- Email: [email protected]
Effect of wire diameter on surface integrity of wire electrical discharge machined Inconel 706 for gas turbine application
Journal of Manufacturing Processes 24 (2016) 170–178
Contents lists available at ScienceDirect
Journal of Manufacturing Processes journal homepage: www.elsevier.com/locate/manpro
Technical Paper
Effect of wire diameter on surface integrity of wire electrical discharge machined Inconel 706 for gas turbine application Priyaranjan Sharma ∗ , D. Chakradhar, S. Narendranath Department of Mechanical Engg., National Institute of Technology Karnataka, Surathkal 575025, India
a r t i c l e
i n f o
Article history: Received 8 January 2016 Accepted 2 September 2016 Keywords: Inconel 706 WEDM Microhardness Topography Microstructure XRD
a b s t r a c t Inconel 706 superalloy has established itself in the field of gas turbine industry because of its easy fabricability combined with high mechanical strength. Due to its high stress rupture and tensile yield strength, conventional machining of this superalloy exhibits poor surface and low dimensional accuracy of the machined components. It is well known that most of the gas turbine components include complex shaped profile with high precision and hence, wire electrical discharge machining (WEDM) of Inconel 706 has been performed to achieve the feasibility in manufacturing of complex shaped components for gas turbine application. In the current investigation, the effect of wire diameter on WEDM performance characteristics such as cutting speed, surface roughness, surface topography, recast layer formation, microhardness, microstructural and metallurgical changes have been evaluated. It was investigated that smaller diameter wire is advantageous over the larger diameter wire since it improves productivity as well as surface quality of the machined components under the same settings of control parameters. In addition, smaller diameter wire has shown comparatively lower recast layer thickness, minimum hardness alteration and shorter manufacturing time. The XRD result has confirmed the presence of residual stress within WED machined component. © 2016 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Inconel 706 is a precipitation hardenable nickel-iron-based superalloy that provides high mechanical strength in combination with good fabricability. Owing to its excellent chemistry balance and less prone to segregation, this superalloy can be produced in large diameter and make it ideal for gas turbine applications. With the improvement in gas turbine engines with high firing temperature and compressor ratio, it became necessary to utilize the Inconel 706 for the rotors [1]. Even though, Inconel 706 has exposed the excellent temperature capability to meet the firing temperature requirement, the conventional machining of this superalloy is fairly complex and get work-harden during machining. Usually, the conventional machining of these nickel-iron-based superalloys offer poor machining performance, low dimensional accuracy and poor surface quality of the machined components. These problems are frequently observed due to high work-hardening tendency, chemical affinity, abrasive nature and low thermal conductivity [2,3]. To overcome these issues, non-conventional machining methods such as laser beam machining (LBM), electrochemical
∗ Corresponding author. E-mail address: [email protected] (P. Sharma).
machining (ECM), abrasive water jet machining (AWJM), electrical discharge machining (EDM) are effectively implemented for machining of these superalloys. However, there are certain issues with non-conventional machining processes such as microcracking, poor surface quality, low dimensional accuracy and significant recast layer formation in the LBM process [4,5]; more chance of corrosion due to acidic electrolyte, comparatively low MRR and require special shaped electrode in the ECM process [6]; impingement of abrasive particles into matrix, crack propagation and burr formation at the edge in the AWJM process [7]; less recast layer formation compare to LBM and require special shaped electrode in EDM process [5,8]. Wire electrical discharge machining (WEDM) is an advanced version of EDM which eliminate the need of special shaped electrode and reduce the recast layer thickness (RLT) significantly with the use of low discharge pulse. Additionally, WEDM is more efficient than the EDM in terms of flexibility and offers low residual stresses on the machined component. In the past few years, WEDM allowed success in the production of gas turbine components which required complex shaped profiles with high precision. The high degree of dimensional accuracy and better surface quality of the machined components make WEDM valuable. Welling [9] has shown the good strength of WED machined component compared to the conventional broaching process while
http://dx.doi.org/10.1016/j.jmapro.2016.09.001 1526-6125/© 2016 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
171
Table 1 Chemical composition of Inconel 706 superalloy [23]. Alloy (%) Inconel 706
Min. Max.
Ni + Co
Cr
Fe
Nb + Ta
Ti
Co
C
Mn
Si
S
Cu
Al
P
39 44
14.5 17.5
Bal
2.5 3.3
1.5 2
1
0.06
0.35
0.35
0.02
0.3
0.4
0.02
producing the fir tree slots through Inconel 718. Further, Klocke et al. [10] have explored the WEDM capability using different wire materials. They have reduced the manufacturing time almost onethird using high speed cutting wire. However, the best surface quality was obtained using standard brass wire. Maher et al. [11] have shown the various improvements in wire material properties for its better utilization in WEDM process. Even though, zinc coated wire has shown improved cooling ability and better flushability compared to conventional brass wire, high cost, straightness issues and environmental hazards have been recorded. Antar et al. [12] have attempted to improve the productivity and surface integrity of WED machined Udimet 720 components. They have achieved 40% improved productivity and 25% thinner recast layer using coated wire. The recast layer formation on the WED machined surface is highly detrimental to aerospace applications. Moreover, detail study has been carried out by Newton et al. [13] to investigate the characteristics of recast layer. They exposed that recast material had lower hardness as well as lower elastic modulus compared to base material. Aspinwall et al. [14] have used minimum damage generator technology (MDGT) to minimize the surface damage during the WEDM process of Inconel 718. With the usage of appropriate trim pass strategy, recast layer became nearly invisible. Similarly, Antar et al. [15] have used the MDGT to improve the fatigue performance of WED machined component of Udimet 720. The white layer is also formed on the WED machined surface, generally observed above the recast layer surface. Li et al. [16] have studied the white layer formed on the WED machined surface while cutting Inconel 718. They found that white layer is thicker at high pulse energy, although more discontinuous and non-uniform in nature. But, under a trim cut mode, no white layer was detected in WED machined surface. The microhardness of white layer was reduced significantly under rough cut as well as trim cut due to substantial thermal degradation during the WEDM process. But, with the usage of MDGT, no significant change in subsurface microhardness was observed while cutting Inconel 718 using WEDM process [14]. Atzeni et al. [17] have evaluated the subsurface alteration of Inconel 718 during the WEDM process. But, no thermal modification was observed on machined surface using the proper setting of discharge energy and wire feed rate. For further improvement in WEDM process, process monitoring tool has been developed by Klocke at al. [18] to correlate the surface integrity of the WEDM process while producing the fir tree slot. It is well known that WEDM involves the complex mechanism of material removal. WEDM control parameters have a significant role to improve the productivity and surface features of the machined component. Sharma et al. [19] have studied the effect of electrical/non-electrical control parameters on WEDM performance characteristics while manufacturing the complex profile slots through Inconel 706. It was observed that pulse on time and servo voltage are major factors influencing the material removal rate (MRR), surface roughness (SR), recast layer thickness and microhardness of the machined component. Some researchers have used various optimization techniques to select the optimum combination of control parameters to improve the WEDM efficiency. Ramakrishnan and Karunamoorthy [20] have developed the artificial neural network (ANN) model based on back-propagation algorithms to predict the WEDM control parameters more accurately while machining Inconel 718. Hewidy et al. [21] have developed the mathematical model to determine the
complex inter-relationships between control parameters and performance characteristics of WEDM process during the cutting of Inconel 718. With the use of mathematical model, maximum MRR of 8 mm3 /min and minimum SR of 0.8 m were observed. Even though, earlier literatures have shown the various improvements in WEDM process using modified generator technology, trim pass strategy, different wire material and various optimization approaches, no significant literature has been found by the author regarding the effect of wire diameter on the performance characteristics of WEDM process. Also, the surface integrity of newly developed superalloys is hardly studied. Therefore, the present study has been carried out to investigate the effect of different diameter wires and evaluate the WEDM performance characteristics of Inconel 706 to achieve the feasibility in manufacturing of complex shaped components for gas turbine applications. 2. Materials and methods 2.1. Material selection Inconel 706, which is a recently developed superalloy in preference to Inconel 718 for turbine wheel application, was selected as an objective material. The chemical composition of Inconel 706, which has been verified by Energy dispersive X-ray spectroscopy (EDAX) analysis, was presented in Table 1. The alloy was procured in the form of a plate of dimension 200 mm × 200 mm × 10 mm from ‘Special Metals’, India. The properties of Inconel 706 are more or less similar to Inconel 718 excluding their lower proportion of alloying elements and improved fabricability. Some physical and mechanical properties of Inconel 706 have been listed in Table 2. Molybdenum, which is existing in Inconel 718 as a solid solution strengthener, was excluded in Inconel 706 to increase forgability. Niobium was reduced to decrease the propensity for segregation and freckle formation. The chromium content was selected to achieve good oxidation resistance and low magnetic permeability. The nickel level was selected as low as possible to reduce the cost, however maintain phase stability and avoid formation of sigma phase [22]. Before WEDM operation, specimens were stress relieved to avoid the distortion due to the presence of residual stresses within the material. Therefore, all specimens were heat treated at a temperature of 800 ◦ C in a tubular furnace for 1 h and then, cooled down at room temperature to minimize the residual stresses. 2.2. Experimental setup The experimental work was carried out on ‘ELECTRONICA ECOCUT’ WED machine. The machine has two distinct modes of Table 2 Physical and mechanical properties of Inconel 706 [23]. Properties of Inconel 706
Specification
Density Melting range Thermal conductivity Modulus of elasticity Tensile strength Yield strength (0.2% offset) Elongation
8.05 g/cm3 1334–1371 ◦ C 12.5 W/mK 210 kN/mm2 1282 MPa 993 MPa 19%
172
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Table 3 Experimental settings of control parameters for different discharge energy. Discharge energy
Pulse on time (s)
Pulse off time (s)
Servo voltage (V)
Wire feed (m/min)
Low (0.0353 J) Medium (0.0369 J) High (0.0386 J)
105 110 115
52 39 26
60 40 20
3 6 9
operation, power pulse mode and fine pulse mode. Usually, the power pulse mode is employed for the basic cutting operation, whereas fine pulse mode is used for fine finishing of machined surface. As per our experimental record, the WED machine turns out to be unstable with the usage of smaller diameter wire at low discharge setting while cutting the complex shaped through Inconel 706. Therefore, control parameters and their levels are selected in such a way that there is no wire rupture and gap short issue during the WEDM operation. Based on the preliminary study [19], four control parameters such as servo voltage, pulse on time, pulse off time and wire feed are selected for the current investigation. Mainly, ‘ELECTRONICA ECOCUT’ WED machine are made for the use of 250 m diameter wire. In order to use the 200 m and 150 m diameter wire for the same experimental setup, separate wire guides were procured. For the comparative investigation of different diameter wires, low, medium and high discharge energy settings were selected as shown in Table 3. The discharge energy for each experimental settings has been calculated using Eq. (1). Based on the low and high discharge energy settings, microstructural changes, surface topography, microhardness and recast surface have been evaluated for different diameter wire. The complex shaped profile of Inconel 706, which is manufactured by WEDM, has been shown in Fig. 1. The current study mainly focuses on selection of best diameter wire based on various performance criteria. The different diameter wires are made up of hard brass with elongation of 1%. The 150 m wire has the tensile strength of 900 N/mm2 while 200 m and 250 m wires have tensile strength of 1000 N/mm2 . It must be noted that 150 m, 200 m and 250 m wires are unable to sustains the pulse on time of more than 118 s, 123 s, 127 s respectively with an average setting of other control parameters. Because of low tensile strength of 150 m wire, there is more propensity of wire breakage at pulse on time more than 118 s. In order to maintain the WED machine stability, some control parameters (Table 4) were kept constant throughout the experimental work.
2.3. Measurement of performance characteristics In the current study, discharge energy was calculated by estimating the average electrical energy per spark which is transformed into heat and it can be expressed by Eq. (1).
Ee =
te
ue (t) · ie (t) · dt ∼ = Ue · Ie · te
(1)
0
Where, Ee = Discharge energy, Ue = Discharge voltage, Ie = Discharge current, and te = Discharge duration. In this study, it is considered that electrical discharge occurs without any ignition delay, then discharge duration become equal to the pulse duration. The value of pulse duration or pulse on time corresponding to different discharge energy has been given in Table 3. However, discharge current of 12 A and discharge voltage of 28 V were kept constant. Further, numerical value of discharge energy can be computed using Eq. (2). Ee = Ue · Ie · ti = P · ti
(2)
Fig. 1. Complex profile slot of Inconel 706 fabricated by WEDM process. Table 4 Constant process parameters used during the WEDM process. Wire material Dielectric fluid Polarity Peak current Peak voltage Servo feed Flushing pressure Current speed (%) Dwell time Corner control factor
Standard brass De-ionized water Positive 12 A 11 V 15 mm/min 1.96 bar 50 3 s 3
where, P = Power and ti = Pulse duration or pulse on time. Discharge energy per spark = (336 × ti ) J
(3)
For the current study, discharge energy per spark has been considered instead of total discharge energy. Further, the discharge energy per spark can be computed by Eq. (3). The weight of each sample was measured very precisely using digital weighing balance with an accuracy of 0.0001 g. Prior to weight measurement, hot air blower was used to remove the moisture from the samples. The machining time was calculated using a digital stopwatch. The surface roughness (SR) of WED machined components were measured using ‘Mitutoyo SJ-301’ surface roughness tester. The average surface roughness, which is commonly used in manufacturing industries, is considered for the current research work. The topography of the WED machined surface was obtained using ‘LEST OLS4100’ 3D laser microscope. In order to reveal the grain structure of Inconel 706, samples were mirror polished using a sequential grade of SiC papers followed by diamond paste polishing. The polished samples were etched using Marble’s reagent (10 g CuSO4 + 50 ml HCl + 50 ml H2 O + few drops of H2 SO4 ) to expose the grain structure. ‘JEO JSM638OLA’ scanning electron microscope (SEM) was used to obtain the microstructure image at 1500X. The Energy Dispersive Xray spectrography (EDAX) analysis was carried out to study the metallurgical changes on the WED machined surface. ‘PAN Analytical’ XRD machine (wavelength: Cu K-alpha, 1.5405 nm) was used to confirm the presence of residual stresses and crystal size changes on WED machined surface. In order to measure the subsurface microhardness and recast layer thickness, samples were cold mounted using acrylic powder and self curing liquid and then polished. ‘OMNI TECH MVH-S-AUTO’ micro Vickers hardness tester
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
173
Fig. 2. Microstructure and elemental analysis of as-received Inconel 706.
was used to calculate the diagonal length of the indentation. For the micro indentations, 10 kgf load was applied for dwell time of 10 s. The microhardness was calculated using the expression [24] given in Eq. (4). Hv =
2F sin(136◦ /2) F = 1.854 2 d2 d
(4)
where, Hv, Vicker hardness of the sample; d, mean of diagonals of the squared rhombus, and F, applied load. 3. Results and discussion 3.1. Microstructure study of as-received Inconel 706 The microstructure graph of as-received Inconel 706 has been shown in Fig. 2(a) which indicates the fine and stable equi-axed grains. The ‘BIOVIS’ software based on linear intercept method was used to calculate the grain size as per the ASTM E112 standard. The average grain size was within the range of 13.5–14.5 m. The chemical composition of Inconel 706 has been confirmed by EDAX analysis as shown in Fig. 2(b). 3.2. Effect of wire diameter on cutting speed and surface roughness The effect of wire diameter on cutting speed as well as surface roughness has been shown in Fig. 3. From Fig. 3(a), it was observed that with the smaller diameter wire, cutting speed is 20% higher compared to larger diameter wire under similar experimental condition. This behavior could be explained by the fact that with
a smaller wire diameter (150 m), the wire transport speed was relatively higher. This in turn, slightly improves the wire feed compared to the large diameter wire. Therefore, more amount of molten metal splashed through the machining zone and hence leading to higher cutting speed. Generally, higher cutting speed resembles lesser time for cutting the profile through the component. But, the smaller diameter wire will remove comparatively lesser amount of material for the same time interval because spark gap is relatively lower in case of smaller diameter wire. This can be better explained by the experimental results that for the high discharge energy setting, 0.4337 g material has been removed in 717 s in case of smaller diameter wire (150 m) whereas 0.6887 g material has been removed in 824 s in case of larger diameter wire (250 m). It means that MRR is comparatively lower for smaller diameter wire, but cutting speed is higher. Even though, small diameter wire has shown the various improvements in WEDM process such as higher cutting speed, minimum corner radius, minimum spark gap and better surface quality. However, there is a more propensity of wire breakage because of its lower tensile strength of 900 N/mm2 as that of larger diameter wire of 1000 N/mm2 . Earlier literatures have shown the improvement in WEDM process using the coated wire [12] but coated wires are restricted to their minimum size of 200 m. Below 200 m, no coated wires are commercially available because of straightness issue and difficulties in processing of smaller diameter coated wire. Fig. 3(b) indicates that surface roughness (SR) of larger diameter wire is 8% higher compared to smaller diameter wire under similar experimental condition. This behavior can be explained by the fact that with larger diameter wire, the relative wire transport speed is lower. This, in turn, marginally reduces the wire feed and
Fig. 3. Effect of discharge energy setting on: (a) cutting speed; (b) surface roughness.
174
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Fig. 4. Microstructure analysis of wire surface.
the amount of molten metal to be splashed out from the machining zone is considerably reduced. Hence, allowing the formation of micro holes and micro globules on the machined surface leading to slightly higher SR. The similar trend for the variation of surface roughness with the different diameter wires are reported in previous literature [25]. Based on the experimental investigation, it was concluded that the smaller diameter wire is well-suited for cutting the complex shape profile with improved cutting speed as well as the better surface finish and found to be useful to maintain the minimum corner radius. The smaller diameter wire is also the best choice for cutting the profile through thin and delicate specimen with minimum spark gap. 3.3. Microstructure study of wire surface With the usage of high discharge energy setting along with smaller diameter wire, the more frequent wire breakage would take place due to its lower tensile strength of 900 N/mm2 . The experiment results revealed that pulse on time and servo voltage are the major factors contributing to the wire rupture during the WEDM process. In the current investigation (Fig. 4), crater, rupture, recast layer, melted debris and micro holes on the wire surface have been detected while cutting the Inconel 706 during WEDM process. The EDAX analysis of wire surface (Fig. 5) exposed the presence of O, Ni and Cr on the wire surface resulting in alter metallurgical properties of wire surface. Owing to insufficient pulse off time, melted debris may stick to the wire surface and reduces the spark gap thus making the spark unstable. This situation leads to an arcing between wire and workpiece resulting in wire breakage. 3.4. Microstructure study of WED machined surface The microstructure of WED machined surface of Inconel 706 revealed the formation of micro holes, micro globules, craters and melted debris as shown in Fig. 6. But, no micro crack was detected on the machined surface due high toughness of Inconel 706 alloy. From Fig. 6(b), (d) and (f), it was observed that micro voids, craters and micro globules are commonly observed due
to the high discharge energy of the spark. Li et al. [16] have also observed similar results under high discharge energy settings. Our investigation revealed that with the appropriate setting of pulse on time, servo voltage and flushing pressure, the formation of micro voids and micro globules can be reduced significantly. Fig. 6(a), (c) and (e) shows the reduced micro void and micro globules due to the low discharge energy of the spark. Further, the formation of micro holes and craters can be explained by the fact that the electrical spark has the temperature in excess of 10,000 ◦ C which is more than sufficient to melt and vaporize the any conductive material, but not enough to create high exploding pressure which can splash all the melted material from the machining zone. When remaining molten material resolidify on the machined surface, some gas bubbles get entrapped in the melted area and thus, forming micro holes and craters on machined surface. 3.5. Evaluation of surface topography The experimental results revealed that the topography of WED machined surface is slightly varied with the wire diameter. From Fig. 7(a) it was observed that a smaller diameter wire relatively offers fine and smooth surface on the machined components with the use of low discharge energy setting. This behavior can be explained by the fact that the smaller diameter wire generally offers lower SR on the machined component as discussed in the previous section. Therefore, improves the surface features of the WED machined components. From Fig. 7(a), (c) and (e), it was observed that the topography of machined surface is fine textured compare to Fig. 7(b), (d) and (f) because of low discharge energy setting. At higher discharge energy setting, more amount of material melts and resolidify on the machined surface, and allow less time for the splashing of molten metal through machining zone leading to the rough surface of the machined component. 3.6. Evaluation of microhardness Microhardness of machined surface is significantly affected due to sudden heating and cooling during the WEDM process. Our investigation revealed that microhardness of machined surface was
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
175
Fig. 5. EDAX analysis of wire surface.
Fig. 6. Microstructure graph of WED machined surface at: (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
Fig. 7. Surface topography of WED machined surface at: (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
176
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Fig. 8. Subsurface microhardness of WED machined surface at low and high discharge energy followed by different diameter wires.
decreased upto a depth of 80 m due to altered material properties. Due to low carbon content (wt. 0.06%) of Inconel 706, it will not make the WEDM surface harder. In addition, the electrolysis may occur during the WEDM process and also contributes to the softening of machined surface. Fig. 8 shows the microhardness profile for different diameter wires as well as different discharge energy setting. It was observed that with the larger diameter wire, there is a higher tendency of microhardness reduction compared to the smaller diameter wire. The microhardness of machined surface has been decreased to 270.6 Hv due to high thermal gradient on the machined component during WEDM process. Under high discharge energy setting, more reduction in microhardness was observed due to significant thermal degradation that occurred during WEDM process. 3.7. Evaluation of recast surface The recast layer generally formed on the machined surface due to re-solidification of molten material followed by rapid heating
and cooling during the WEDM process. The SEM graph of the crosssection of the WED machined surface revealed that there is a more tendency of recast layer formation at high discharge energy setting accompanied by inadequate flushing conditions. From Fig. 9(a–f), it was revealed that the recast layer thickness (RLT) is slightly larger with larger wire diameter and vice versa. Since, larger wire diameter offers slightly rough surface on the machined component and thus, offers relatively thick recast layer on the machined surface. From Fig. 9(a), (c) and (e), it was observed that RLT has been reduced significantly due low discharge energy of the spark. However, a thick recast layer has been detected under high discharge energy setting as shown in Fig. 9(b), (d) and (f). This is because, under high discharge energy setting, workpiece material is subjected to high amount thermal energy which melts comparatively more volume of material from the machined surface. A part of this molten metal is flushed away by pressurized waves generated during WEDM process. Due to the high temperature of electrical discharge, metallurgical changes occur on the recast surface. The EDAX analysis exposed the depletion of Ni, Fe and Cr, and the addition of Cu, Zn and O to the recast surface as shown in Fig. 10. The microhardness of the recast surface is significantly reduced due to altered material properties.
3.8. XRD analysis Based on the earlier study in Section 3.2, it was observed that smaller diameter wire has shown improved WEDM performance compared to the larger diameter wire. Therefore, smaller diameter wire has chosen for further investigation. Fig. 11 shows the XRD analysis of the WED machined surface of Inconel 706 machined by smaller diameter wire. From Fig. 11, it was observed that the peak intensity of WED machined at high discharge energy has been decreased compared to low discharge energy. It means that crystal size of WED machined surface has been reduced at high discharge energy. Moreover, the XRD peak shift has been observed toward right side when increasing the discharge energy which indicates the presence of residual stresses within the WED machined component.
Fig. 9. Recast surface of WED machined component at (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
177
Fig. 10. EDAX analysis of recast surface.
Fig. 11. XRD analysis of WED machined surface: (a) low discharge energy (150 m); (b) high discharge energy (150 m).
4. Conclusions In the current research work, the effect of wire diameter on WEDM machinability of Inconel 706 superalloy has been investigated. The performance characteristics such as cutting speed, surface roughness, recast layer thickness, microstructure, wire topography, microhardness, WED machined surface topography, metallurgical and elemental changes have been evaluated. Based on the experimental investigation, the conclusions are summarized as follows:
• The smaller diameter wire improves the cutting speed as well as surface finish of the machined component for the same setting of control parameters. That’s because, with the smaller diameter wire, the relative wire transport speed is comparatively higher. This, in turn, increases the splashing of molten material from the machining region leading to improved WEDM performance. With the use of smaller diameter wire, manufacturing time has been reduced to 30% compare to larger diameter wire. • Although, the smaller diameter wire performs better compared to larger diameter wire in terms of minimum corner radius, minimum production time, minimum amplitude of vibrations, minimum spark gap and improved surface quality of the machined parts. But, more frequent wire breakage was observed because of its lower tensile strength. • Pulse on time and servo voltage were found to be major factors which significantly contribute to the wire rupture. The EDAX analysis exposed the presence of Ni, Cr, Cu, Zn and O on the wire
•
•
•
•
•
•
surface and exhibits different wire properties leading to wire rupture. As a result of SEM analysis, micro voids, micro globules, craters and melted debris were observed on the machined surface, but no micro cracks were observed due to high toughness of Inconel 706. The formation of micro voids, craters and micro globules were reduced significantly at low discharge setting. The topography of machined surface revealed that with the usage of smaller diameter wire, more fine and smooth surface can be obtained at low discharge energy setting. However, the topography of machined surface turns out to be more rough at high discharge energy setting. The microhardness of WED machined surface was changed below the depth of 80 m. The smaller diameter wire has shown the minimum hardness alteration of Inconel 706 during WEDM operation. However, more reduction in microhardeness was observed at high discharge energy setting. The recast layer thickness was comparatively lower with the smaller diameter wire and vice versa. But, there is a more tendency of recast layer formation at high discharge energy setting due to significant thermal degradation. The EDAX analysis has exposed the addition of Cu, O and Zn, and depletion of Fe, Ni and Cr from the machined surface and thus exhibits different material properties leading to lower hardness of recast surface. The XRD analysis has shown reduced crystal size of WED machined surface at high discharge energy compared to low discharge energy. Moreover, XRD peak widening and peak toward
178
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
right side has shown the presence of residual stresses within the WED machined material. 5. Future scope The current research work has implied the future scope for the development of smaller diameter coated wire to get the further improvement in the WEDM process in terms of improved tensile strength of wire, minimum wire breakage, higher cutting speed, minimum corner radius, minimum amplitude of vibrations, minimum spark gap and improved surface quality of machined parts while cutting the complex shape with high precision for the specific application. Acknowledgements This work is partially supported by the Department of Science and Technology (DST), Government of India under the project reference number SB/S3/MMER/0067/2013. The authors would like to thank the DST for its funding support. References [1] Muktinutalapati NR. Materials for gas turbines – an overview. INTECH Open Access Publisher; 2011. [2] Arunachalam R, Mannan MA. Machinability of nickel-based high temperature alloys. Mach Sci Technol 2000;4(1):127–68. [3] Ezugwu EO, Bonney J, Yamane Y. An overview of the machinability of aeroengine alloys. J Mater Process Technol 2003;134(2):233–53. [4] Zhong M, Sun H, Liu W, Zhu X, He J. Boundary liquation and interface cracking characterization in laser deposition of Inconel 738 on directionally solidified Ni-based superalloy. Scr Mater 2005;53(2):159–64. [5] Rasheed MS. Comparison of micro-holes produced by micro-EDM with laser machining. Int J Sci Mod Eng 2013;1:3. [6] El-Hofy H. Advanced machining processes: nontraditional and hybrid machining processes. McGraw Hill Professional; 2005. [7] Singh J, Jain SC. Mechanical issues in laser and abrasive water jet cutting. JOM 1995;47(1):28–30. [8] Song KY, Park MS, Chu CN. Electrical discharge machining using a strip electrode. Precis Eng 2013;37(3):738–45.
[9] Welling D. Results of surface integrity and fatigue study of wire-EDM compared to broaching and grinding for demanding jet engine components made of Inconel 718. Proc CIRP 2014;13:339–44. [10] Klocke F, Welling D, Klink A, Veselovac D, Nöthe T, Perez R. Evaluation of advanced wire-EDM capabilities for the manufacture of fir tree slots in Inconel 718. Proc CIRP 2014;14:430–5. [11] Maher I, Sarhan AA, Hamdi M. Review of improvements in wire electrode properties for longer working time and utilization in wire EDM machining. Int J Adv Manuf Technol 2014;76(1–4):329–51. [12] Antar MT, Soo SL, Aspinwall DK, Jones D, Perez R. Productivity and workpiece surface integrity when WEDM aerospace alloys using coated wires. Proc Eng 2011;19:3–8. [13] Newton TR, Melkote SN, Watkins TR, Trejo RM, Reister L. Investigation of the effect of process parameters on the formation and characteristics of recast layer in wire-EDM of Inconel 718. Mater Sci Eng A 2009;513:208–15. [14] Aspinwall DK, Soo SL, Berrisford AE, Walder G. Workpiece surface roughness and integrity after WEDM of Ti–6Al–4V and Inconel 718 using minimum damage generator technology. CIRP Ann – Manuf Technol 2008;57(1):187–90. [15] Antar MT, Soo SL, Aspinwall DK, Sage C, Cuttell M, Perez R, et al. Fatigue response of Udimet 720 following minimum damage wire electrical discharge machining. Mater Des 2012;42:295–300. [16] Li L, Guo YB, Wei XT, Li W. Surface integrity characteristics in wire-EDM of Inconel 718 at different discharge energy. Proc CIRP 2013;6:220–5. [17] Atzeni E, Bassoli E, Gatto A, Iuliano L, Minetola P, Salmi A. Surface and sub surface evaluation in coated-wire electrical discharge machining (WEDM) of INCONEL® alloy 718. Proc CIRP 2014;33:389–94. [18] Klocke F, Welling D, Klink A, Perez R. Quality assessment through in-process monitoring of wire-EDM for fir tree slot production. Proc CIRP 2014;24:97–102. [19] Sharma P, Chakradhar D, Narendranath S. Evaluation of WEDM performance characteristics of Inconel 706 for turbine disk application. Mater Des 2015;88:558–66, http://dx.doi.org/10.1016/j.matdes.2015.09.036. [20] Ramakrishnan R, Karunamoorthy L. Modeling and multi-response optimization of Inconel 718 on machining of CNC WEDM process. J Mater Process Technol 2008;207(1):343–9. [21] Hewidy MS, El-Taweel TA, El-Safty MF. Modelling the machining parameters of wire electrical discharge machining of Inconel 601 using RSM. J Mater Process Technol 2005;169(2):328–36. [22] Schilke PW, Pepe JJ, Schwant RC. Alloy 706 metallurgy and turbine wheel application. Superalloys 1994;718(625,706):1. [23] Technical bulletin: Inconel alloy 706 (Publication number SMC-091), Special Metals Corporation. Retrieved from: http://www.specialmetals.com/assets/ documents/alloys/inconel/inconel-alloy-706.pdf. [24] Dieter GE, Bacon D. Mechanical metallurgy, 3. New York: McGraw-Hill; 1986. p. 331–2. [25] Newton TR. Investigation of the effect of process parameters on the formation of recast layer in wire-EDM of Inconel 718, 51(Master’s dissertation). Atlanta, Georgia: Georgia Institute of Technology; 2008.
Contents lists available at ScienceDirect
Journal of Manufacturing Processes journal homepage: www.elsevier.com/locate/manpro
Technical Paper
Effect of wire diameter on surface integrity of wire electrical discharge machined Inconel 706 for gas turbine application Priyaranjan Sharma ∗ , D. Chakradhar, S. Narendranath Department of Mechanical Engg., National Institute of Technology Karnataka, Surathkal 575025, India
a r t i c l e
i n f o
Article history: Received 8 January 2016 Accepted 2 September 2016 Keywords: Inconel 706 WEDM Microhardness Topography Microstructure XRD
a b s t r a c t Inconel 706 superalloy has established itself in the field of gas turbine industry because of its easy fabricability combined with high mechanical strength. Due to its high stress rupture and tensile yield strength, conventional machining of this superalloy exhibits poor surface and low dimensional accuracy of the machined components. It is well known that most of the gas turbine components include complex shaped profile with high precision and hence, wire electrical discharge machining (WEDM) of Inconel 706 has been performed to achieve the feasibility in manufacturing of complex shaped components for gas turbine application. In the current investigation, the effect of wire diameter on WEDM performance characteristics such as cutting speed, surface roughness, surface topography, recast layer formation, microhardness, microstructural and metallurgical changes have been evaluated. It was investigated that smaller diameter wire is advantageous over the larger diameter wire since it improves productivity as well as surface quality of the machined components under the same settings of control parameters. In addition, smaller diameter wire has shown comparatively lower recast layer thickness, minimum hardness alteration and shorter manufacturing time. The XRD result has confirmed the presence of residual stress within WED machined component. © 2016 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Inconel 706 is a precipitation hardenable nickel-iron-based superalloy that provides high mechanical strength in combination with good fabricability. Owing to its excellent chemistry balance and less prone to segregation, this superalloy can be produced in large diameter and make it ideal for gas turbine applications. With the improvement in gas turbine engines with high firing temperature and compressor ratio, it became necessary to utilize the Inconel 706 for the rotors [1]. Even though, Inconel 706 has exposed the excellent temperature capability to meet the firing temperature requirement, the conventional machining of this superalloy is fairly complex and get work-harden during machining. Usually, the conventional machining of these nickel-iron-based superalloys offer poor machining performance, low dimensional accuracy and poor surface quality of the machined components. These problems are frequently observed due to high work-hardening tendency, chemical affinity, abrasive nature and low thermal conductivity [2,3]. To overcome these issues, non-conventional machining methods such as laser beam machining (LBM), electrochemical
∗ Corresponding author. E-mail address: [email protected] (P. Sharma).
machining (ECM), abrasive water jet machining (AWJM), electrical discharge machining (EDM) are effectively implemented for machining of these superalloys. However, there are certain issues with non-conventional machining processes such as microcracking, poor surface quality, low dimensional accuracy and significant recast layer formation in the LBM process [4,5]; more chance of corrosion due to acidic electrolyte, comparatively low MRR and require special shaped electrode in the ECM process [6]; impingement of abrasive particles into matrix, crack propagation and burr formation at the edge in the AWJM process [7]; less recast layer formation compare to LBM and require special shaped electrode in EDM process [5,8]. Wire electrical discharge machining (WEDM) is an advanced version of EDM which eliminate the need of special shaped electrode and reduce the recast layer thickness (RLT) significantly with the use of low discharge pulse. Additionally, WEDM is more efficient than the EDM in terms of flexibility and offers low residual stresses on the machined component. In the past few years, WEDM allowed success in the production of gas turbine components which required complex shaped profiles with high precision. The high degree of dimensional accuracy and better surface quality of the machined components make WEDM valuable. Welling [9] has shown the good strength of WED machined component compared to the conventional broaching process while
http://dx.doi.org/10.1016/j.jmapro.2016.09.001 1526-6125/© 2016 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
171
Table 1 Chemical composition of Inconel 706 superalloy [23]. Alloy (%) Inconel 706
Min. Max.
Ni + Co
Cr
Fe
Nb + Ta
Ti
Co
C
Mn
Si
S
Cu
Al
P
39 44
14.5 17.5
Bal
2.5 3.3
1.5 2
1
0.06
0.35
0.35
0.02
0.3
0.4
0.02
producing the fir tree slots through Inconel 718. Further, Klocke et al. [10] have explored the WEDM capability using different wire materials. They have reduced the manufacturing time almost onethird using high speed cutting wire. However, the best surface quality was obtained using standard brass wire. Maher et al. [11] have shown the various improvements in wire material properties for its better utilization in WEDM process. Even though, zinc coated wire has shown improved cooling ability and better flushability compared to conventional brass wire, high cost, straightness issues and environmental hazards have been recorded. Antar et al. [12] have attempted to improve the productivity and surface integrity of WED machined Udimet 720 components. They have achieved 40% improved productivity and 25% thinner recast layer using coated wire. The recast layer formation on the WED machined surface is highly detrimental to aerospace applications. Moreover, detail study has been carried out by Newton et al. [13] to investigate the characteristics of recast layer. They exposed that recast material had lower hardness as well as lower elastic modulus compared to base material. Aspinwall et al. [14] have used minimum damage generator technology (MDGT) to minimize the surface damage during the WEDM process of Inconel 718. With the usage of appropriate trim pass strategy, recast layer became nearly invisible. Similarly, Antar et al. [15] have used the MDGT to improve the fatigue performance of WED machined component of Udimet 720. The white layer is also formed on the WED machined surface, generally observed above the recast layer surface. Li et al. [16] have studied the white layer formed on the WED machined surface while cutting Inconel 718. They found that white layer is thicker at high pulse energy, although more discontinuous and non-uniform in nature. But, under a trim cut mode, no white layer was detected in WED machined surface. The microhardness of white layer was reduced significantly under rough cut as well as trim cut due to substantial thermal degradation during the WEDM process. But, with the usage of MDGT, no significant change in subsurface microhardness was observed while cutting Inconel 718 using WEDM process [14]. Atzeni et al. [17] have evaluated the subsurface alteration of Inconel 718 during the WEDM process. But, no thermal modification was observed on machined surface using the proper setting of discharge energy and wire feed rate. For further improvement in WEDM process, process monitoring tool has been developed by Klocke at al. [18] to correlate the surface integrity of the WEDM process while producing the fir tree slot. It is well known that WEDM involves the complex mechanism of material removal. WEDM control parameters have a significant role to improve the productivity and surface features of the machined component. Sharma et al. [19] have studied the effect of electrical/non-electrical control parameters on WEDM performance characteristics while manufacturing the complex profile slots through Inconel 706. It was observed that pulse on time and servo voltage are major factors influencing the material removal rate (MRR), surface roughness (SR), recast layer thickness and microhardness of the machined component. Some researchers have used various optimization techniques to select the optimum combination of control parameters to improve the WEDM efficiency. Ramakrishnan and Karunamoorthy [20] have developed the artificial neural network (ANN) model based on back-propagation algorithms to predict the WEDM control parameters more accurately while machining Inconel 718. Hewidy et al. [21] have developed the mathematical model to determine the
complex inter-relationships between control parameters and performance characteristics of WEDM process during the cutting of Inconel 718. With the use of mathematical model, maximum MRR of 8 mm3 /min and minimum SR of 0.8 m were observed. Even though, earlier literatures have shown the various improvements in WEDM process using modified generator technology, trim pass strategy, different wire material and various optimization approaches, no significant literature has been found by the author regarding the effect of wire diameter on the performance characteristics of WEDM process. Also, the surface integrity of newly developed superalloys is hardly studied. Therefore, the present study has been carried out to investigate the effect of different diameter wires and evaluate the WEDM performance characteristics of Inconel 706 to achieve the feasibility in manufacturing of complex shaped components for gas turbine applications. 2. Materials and methods 2.1. Material selection Inconel 706, which is a recently developed superalloy in preference to Inconel 718 for turbine wheel application, was selected as an objective material. The chemical composition of Inconel 706, which has been verified by Energy dispersive X-ray spectroscopy (EDAX) analysis, was presented in Table 1. The alloy was procured in the form of a plate of dimension 200 mm × 200 mm × 10 mm from ‘Special Metals’, India. The properties of Inconel 706 are more or less similar to Inconel 718 excluding their lower proportion of alloying elements and improved fabricability. Some physical and mechanical properties of Inconel 706 have been listed in Table 2. Molybdenum, which is existing in Inconel 718 as a solid solution strengthener, was excluded in Inconel 706 to increase forgability. Niobium was reduced to decrease the propensity for segregation and freckle formation. The chromium content was selected to achieve good oxidation resistance and low magnetic permeability. The nickel level was selected as low as possible to reduce the cost, however maintain phase stability and avoid formation of sigma phase [22]. Before WEDM operation, specimens were stress relieved to avoid the distortion due to the presence of residual stresses within the material. Therefore, all specimens were heat treated at a temperature of 800 ◦ C in a tubular furnace for 1 h and then, cooled down at room temperature to minimize the residual stresses. 2.2. Experimental setup The experimental work was carried out on ‘ELECTRONICA ECOCUT’ WED machine. The machine has two distinct modes of Table 2 Physical and mechanical properties of Inconel 706 [23]. Properties of Inconel 706
Specification
Density Melting range Thermal conductivity Modulus of elasticity Tensile strength Yield strength (0.2% offset) Elongation
8.05 g/cm3 1334–1371 ◦ C 12.5 W/mK 210 kN/mm2 1282 MPa 993 MPa 19%
172
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Table 3 Experimental settings of control parameters for different discharge energy. Discharge energy
Pulse on time (s)
Pulse off time (s)
Servo voltage (V)
Wire feed (m/min)
Low (0.0353 J) Medium (0.0369 J) High (0.0386 J)
105 110 115
52 39 26
60 40 20
3 6 9
operation, power pulse mode and fine pulse mode. Usually, the power pulse mode is employed for the basic cutting operation, whereas fine pulse mode is used for fine finishing of machined surface. As per our experimental record, the WED machine turns out to be unstable with the usage of smaller diameter wire at low discharge setting while cutting the complex shaped through Inconel 706. Therefore, control parameters and their levels are selected in such a way that there is no wire rupture and gap short issue during the WEDM operation. Based on the preliminary study [19], four control parameters such as servo voltage, pulse on time, pulse off time and wire feed are selected for the current investigation. Mainly, ‘ELECTRONICA ECOCUT’ WED machine are made for the use of 250 m diameter wire. In order to use the 200 m and 150 m diameter wire for the same experimental setup, separate wire guides were procured. For the comparative investigation of different diameter wires, low, medium and high discharge energy settings were selected as shown in Table 3. The discharge energy for each experimental settings has been calculated using Eq. (1). Based on the low and high discharge energy settings, microstructural changes, surface topography, microhardness and recast surface have been evaluated for different diameter wire. The complex shaped profile of Inconel 706, which is manufactured by WEDM, has been shown in Fig. 1. The current study mainly focuses on selection of best diameter wire based on various performance criteria. The different diameter wires are made up of hard brass with elongation of 1%. The 150 m wire has the tensile strength of 900 N/mm2 while 200 m and 250 m wires have tensile strength of 1000 N/mm2 . It must be noted that 150 m, 200 m and 250 m wires are unable to sustains the pulse on time of more than 118 s, 123 s, 127 s respectively with an average setting of other control parameters. Because of low tensile strength of 150 m wire, there is more propensity of wire breakage at pulse on time more than 118 s. In order to maintain the WED machine stability, some control parameters (Table 4) were kept constant throughout the experimental work.
2.3. Measurement of performance characteristics In the current study, discharge energy was calculated by estimating the average electrical energy per spark which is transformed into heat and it can be expressed by Eq. (1).
Ee =
te
ue (t) · ie (t) · dt ∼ = Ue · Ie · te
(1)
0
Where, Ee = Discharge energy, Ue = Discharge voltage, Ie = Discharge current, and te = Discharge duration. In this study, it is considered that electrical discharge occurs without any ignition delay, then discharge duration become equal to the pulse duration. The value of pulse duration or pulse on time corresponding to different discharge energy has been given in Table 3. However, discharge current of 12 A and discharge voltage of 28 V were kept constant. Further, numerical value of discharge energy can be computed using Eq. (2). Ee = Ue · Ie · ti = P · ti
(2)
Fig. 1. Complex profile slot of Inconel 706 fabricated by WEDM process. Table 4 Constant process parameters used during the WEDM process. Wire material Dielectric fluid Polarity Peak current Peak voltage Servo feed Flushing pressure Current speed (%) Dwell time Corner control factor
Standard brass De-ionized water Positive 12 A 11 V 15 mm/min 1.96 bar 50 3 s 3
where, P = Power and ti = Pulse duration or pulse on time. Discharge energy per spark = (336 × ti ) J
(3)
For the current study, discharge energy per spark has been considered instead of total discharge energy. Further, the discharge energy per spark can be computed by Eq. (3). The weight of each sample was measured very precisely using digital weighing balance with an accuracy of 0.0001 g. Prior to weight measurement, hot air blower was used to remove the moisture from the samples. The machining time was calculated using a digital stopwatch. The surface roughness (SR) of WED machined components were measured using ‘Mitutoyo SJ-301’ surface roughness tester. The average surface roughness, which is commonly used in manufacturing industries, is considered for the current research work. The topography of the WED machined surface was obtained using ‘LEST OLS4100’ 3D laser microscope. In order to reveal the grain structure of Inconel 706, samples were mirror polished using a sequential grade of SiC papers followed by diamond paste polishing. The polished samples were etched using Marble’s reagent (10 g CuSO4 + 50 ml HCl + 50 ml H2 O + few drops of H2 SO4 ) to expose the grain structure. ‘JEO JSM638OLA’ scanning electron microscope (SEM) was used to obtain the microstructure image at 1500X. The Energy Dispersive Xray spectrography (EDAX) analysis was carried out to study the metallurgical changes on the WED machined surface. ‘PAN Analytical’ XRD machine (wavelength: Cu K-alpha, 1.5405 nm) was used to confirm the presence of residual stresses and crystal size changes on WED machined surface. In order to measure the subsurface microhardness and recast layer thickness, samples were cold mounted using acrylic powder and self curing liquid and then polished. ‘OMNI TECH MVH-S-AUTO’ micro Vickers hardness tester
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
173
Fig. 2. Microstructure and elemental analysis of as-received Inconel 706.
was used to calculate the diagonal length of the indentation. For the micro indentations, 10 kgf load was applied for dwell time of 10 s. The microhardness was calculated using the expression [24] given in Eq. (4). Hv =
2F sin(136◦ /2) F = 1.854 2 d2 d
(4)
where, Hv, Vicker hardness of the sample; d, mean of diagonals of the squared rhombus, and F, applied load. 3. Results and discussion 3.1. Microstructure study of as-received Inconel 706 The microstructure graph of as-received Inconel 706 has been shown in Fig. 2(a) which indicates the fine and stable equi-axed grains. The ‘BIOVIS’ software based on linear intercept method was used to calculate the grain size as per the ASTM E112 standard. The average grain size was within the range of 13.5–14.5 m. The chemical composition of Inconel 706 has been confirmed by EDAX analysis as shown in Fig. 2(b). 3.2. Effect of wire diameter on cutting speed and surface roughness The effect of wire diameter on cutting speed as well as surface roughness has been shown in Fig. 3. From Fig. 3(a), it was observed that with the smaller diameter wire, cutting speed is 20% higher compared to larger diameter wire under similar experimental condition. This behavior could be explained by the fact that with
a smaller wire diameter (150 m), the wire transport speed was relatively higher. This in turn, slightly improves the wire feed compared to the large diameter wire. Therefore, more amount of molten metal splashed through the machining zone and hence leading to higher cutting speed. Generally, higher cutting speed resembles lesser time for cutting the profile through the component. But, the smaller diameter wire will remove comparatively lesser amount of material for the same time interval because spark gap is relatively lower in case of smaller diameter wire. This can be better explained by the experimental results that for the high discharge energy setting, 0.4337 g material has been removed in 717 s in case of smaller diameter wire (150 m) whereas 0.6887 g material has been removed in 824 s in case of larger diameter wire (250 m). It means that MRR is comparatively lower for smaller diameter wire, but cutting speed is higher. Even though, small diameter wire has shown the various improvements in WEDM process such as higher cutting speed, minimum corner radius, minimum spark gap and better surface quality. However, there is a more propensity of wire breakage because of its lower tensile strength of 900 N/mm2 as that of larger diameter wire of 1000 N/mm2 . Earlier literatures have shown the improvement in WEDM process using the coated wire [12] but coated wires are restricted to their minimum size of 200 m. Below 200 m, no coated wires are commercially available because of straightness issue and difficulties in processing of smaller diameter coated wire. Fig. 3(b) indicates that surface roughness (SR) of larger diameter wire is 8% higher compared to smaller diameter wire under similar experimental condition. This behavior can be explained by the fact that with larger diameter wire, the relative wire transport speed is lower. This, in turn, marginally reduces the wire feed and
Fig. 3. Effect of discharge energy setting on: (a) cutting speed; (b) surface roughness.
174
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Fig. 4. Microstructure analysis of wire surface.
the amount of molten metal to be splashed out from the machining zone is considerably reduced. Hence, allowing the formation of micro holes and micro globules on the machined surface leading to slightly higher SR. The similar trend for the variation of surface roughness with the different diameter wires are reported in previous literature [25]. Based on the experimental investigation, it was concluded that the smaller diameter wire is well-suited for cutting the complex shape profile with improved cutting speed as well as the better surface finish and found to be useful to maintain the minimum corner radius. The smaller diameter wire is also the best choice for cutting the profile through thin and delicate specimen with minimum spark gap. 3.3. Microstructure study of wire surface With the usage of high discharge energy setting along with smaller diameter wire, the more frequent wire breakage would take place due to its lower tensile strength of 900 N/mm2 . The experiment results revealed that pulse on time and servo voltage are the major factors contributing to the wire rupture during the WEDM process. In the current investigation (Fig. 4), crater, rupture, recast layer, melted debris and micro holes on the wire surface have been detected while cutting the Inconel 706 during WEDM process. The EDAX analysis of wire surface (Fig. 5) exposed the presence of O, Ni and Cr on the wire surface resulting in alter metallurgical properties of wire surface. Owing to insufficient pulse off time, melted debris may stick to the wire surface and reduces the spark gap thus making the spark unstable. This situation leads to an arcing between wire and workpiece resulting in wire breakage. 3.4. Microstructure study of WED machined surface The microstructure of WED machined surface of Inconel 706 revealed the formation of micro holes, micro globules, craters and melted debris as shown in Fig. 6. But, no micro crack was detected on the machined surface due high toughness of Inconel 706 alloy. From Fig. 6(b), (d) and (f), it was observed that micro voids, craters and micro globules are commonly observed due
to the high discharge energy of the spark. Li et al. [16] have also observed similar results under high discharge energy settings. Our investigation revealed that with the appropriate setting of pulse on time, servo voltage and flushing pressure, the formation of micro voids and micro globules can be reduced significantly. Fig. 6(a), (c) and (e) shows the reduced micro void and micro globules due to the low discharge energy of the spark. Further, the formation of micro holes and craters can be explained by the fact that the electrical spark has the temperature in excess of 10,000 ◦ C which is more than sufficient to melt and vaporize the any conductive material, but not enough to create high exploding pressure which can splash all the melted material from the machining zone. When remaining molten material resolidify on the machined surface, some gas bubbles get entrapped in the melted area and thus, forming micro holes and craters on machined surface. 3.5. Evaluation of surface topography The experimental results revealed that the topography of WED machined surface is slightly varied with the wire diameter. From Fig. 7(a) it was observed that a smaller diameter wire relatively offers fine and smooth surface on the machined components with the use of low discharge energy setting. This behavior can be explained by the fact that the smaller diameter wire generally offers lower SR on the machined component as discussed in the previous section. Therefore, improves the surface features of the WED machined components. From Fig. 7(a), (c) and (e), it was observed that the topography of machined surface is fine textured compare to Fig. 7(b), (d) and (f) because of low discharge energy setting. At higher discharge energy setting, more amount of material melts and resolidify on the machined surface, and allow less time for the splashing of molten metal through machining zone leading to the rough surface of the machined component. 3.6. Evaluation of microhardness Microhardness of machined surface is significantly affected due to sudden heating and cooling during the WEDM process. Our investigation revealed that microhardness of machined surface was
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
175
Fig. 5. EDAX analysis of wire surface.
Fig. 6. Microstructure graph of WED machined surface at: (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
Fig. 7. Surface topography of WED machined surface at: (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
176
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
Fig. 8. Subsurface microhardness of WED machined surface at low and high discharge energy followed by different diameter wires.
decreased upto a depth of 80 m due to altered material properties. Due to low carbon content (wt. 0.06%) of Inconel 706, it will not make the WEDM surface harder. In addition, the electrolysis may occur during the WEDM process and also contributes to the softening of machined surface. Fig. 8 shows the microhardness profile for different diameter wires as well as different discharge energy setting. It was observed that with the larger diameter wire, there is a higher tendency of microhardness reduction compared to the smaller diameter wire. The microhardness of machined surface has been decreased to 270.6 Hv due to high thermal gradient on the machined component during WEDM process. Under high discharge energy setting, more reduction in microhardness was observed due to significant thermal degradation that occurred during WEDM process. 3.7. Evaluation of recast surface The recast layer generally formed on the machined surface due to re-solidification of molten material followed by rapid heating
and cooling during the WEDM process. The SEM graph of the crosssection of the WED machined surface revealed that there is a more tendency of recast layer formation at high discharge energy setting accompanied by inadequate flushing conditions. From Fig. 9(a–f), it was revealed that the recast layer thickness (RLT) is slightly larger with larger wire diameter and vice versa. Since, larger wire diameter offers slightly rough surface on the machined component and thus, offers relatively thick recast layer on the machined surface. From Fig. 9(a), (c) and (e), it was observed that RLT has been reduced significantly due low discharge energy of the spark. However, a thick recast layer has been detected under high discharge energy setting as shown in Fig. 9(b), (d) and (f). This is because, under high discharge energy setting, workpiece material is subjected to high amount thermal energy which melts comparatively more volume of material from the machined surface. A part of this molten metal is flushed away by pressurized waves generated during WEDM process. Due to the high temperature of electrical discharge, metallurgical changes occur on the recast surface. The EDAX analysis exposed the depletion of Ni, Fe and Cr, and the addition of Cu, Zn and O to the recast surface as shown in Fig. 10. The microhardness of the recast surface is significantly reduced due to altered material properties.
3.8. XRD analysis Based on the earlier study in Section 3.2, it was observed that smaller diameter wire has shown improved WEDM performance compared to the larger diameter wire. Therefore, smaller diameter wire has chosen for further investigation. Fig. 11 shows the XRD analysis of the WED machined surface of Inconel 706 machined by smaller diameter wire. From Fig. 11, it was observed that the peak intensity of WED machined at high discharge energy has been decreased compared to low discharge energy. It means that crystal size of WED machined surface has been reduced at high discharge energy. Moreover, the XRD peak shift has been observed toward right side when increasing the discharge energy which indicates the presence of residual stresses within the WED machined component.
Fig. 9. Recast surface of WED machined component at (a) low discharge energy (150 m); (b) high discharge energy (150 m); (c) low discharge energy (200 m); (d) high discharge energy (200 m); (e) low discharge energy (250 m); (f) high discharge energy (250 m).
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
177
Fig. 10. EDAX analysis of recast surface.
Fig. 11. XRD analysis of WED machined surface: (a) low discharge energy (150 m); (b) high discharge energy (150 m).
4. Conclusions In the current research work, the effect of wire diameter on WEDM machinability of Inconel 706 superalloy has been investigated. The performance characteristics such as cutting speed, surface roughness, recast layer thickness, microstructure, wire topography, microhardness, WED machined surface topography, metallurgical and elemental changes have been evaluated. Based on the experimental investigation, the conclusions are summarized as follows:
• The smaller diameter wire improves the cutting speed as well as surface finish of the machined component for the same setting of control parameters. That’s because, with the smaller diameter wire, the relative wire transport speed is comparatively higher. This, in turn, increases the splashing of molten material from the machining region leading to improved WEDM performance. With the use of smaller diameter wire, manufacturing time has been reduced to 30% compare to larger diameter wire. • Although, the smaller diameter wire performs better compared to larger diameter wire in terms of minimum corner radius, minimum production time, minimum amplitude of vibrations, minimum spark gap and improved surface quality of the machined parts. But, more frequent wire breakage was observed because of its lower tensile strength. • Pulse on time and servo voltage were found to be major factors which significantly contribute to the wire rupture. The EDAX analysis exposed the presence of Ni, Cr, Cu, Zn and O on the wire
•
•
•
•
•
•
surface and exhibits different wire properties leading to wire rupture. As a result of SEM analysis, micro voids, micro globules, craters and melted debris were observed on the machined surface, but no micro cracks were observed due to high toughness of Inconel 706. The formation of micro voids, craters and micro globules were reduced significantly at low discharge setting. The topography of machined surface revealed that with the usage of smaller diameter wire, more fine and smooth surface can be obtained at low discharge energy setting. However, the topography of machined surface turns out to be more rough at high discharge energy setting. The microhardness of WED machined surface was changed below the depth of 80 m. The smaller diameter wire has shown the minimum hardness alteration of Inconel 706 during WEDM operation. However, more reduction in microhardeness was observed at high discharge energy setting. The recast layer thickness was comparatively lower with the smaller diameter wire and vice versa. But, there is a more tendency of recast layer formation at high discharge energy setting due to significant thermal degradation. The EDAX analysis has exposed the addition of Cu, O and Zn, and depletion of Fe, Ni and Cr from the machined surface and thus exhibits different material properties leading to lower hardness of recast surface. The XRD analysis has shown reduced crystal size of WED machined surface at high discharge energy compared to low discharge energy. Moreover, XRD peak widening and peak toward
178
P. Sharma et al. / Journal of Manufacturing Processes 24 (2016) 170–178
right side has shown the presence of residual stresses within the WED machined material. 5. Future scope The current research work has implied the future scope for the development of smaller diameter coated wire to get the further improvement in the WEDM process in terms of improved tensile strength of wire, minimum wire breakage, higher cutting speed, minimum corner radius, minimum amplitude of vibrations, minimum spark gap and improved surface quality of machined parts while cutting the complex shape with high precision for the specific application. Acknowledgements This work is partially supported by the Department of Science and Technology (DST), Government of India under the project reference number SB/S3/MMER/0067/2013. The authors would like to thank the DST for its funding support. References [1] Muktinutalapati NR. Materials for gas turbines – an overview. INTECH Open Access Publisher; 2011. [2] Arunachalam R, Mannan MA. Machinability of nickel-based high temperature alloys. Mach Sci Technol 2000;4(1):127–68. [3] Ezugwu EO, Bonney J, Yamane Y. An overview of the machinability of aeroengine alloys. J Mater Process Technol 2003;134(2):233–53. [4] Zhong M, Sun H, Liu W, Zhu X, He J. Boundary liquation and interface cracking characterization in laser deposition of Inconel 738 on directionally solidified Ni-based superalloy. Scr Mater 2005;53(2):159–64. [5] Rasheed MS. Comparison of micro-holes produced by micro-EDM with laser machining. Int J Sci Mod Eng 2013;1:3. [6] El-Hofy H. Advanced machining processes: nontraditional and hybrid machining processes. McGraw Hill Professional; 2005. [7] Singh J, Jain SC. Mechanical issues in laser and abrasive water jet cutting. JOM 1995;47(1):28–30. [8] Song KY, Park MS, Chu CN. Electrical discharge machining using a strip electrode. Precis Eng 2013;37(3):738–45.
[9] Welling D. Results of surface integrity and fatigue study of wire-EDM compared to broaching and grinding for demanding jet engine components made of Inconel 718. Proc CIRP 2014;13:339–44. [10] Klocke F, Welling D, Klink A, Veselovac D, Nöthe T, Perez R. Evaluation of advanced wire-EDM capabilities for the manufacture of fir tree slots in Inconel 718. Proc CIRP 2014;14:430–5. [11] Maher I, Sarhan AA, Hamdi M. Review of improvements in wire electrode properties for longer working time and utilization in wire EDM machining. Int J Adv Manuf Technol 2014;76(1–4):329–51. [12] Antar MT, Soo SL, Aspinwall DK, Jones D, Perez R. Productivity and workpiece surface integrity when WEDM aerospace alloys using coated wires. Proc Eng 2011;19:3–8. [13] Newton TR, Melkote SN, Watkins TR, Trejo RM, Reister L. Investigation of the effect of process parameters on the formation and characteristics of recast layer in wire-EDM of Inconel 718. Mater Sci Eng A 2009;513:208–15. [14] Aspinwall DK, Soo SL, Berrisford AE, Walder G. Workpiece surface roughness and integrity after WEDM of Ti–6Al–4V and Inconel 718 using minimum damage generator technology. CIRP Ann – Manuf Technol 2008;57(1):187–90. [15] Antar MT, Soo SL, Aspinwall DK, Sage C, Cuttell M, Perez R, et al. Fatigue response of Udimet 720 following minimum damage wire electrical discharge machining. Mater Des 2012;42:295–300. [16] Li L, Guo YB, Wei XT, Li W. Surface integrity characteristics in wire-EDM of Inconel 718 at different discharge energy. Proc CIRP 2013;6:220–5. [17] Atzeni E, Bassoli E, Gatto A, Iuliano L, Minetola P, Salmi A. Surface and sub surface evaluation in coated-wire electrical discharge machining (WEDM) of INCONEL® alloy 718. Proc CIRP 2014;33:389–94. [18] Klocke F, Welling D, Klink A, Perez R. Quality assessment through in-process monitoring of wire-EDM for fir tree slot production. Proc CIRP 2014;24:97–102. [19] Sharma P, Chakradhar D, Narendranath S. Evaluation of WEDM performance characteristics of Inconel 706 for turbine disk application. Mater Des 2015;88:558–66, http://dx.doi.org/10.1016/j.matdes.2015.09.036. [20] Ramakrishnan R, Karunamoorthy L. Modeling and multi-response optimization of Inconel 718 on machining of CNC WEDM process. J Mater Process Technol 2008;207(1):343–9. [21] Hewidy MS, El-Taweel TA, El-Safty MF. Modelling the machining parameters of wire electrical discharge machining of Inconel 601 using RSM. J Mater Process Technol 2005;169(2):328–36. [22] Schilke PW, Pepe JJ, Schwant RC. Alloy 706 metallurgy and turbine wheel application. Superalloys 1994;718(625,706):1. [23] Technical bulletin: Inconel alloy 706 (Publication number SMC-091), Special Metals Corporation. Retrieved from: http://www.specialmetals.com/assets/ documents/alloys/inconel/inconel-alloy-706.pdf. [24] Dieter GE, Bacon D. Mechanical metallurgy, 3. New York: McGraw-Hill; 1986. p. 331–2. [25] Newton TR. Investigation of the effect of process parameters on the formation of recast layer in wire-EDM of Inconel 718, 51(Master’s dissertation). Atlanta, Georgia: Georgia Institute of Technology; 2008.