open voltage. In other words, more discharge energy is
accumulated in constant distance while increasing the open volt- [1] Philpott, W., Jepson, M. A., & Thomson, R. C. (2016). Comparison of the effects
of a conventional heat treatment between cast and selective laser melted
age and eventually results in intensified material removal. Hence, IN939 alloy.
sub-surface defects as shown in Fig. 2 may occur while intensive [2] P. Sharma, D. Chakradhar, S. Narendranath, Effect of wire diameter on surface
voltage is in use. The sample 10 and sample 16 that machined with integrity of wire electrical discharge machined Inconel 706 for gas turbine
application, J. Manuf. Processes 24 (2016) 170–178.
200 V has low tensile strength due to crack initiation as shown in [3] D. Welling, Results of surface integrity and fatigue study of wire-EDM
Fig. 2. compared to broaching and grinding for demanding jet engine components
made of Inconel 718, Procedia CIRP 13 (2014) 339–344.
[4] L. Li, Y.B. Guo, X.T. Wei, W. Li, Surface integrity characteristics in wire-EDM of
3. Conclusion Inconel 718 at different discharge energy, Procedia CIRP 6 (2013) 220–225.
[5] D. Reddy, H. Soni, S. Narendranath, Experimental Investigation and
In summary, current study accomplishes the experimental Optimization of WEDM process parameters for Ti50Ni48Co2 Shape Memory
Alloy, Mater. Today. Proc. 5 (9) (2018) 19063–19072.
investigation on the surface integrity and mechanical characteris- [6] Y. Tian, D. Tomus, P. Rometsch, X. Wu, Influences of processing parameters on
tics of additive manufactured Inconel 939 specimens processed surface roughness of Hastelloy X produced by selective laser melting, Addit.
by Wire EDM. The findings can be summarized as follows; Manuf. 13 (2017) 103–112.
[7] Majeed, A., Lv, J., Zhang, Y., Muzamil, M., Waqas, A., Shamim, K., ... & Zafar, F.
(2019, January). An investigation into the influence of processing parameters
The surface properties of Inconel 939 are considerably affected on the surface quality of AlSi10Mg parts by SLM process. In 2019 16th
by Wire EDM parameters as the main inputs such as pulse on, international bhurban conference on applied sciences and technology (IBCAST)
(pp. 143-147). IEEE.
pulse off and voltage leads noteworthy change in surface tex- [8] M. Anthony Xavior, Ashwath p., Fatigue life and fracture morphology of
ture that ultimately yields varying mechanical characteristics. Inconel 718 machined by spark EDM process, Procedia Manuf. 30 (2019) 292–
It is observed that pulse on has highest effect on recast layer 299.
[9] M. Subrahmanyam, T. Nancharaiah, Optimization of process parameters in
growth and surface roughness based on main effect plots. wire-cut EDM of Inconel 625 using Taguchi’s approach, Mater. Today. Proc. 23
In addition to examining the Wire EDM parameters individu- (2020) 642–646.
ally, it was found that surface integrity is affected by ratio with [10] T.B. Gibbons, R. Stickler, IN939: metallurgy, properties and performance, in:
High Temperature Alloys for Gas Turbines 1982, Springer, Dordrecht, 1982, pp.
pulse off and pulse on.
369–393.
Recast layer thickness and surface roughness show increasing [11] Anish, Kumar & Kumar, Dr & Kumar, Jatinder. (2011). A review on effect of
trend with higher pulse on values. Additionally, an increase in wedm performance, recast layer and wire accuracy.
thickness of recast layer results in gradually higher surface [12] P. Hanzl, M. Zetek, T. Bakša, T. Kroupa, The influence of processing parameters
on the mechanical properties of SLM parts, Procedia Eng. 100 (2015) 1405–
roughness where lower tensile strength is attained resulted 1413.
from rougher surface texture.
1865