- Email: [email protected]
Fundamental Investigation of EDM Plasmas, Part II: Parametric Analysis of Electric Discharges in Gaseous Dielectric Medium
Available online at www.sciencedirect.com
ScienceDirect Procedia CIRP 68 (2018) 336 – 341
19th CIRP Conference on Electro Physical and Chemical Machining, 23-27 April 2018, Bilbao, Spain
Fundamental Investigation of EDM Plasmas, Part II: Parametric Analysis of Electric Discharges in Gaseous Dielectric Medium Macedoa F. T. B.*, Wiessnera M., Bernardellia G. C., Kustera F., Wegenera K. a
Institute of machine tools and manufacturing (IWF), ETH Zurich, Switzerland
* Felipe Tadeu Barata de Macedo. Tel.: +41-0-44-632-5923; fax: +41-0-44-632-1125. E-mail address: [email protected]
Abstract The parametric investigation presented in this paper provides an analysis of dry electrical discharge machining (DEDM) under different processing conditions by advanced plasma diagnostics. Optical emission spectroscopy interpretation supported by a collisional-radiative model indicates that DEDM discharges performed with point-type cathode electrode are dominated by the anode material, independently of the used electrodes materials and anode electrode geometry. Moreover, plasma diagnostics suggests substantial influence of the adopted electric current, anode electrode geometry and electrode materials in several discharge properties, such as electron density, fraction of accelerated electrons, expansion and composition of the plasma. The achieved results support a better understanding of the effects of different working conditions on the DEDM discharges, necessary for future industrial development of this process. © Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2018 2018The The Authors. Published by Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining. Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining
Keywords: dry EDM; optical emission spectroscopy; emission spectra simultion; plasma diagnostics
1. Introduction The part I [1] of this paper presents a comparison between electric discharge plasmas in gaseous and liquid dielectric media by advanced plasma diagnostics. The present part II reports a parametric investigation of dry electrical discharge machining (DEDM) under different working conditions. The studied discharges are generated by applying different electric currents, electrode geometries and materials. A spectroscopic analysis of μ-DEDM discharge plasmas under different processing conditions was published by Kanmani Subbu et al. [2], who infer that DEDM plasmas are cold and less dense than EDM plasmas in liquid dielectric media. Spatially-resolved optical emission spectroscopy of EDM discharge plasmas was performed by Natsu et al. [3], applying the Abel inversion technique. Their results suggest the existence of an electron temperature profile in DEDM plasmas. Spatially-resolved optical emission spectroscopy of EDM discharges was also performed in air as dielectric by Kojima et al. [4], who
present a dependence between the discharge current, crater diameter and plasma thickness, whereas a correlation between the electron temperature and ionization of the plasma was also observed. Electron temperatures reported in the cited investigations are calculated assuming plasmas in local thermodynamic equilibrium (LTE) by the two-line Boltzmann method, described by Griem [5]. Macedo et al. [6] reported that DEDM discharges have properties similar to a specific type of vacuum discharge, the hot anode vacuum arcs (HAVA). Later, the same authors [7] presented collisional-radiative (CR) models as a successful tool for advanced interpretation of optical emission spectroscopy of DEDM discharge plasmas. Despite the efforts made in order to characterize DEDM plasmas, advanced analysis of these plasmas under different processing conditions is still lacking. Knowledge about the DEDM discharges performed under different conditions is crucial for further development of this new manufacturing technology, since the plasma-material interactions govern the workpiece material removal and tool wear. Therefore, an investigation of single DEDM discharges, performed with
2212-8271 © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining doi:10.1016/j.procir.2017.12.074
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
different electric currents, electrodes geometries and materials, is made using advanced plasma diagnostics. Nomenclature Uopen I t ܶ ܧܫ
Open voltage [V] Current [A] Pulse duration [μs] Temperature of electrons [K] Ionization energy [eV]
337
Further descriptions of the EDM machine, high-speed imaging, optical emission spectroscopy and electrical parameter measuring equipment are reported in the part I of this paper [1]. In addition, part I contains also information about the used collisional-radiative model, PrismSPECT [9]. The next two chapters present the plasma diagnostics outcomes and a discussion concerning the effects of the working conditions on the DEDM plasma. 3. Results 3.1. Electric discharge plasma expansion
2. Experimental setup 2.1. High-speed imaging and optical emission spectroscopy Different electrode geometry configurations are used in the present work, highlighted in Fig. 1 as setups 1 and 2, whereas the applied electric discharge conditions are summarized in Tab. 1. Setup 1
Setup 2
Cu point electrode
Cu point electrode
Plasma
Plasma
Al plane electrode
Al point electrode
High-speed imaging measurements show that, for square-shaped current pulses, the DEDM plasma plume reaches its maximum dimension already after the first microseconds of the discharge ignition. Moreover, higher electric currents lead to larger plasma plume dimensions, as shown in Fig. 2. These results are in agreement with the plasma diameter measurements reported by Kojima et al. [4], which shows similar behavior for discharges in air under a plane-to-plane electrode geometry configuration.
Oscilloscope
High-speed camera
Spectrometer
Computer Fig. 1. Optical emission spectroscopy experimental setup Table 1. Electric discharge conditions Variables of the process
Working conditions
Open voltage (Uopen)
250 V
Electric current (I)
10, 20 and 40 A
Electric current pulse shape
Square-shaped pulse; Slope-shaped pulse.
Pulse duration (t)
316 μs
Electrodes geometry
Point-to-plane configuration Point-to-point configuration
Point-type electrode polarity
Positive and negative
Dielectric medium
Air at atmospheric pressure
The single electric discharges are performed in gaps smaller than 5 μm, moving one of the electrodes towards the other with the applied open voltage. The discharges are triggered by a mechanism similar to a vacuum breakdown due to the high electric field formed in the small gap, as explained by Klas et al. [8].
Fig. 2. Dependence of the average plasma expansion on the electric current and electrodes polarity; Square-shaped current pulses.
Electrode materials, geometries and polarities also play an important role in the plasma expansion. Application of a point-type aluminum anode leads to plasma expansions almost two times larger than discharges that apply a planetype aluminum anode. Unlikely, the plasma expansion of discharges that use copper as anode reaches values just slightly different for the different used electrode geometry configurations under the same electrical parameter set. 3.2. Electron temperature and electron beam Optical emission spectra of DEDM discharges performed with a point cathode and a plane-type anode can only be explained assuming two possible hypothesis: the presence of a temperature profile peaking at the plasma center; the formation of an electron beam within the discharge, as presented by Macedo et al. [7]. The part I [1] of this paper reports discharge properties calculated assuming the first hypothesis, of an electron temperature profile peaking at the plasma center in LTE.
338
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
Hot anode vacuum arcs have high electron energy and positive plasma potential in the vicinity of the anode surface due to the concentration of the discharge to a hot anode spot, as shown by Bacon et al. [10]. Moreover, Baalrud et al. [11] infer that the concentration of the discharge to the anode spot can lead to the generation of an electron beam. Since DEDM discharges are similar to a HAVA, the present work estimates DEDM plasma properties considering the second mentioned hypothesis, which assumes the presence of an electron beam within the discharges. The anode fall, which is a high potential gradient in a thin region in front of the anode surface, corresponds to the ionization energy of the neutral gas in the gap, as described by Baalrud et al. [11]. Since DEDM discharges performed with a point-type cathode are dominated by anode metal vapor, the anode fall can be assumed to correspond to the ionization energy of the anode material, IEAluminum = 5.98 eV [12]. Furthermore, a hot spot with diameter Ø 80 μm is assumed to be formed in DEDM discharges, as indicated by emission spectra simulations reported by Macedo et al. [7]. Several plasma properties can be deduced from the fit between observed optical emission spectra and simulated synthetic spectra, as described in detail by Macedo et al. [7]. A fit between an observed DEDM plasma emission spectrum and synthetic spectra is shown in Fig. 3. The synthetic spectra are simulated for an electron temperature ܶ = 13,800 K and an electron beam region with 12% of accelerated electrons within the discharge. The inlet of Fig. 3 shows that Al III lines are uniquely emitted from the electron beam region (green curve), which agrees with the optical emission spectroscopy reported by Bacon et al. [13], and is compatible with formation of a hot anode spot.
smaller for discharges performed with current I = 10 A, if compared with discharges applying I = 20 and 40 A. The electron temperature of the plasma is also calculated by the emission spectra simulations considering different electrode geometries and polarities. Increasing electron temperature is observed when an aluminum point-type cathode is used for the discharges, as emphasized with a round marker in a curve of Fig. 6. In addition, the electron temperature is relatively stable during the time for the other applied electrode geometry configurations and polarities.
Fig. 4. Electron temperature for different currents; Cu point cathode and Al plane-type anode; Square-shaped current pulses.
Al I
Fig. 5. Fraction of accelerated electrons for different currents; Cu point cathode and Al plane-type anode electrodes; Square-shaped current pulses. Al II Al III
Al III Cu I
Experimental spectrum EB spectrum simulation
Al II
Hα
Plume spectrum simulation Summed simulations
Fig. 3. Experimental spectrum, plasma plume and electron beam computed spectra at the 160 μs after ignition of the discharge; Cu point cathode and Al plane-type anode electrodes; Slope-shaped current pulse, I = 20 A.
Assuming the formation of an electron beam within the discharge, emission spectra simulations indicate that the electron temperature does not significantly depend on the applied electric current and just slightly vary during the time, as presented in Fig. 4. Fig. 5 shows that the fraction of accelerated electrons within the plasma is noticeably
Fig. 6. Electron temperature for different electrode polarities and geometries; Square-shaped pulses, I = 20 A.
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
3.3. Electron density and electric current Slope and square-shaped current pulses are applied in the present work in order to analyze their effects on the discharge plasma. Slope-shaped currents are commonly used in EDM to reduce the tool wear, as explained by Tsai et al. [14]. Fig. 7 shows that the electron density of the plasma follows the time behavior of the electric current for the experimental conditions adopted here. The electron density is calculated from the full width of a half maximum (FWHM) of the Hα emission line, according to the theories of Gigosos et al. [15].
339
fraction of cathode material in the plasma reaches its maximum value immediately after the discharge ignition for both electrode polarities, dropping afterwards.
Fig. 9. Al, H and Cu fractions of the plasma composition: (a) Cu point anode and Al point-type cathode; (b) Cu point cathode and Al point-type anode; Square-shaped current pulse, I = 20 A.
Fig. 7. Electron densities obtained from different electric current pulse shapes, Cu point cathode and Al plane-type anode; I = 20 A.
Increasing electron density is detected in discharges performed between copper anode and aluminum cathode point-type electrodes, as highlighted with a round marker in Fig. 8. Moreover, the point-to-plane configuration leads to larger electron densities than the point-to-point one.
Higher content of the cathode material is present in the plasma during the start of the discharge when aluminum is used as cathode, if compared with discharges that apply copper as cathode. Furthermore, a significant fraction of the plasma is composed of hydrogen, which varies between 15% and 30% during the time for both electrode polarities using the point-to-point geometry. The fraction of the copper point-type cathode electrode material in the plasma for discharges made under the pointto-point configuration, presented in Fig. 9b, has a time behavior similar to the point-to-plane one, shown in Fig. 10b. Nonetheless, smaller fraction of copper cathode material is observed under the point-to-plane configuration. Similar results are observed for discharges performed with I = 10 and 40 A under the same working conditions.
Al H Cu
Fig. 10. Al, H and Cu fractions of the plasma composition: (a) Cu point anode and Al plane-type cathode; (b) Cu point cathode and Al plane-type anode; Square-shaped current pulse, I = 20 A [1].
Fig. 8. Plasma electron density observed for different electrode geometries and polarities; Square-shaped current pulses, I = 20 A.
3.4. Plasma composition Emission spectra simulations indicate that DEDM discharges performed between point-type electrodes are dominated by the anode material, as shown in Fig. 9. The
Aluminum and hydrogen fractions of the plasma of discharges performed with point copper cathode and planetype aluminum anode stay stable during the time for I = 10 A; however, the hydrogen fraction rises and aluminum fraction drops for I = 20 and 40 A, as indicated by the arrow in Fig. 11. Furthermore, similar correlation is observed for discharges applying copper as cathode under the point-to-point electrode geometry configuration and electric current I = 20 A. Hydrogen has a high solubility in molten aluminum due to reactions of this metal with moisture of the melting environment, as reported by Lin et al. [16]. Thus, the hydrogen fraction of the plasma is probably originated from the aluminum anode material.
340
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
Fig. 11. Correlation between Al and H fractions of the plasma; Cu point cathode and Al point and plane-type anode; Square-shaped currents.
A correlation between the copper and aluminum fractions of the plasma is also observed in discharges performed with copper anode and aluminum cathode pointtype electrodes. The copper anode fraction of the plasma increases, while the aluminum cathode fraction decreases, as the arrow in Fig. 12 indicates. Moreover, Fig. 12 shows that there is no significant correlation between the copper and aluminum fractions of the plasma for discharges applying copper as cathode and aluminum as anode pointtype electrodes, as well as for both polarities of discharges performed under the point-to-plane electrode configuration.
Fig. 12. Correlation between the Cu and Al fractions of the plasma; Square-shaped currents, I = 20 A.
4. Discussion 4.1. Effects of the electric current The discharge current has effect on several plasma properties, such as fraction of accelerated electrons, electron density, expansion and composition. The fraction of accelerated electrons of discharges performed with electric current I = 10 A is lower than the ones that apply I = 20 and 40 A, as shown in Fig. 5. Since a hot spot diameter Ø 80 μm is assumed for all the emission spectra simulations, the referred lower fraction of accelerated electrons suggest that hot spots of discharges made with electric current I = 10 A have smaller dimensions than with I = 20 and 40 A. An accurate estimation of the hot spot
dimension could be obtained by spatially-resolved optical emission spectroscopy together with the Abel inversion method, as described by Kunze [17]. Fig. 11 presents a correlation between aluminum and hydrogen contents of the plasma. This correlation indicates that the aluminum fraction decreases, while the hydrogen fraction increases during the development of electric discharges performed applying different electric currents with copper point-type cathode electrode. This might be explained by the plasma analysis reported by Kharin [18], who infers that strong anode metal evaporation occurs during the start of a HAVA, followed by an erosion gap enlargement and metal evaporation cooling effect. These phenomena can lead to a drop in the anode spot temperature and metal evaporation, and could be a possible reason for the reduction of the aluminum fraction of the plasma composition during the discharge development; however, additional investigation is still necessary in order to properly clarify those observed plasma composition correlations. 4.2. Effects of the electrodes geometry and polarity Electric discharges performed with point-type cathode are dominated by the anode material, independently of the applied electrode materials and anode electrode geometry. The point-type cathode concentrates the discharge to the anode spot, which becomes an intense source of metal vapor. High fraction of the aluminum point-type cathode is present in the plasma during the beginning of the discharge, which progressively becomes dominated by the copper anode material, as shown in Fig. 12. This can be explained by the characteristics of a HAVA, which starts burning in the cathode material, as described by Boxman et al. [19]. The plasma composition of discharges performed with copper point anode and aluminum plane-type cathode reaches a copper fraction of around 20%. This indicates that an aluminum plane-type cathode does not lead to an anode dominated discharge, since its plasma is mainly formed of components originated from the cathode material, as shown in Fig. 10a. Furthermore, this observation supports the thesis that a point-type cathode is necessary for concentrating the discharge to the anode spot, proposed in Macedo et al. [6]. Application of a point-type aluminum anode leads to discharge plasmas with larger expansion, lower electron density and electron temperature than discharges with a plane-type aluminum anode electrode. The point-type anode allows the discharge to be concentrated to a relatively small region, which increases the anode material erosion and evaporation. Smaller plasma thickness, increasing electron density and temperature, and larger fraction of the cathode material in the plasma of discharges performed with aluminum point-type cathode are observed, when compared with discharges made with copper point-type cathode electrode. These observations agree with the tool wear analysis reported by Tsai et al. [20], who infer that larger fraction of the aluminum point-type cathode is eroded by the discharge
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
than the copper point-type one, since aluminum has lower melting and evaporation points, and lower electrical and thermal conductivities than copper.
References [1]
Wiessner, M., Macedo, F. T. B., Martendal, C. P., Kuster, F., Wegener, K., 2017, Fundamental investigation of EDM plasmas, part I: a comparison between electric discharges in gaseous and liquid dielectric media, Procedia CIRP, Manuscript submitted for publication/.
[2]
Kanmani Subbu, S., Karthikeyan, G., Ramkumar, J., Dhamodaran, S., 2011, Plasma characterization of dry μ-EDM, The International Journal of Advanced Manufacturing Technology, 56/1-4:187-95.
[3]
Natsu, W., Ojima, S., Kobayashi, T., Kunieda, M., 2004, Temperature distribution measurement in EDM arc plasma using spectroscopy, JSME International Journal Series C, 47/1:384-90.
[4]
Kojima, A., Natsu, W., Kunieda, M., 2008, Spectroscopic measurement of arc plasma diameter in EDM, CIRP Annals Manufacturing Technology, 57/1:203-7.
[5]
Griem, H. R., 1997, Principles of Plasma Spectroscopy: Cambridge University Press.
[6]
Macedo, F. T. B., Wiessner, M., Hollenstein, C., Kuster, F., Wegener, K., 2016, Investigation of the Fundamentals of Tool Electrode Wear in Dry EDM, Procedia CIRP, 46/:55-8.
[7]
Macedo, F. T. B., Wiessner, M., Hollenstein, C., Esteves, P. M. B., Wegener, K., 2017, Fundamental investigation of dry electrical discharge machining (DEDM) by optical emission spectroscopy and its numerical interpretation, The International Journal of Advanced Manufacturing Technology, 90/9:3697-709.
[8]
Klas, M., Š, M., Radjenović, B., Radmilović-Radjenović, M., 2011, Experimental and theoretical studies of the breakdown voltage characteristics at micrometre separations in air, EPL (Europhysics Letters), 95/3:35002.
[9]
MacFarlane, J. J., Golovkin, I. E., Wang, P., Woodruff, P. R., Pereyra, N. A., 2007, SPECT3D – A multi-dimensional collisionalradiative code for generating diagnostic signatures based on hydrodynamics and PIC simulation output, High Energy Density Physics, 3/1–2:181-90.
5. Summary The present work investigates DEDM electric discharges under different processing conditions, analyzing the effect of electric current, electrode materials, geometries and polarities on the properties of the plasma. The plasma diagnostics reported here provide important information for better understanding of different working conditions on the DEDM electric discharge physics, necessary for further development of this new manufacturing technology. DEDM plasma temperature is estimated by emission spectra simulations assuming the formation of an electron beam within the discharges. The electron temperature of discharges performed with copper point cathode and aluminum plane-type anode electrodes just slightly varies during the time, and does not significantly depend on the applied electric current. Differently, the electron density follows the time behavior of the applied electric current shape. Furthermore, smaller fractions of accelerated electrons of electric discharges performed with current I = 10 A suggest the formation of smaller hot anode spots than discharges performed with currents I = 20 and 40 A. The adopted electrode materials and geometry configurations can affect several properties of the DEDM plasma. Increasing electron temperature, electron density and anode material fraction of the plasma are detected when an aluminum point cathode is used, whereas these plasma properties stay fairly constant during the time when a copper point-type cathode is applied under the same electrical parameter set. In addition, electric discharges performed under a point-to-plane configuration have higher electron densities than the ones made under a point-to-point electrode configuration. Analysis of the DEDM plasma composition reveal that the application of a point cathode electrode leads to the formation of anode dominated discharges independently of the used electric current, electrode materials and anode geometry, whereas a plane-type cathode electrode does not concentrate the discharge to the anode spot. Moreover, analysis of discharges performed under different electric currents with copper point cathode and aluminum point and plane-type anode electrodes indicate that the fraction of aluminum content of the plasma drops during the time, while the hydrogen fraction increases. Considerably higher fraction of the cathode material is present in the plasma composition of discharges performed with a point aluminum cathode than the ones made with a point-type copper cathode. These results are supported by tool wear observations reported in the literature. Acknowledgements We would like to thank Dr. Christoph Hollenstein for his great collaboration.
341
[10] Bacon, F. M., Watts, H. A., 1975, Vacuum arc anode plasma 2. Collisional-radiative model and comparison with experiment, Journal of Applied Physics, 46/11:4758-66. [11] Baalrud, S., Longmier, B., Hershkowitz, N., 2009, Equilibrium states of anodic double layers, Plasma Sources Science and Technology, 18/3:035002. [12] Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L., et al., 2012, NIST Cloud Computing Reference Architecture: Recommendations of the National Institute of Standards and Technology (Special Publication 500-292): CreateSpace Independent Publishing Platform. [13] Bacon, F. M., 1975, Vacuum arc anode plasma. I. Spectroscopic investigation, Journal of Applied Physics, 46/11:4750-7. [14] Tsai, Y., Lu, C., 2007, Influence of current impulse on machining characteristics in EDM, Journal of mechanical science and technology, 21/10:1617-21. [15] Gigosos, M. A., Cardeñoso, V., 1996, New plasma diagnosis tables of hydrogen Stark broadening including ion dynamics, Journal of Physics B: Atomic, Molecular and Optical Physics, 29/20:4795. [16] Lin, R., Hoch, M., 1989, The solubility of hydrogen in molten aluminum alloys, Metallurgical and Materials Transactions A, 20/9:1785-91. [17] Kunze, H. J., 2009, Introduction to Plasma Spectroscopy: Springer Berlin Heidelberg. [18] Kharin, S., 1997, Mathematical model of the short arc phenomena at the initial stage, Electrical Contacts, 1997, Proceedings of the FortyThird IEEE Holm Conference on:289-305. [19] Boxman, R. L., Sanders, D. M., Martin, P. J., 1995, Handbook of Vacuum Arc Science and Technology: Fundamentals and Applications: Noyes Publications. [20] Tsai, Y.-Y., Masuzawa, T., 2004, An index to evaluate the wear resistance of the electrode in micro-EDM, Journal of Materials Processing Technology, 149/1:304-9.
ScienceDirect Procedia CIRP 68 (2018) 336 – 341
19th CIRP Conference on Electro Physical and Chemical Machining, 23-27 April 2018, Bilbao, Spain
Fundamental Investigation of EDM Plasmas, Part II: Parametric Analysis of Electric Discharges in Gaseous Dielectric Medium Macedoa F. T. B.*, Wiessnera M., Bernardellia G. C., Kustera F., Wegenera K. a
Institute of machine tools and manufacturing (IWF), ETH Zurich, Switzerland
* Felipe Tadeu Barata de Macedo. Tel.: +41-0-44-632-5923; fax: +41-0-44-632-1125. E-mail address: [email protected]
Abstract The parametric investigation presented in this paper provides an analysis of dry electrical discharge machining (DEDM) under different processing conditions by advanced plasma diagnostics. Optical emission spectroscopy interpretation supported by a collisional-radiative model indicates that DEDM discharges performed with point-type cathode electrode are dominated by the anode material, independently of the used electrodes materials and anode electrode geometry. Moreover, plasma diagnostics suggests substantial influence of the adopted electric current, anode electrode geometry and electrode materials in several discharge properties, such as electron density, fraction of accelerated electrons, expansion and composition of the plasma. The achieved results support a better understanding of the effects of different working conditions on the DEDM discharges, necessary for future industrial development of this process. © Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2018 2018The The Authors. Published by Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining. Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining
Keywords: dry EDM; optical emission spectroscopy; emission spectra simultion; plasma diagnostics
1. Introduction The part I [1] of this paper presents a comparison between electric discharge plasmas in gaseous and liquid dielectric media by advanced plasma diagnostics. The present part II reports a parametric investigation of dry electrical discharge machining (DEDM) under different working conditions. The studied discharges are generated by applying different electric currents, electrode geometries and materials. A spectroscopic analysis of μ-DEDM discharge plasmas under different processing conditions was published by Kanmani Subbu et al. [2], who infer that DEDM plasmas are cold and less dense than EDM plasmas in liquid dielectric media. Spatially-resolved optical emission spectroscopy of EDM discharge plasmas was performed by Natsu et al. [3], applying the Abel inversion technique. Their results suggest the existence of an electron temperature profile in DEDM plasmas. Spatially-resolved optical emission spectroscopy of EDM discharges was also performed in air as dielectric by Kojima et al. [4], who
present a dependence between the discharge current, crater diameter and plasma thickness, whereas a correlation between the electron temperature and ionization of the plasma was also observed. Electron temperatures reported in the cited investigations are calculated assuming plasmas in local thermodynamic equilibrium (LTE) by the two-line Boltzmann method, described by Griem [5]. Macedo et al. [6] reported that DEDM discharges have properties similar to a specific type of vacuum discharge, the hot anode vacuum arcs (HAVA). Later, the same authors [7] presented collisional-radiative (CR) models as a successful tool for advanced interpretation of optical emission spectroscopy of DEDM discharge plasmas. Despite the efforts made in order to characterize DEDM plasmas, advanced analysis of these plasmas under different processing conditions is still lacking. Knowledge about the DEDM discharges performed under different conditions is crucial for further development of this new manufacturing technology, since the plasma-material interactions govern the workpiece material removal and tool wear. Therefore, an investigation of single DEDM discharges, performed with
2212-8271 © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 19th CIRP Conference on Electro Physical and Chemical Machining doi:10.1016/j.procir.2017.12.074
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
different electric currents, electrodes geometries and materials, is made using advanced plasma diagnostics. Nomenclature Uopen I t ܶ ܧܫ
Open voltage [V] Current [A] Pulse duration [μs] Temperature of electrons [K] Ionization energy [eV]
337
Further descriptions of the EDM machine, high-speed imaging, optical emission spectroscopy and electrical parameter measuring equipment are reported in the part I of this paper [1]. In addition, part I contains also information about the used collisional-radiative model, PrismSPECT [9]. The next two chapters present the plasma diagnostics outcomes and a discussion concerning the effects of the working conditions on the DEDM plasma. 3. Results 3.1. Electric discharge plasma expansion
2. Experimental setup 2.1. High-speed imaging and optical emission spectroscopy Different electrode geometry configurations are used in the present work, highlighted in Fig. 1 as setups 1 and 2, whereas the applied electric discharge conditions are summarized in Tab. 1. Setup 1
Setup 2
Cu point electrode
Cu point electrode
Plasma
Plasma
Al plane electrode
Al point electrode
High-speed imaging measurements show that, for square-shaped current pulses, the DEDM plasma plume reaches its maximum dimension already after the first microseconds of the discharge ignition. Moreover, higher electric currents lead to larger plasma plume dimensions, as shown in Fig. 2. These results are in agreement with the plasma diameter measurements reported by Kojima et al. [4], which shows similar behavior for discharges in air under a plane-to-plane electrode geometry configuration.
Oscilloscope
High-speed camera
Spectrometer
Computer Fig. 1. Optical emission spectroscopy experimental setup Table 1. Electric discharge conditions Variables of the process
Working conditions
Open voltage (Uopen)
250 V
Electric current (I)
10, 20 and 40 A
Electric current pulse shape
Square-shaped pulse; Slope-shaped pulse.
Pulse duration (t)
316 μs
Electrodes geometry
Point-to-plane configuration Point-to-point configuration
Point-type electrode polarity
Positive and negative
Dielectric medium
Air at atmospheric pressure
The single electric discharges are performed in gaps smaller than 5 μm, moving one of the electrodes towards the other with the applied open voltage. The discharges are triggered by a mechanism similar to a vacuum breakdown due to the high electric field formed in the small gap, as explained by Klas et al. [8].
Fig. 2. Dependence of the average plasma expansion on the electric current and electrodes polarity; Square-shaped current pulses.
Electrode materials, geometries and polarities also play an important role in the plasma expansion. Application of a point-type aluminum anode leads to plasma expansions almost two times larger than discharges that apply a planetype aluminum anode. Unlikely, the plasma expansion of discharges that use copper as anode reaches values just slightly different for the different used electrode geometry configurations under the same electrical parameter set. 3.2. Electron temperature and electron beam Optical emission spectra of DEDM discharges performed with a point cathode and a plane-type anode can only be explained assuming two possible hypothesis: the presence of a temperature profile peaking at the plasma center; the formation of an electron beam within the discharge, as presented by Macedo et al. [7]. The part I [1] of this paper reports discharge properties calculated assuming the first hypothesis, of an electron temperature profile peaking at the plasma center in LTE.
338
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
Hot anode vacuum arcs have high electron energy and positive plasma potential in the vicinity of the anode surface due to the concentration of the discharge to a hot anode spot, as shown by Bacon et al. [10]. Moreover, Baalrud et al. [11] infer that the concentration of the discharge to the anode spot can lead to the generation of an electron beam. Since DEDM discharges are similar to a HAVA, the present work estimates DEDM plasma properties considering the second mentioned hypothesis, which assumes the presence of an electron beam within the discharges. The anode fall, which is a high potential gradient in a thin region in front of the anode surface, corresponds to the ionization energy of the neutral gas in the gap, as described by Baalrud et al. [11]. Since DEDM discharges performed with a point-type cathode are dominated by anode metal vapor, the anode fall can be assumed to correspond to the ionization energy of the anode material, IEAluminum = 5.98 eV [12]. Furthermore, a hot spot with diameter Ø 80 μm is assumed to be formed in DEDM discharges, as indicated by emission spectra simulations reported by Macedo et al. [7]. Several plasma properties can be deduced from the fit between observed optical emission spectra and simulated synthetic spectra, as described in detail by Macedo et al. [7]. A fit between an observed DEDM plasma emission spectrum and synthetic spectra is shown in Fig. 3. The synthetic spectra are simulated for an electron temperature ܶ = 13,800 K and an electron beam region with 12% of accelerated electrons within the discharge. The inlet of Fig. 3 shows that Al III lines are uniquely emitted from the electron beam region (green curve), which agrees with the optical emission spectroscopy reported by Bacon et al. [13], and is compatible with formation of a hot anode spot.
smaller for discharges performed with current I = 10 A, if compared with discharges applying I = 20 and 40 A. The electron temperature of the plasma is also calculated by the emission spectra simulations considering different electrode geometries and polarities. Increasing electron temperature is observed when an aluminum point-type cathode is used for the discharges, as emphasized with a round marker in a curve of Fig. 6. In addition, the electron temperature is relatively stable during the time for the other applied electrode geometry configurations and polarities.
Fig. 4. Electron temperature for different currents; Cu point cathode and Al plane-type anode; Square-shaped current pulses.
Al I
Fig. 5. Fraction of accelerated electrons for different currents; Cu point cathode and Al plane-type anode electrodes; Square-shaped current pulses. Al II Al III
Al III Cu I
Experimental spectrum EB spectrum simulation
Al II
Hα
Plume spectrum simulation Summed simulations
Fig. 3. Experimental spectrum, plasma plume and electron beam computed spectra at the 160 μs after ignition of the discharge; Cu point cathode and Al plane-type anode electrodes; Slope-shaped current pulse, I = 20 A.
Assuming the formation of an electron beam within the discharge, emission spectra simulations indicate that the electron temperature does not significantly depend on the applied electric current and just slightly vary during the time, as presented in Fig. 4. Fig. 5 shows that the fraction of accelerated electrons within the plasma is noticeably
Fig. 6. Electron temperature for different electrode polarities and geometries; Square-shaped pulses, I = 20 A.
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
3.3. Electron density and electric current Slope and square-shaped current pulses are applied in the present work in order to analyze their effects on the discharge plasma. Slope-shaped currents are commonly used in EDM to reduce the tool wear, as explained by Tsai et al. [14]. Fig. 7 shows that the electron density of the plasma follows the time behavior of the electric current for the experimental conditions adopted here. The electron density is calculated from the full width of a half maximum (FWHM) of the Hα emission line, according to the theories of Gigosos et al. [15].
339
fraction of cathode material in the plasma reaches its maximum value immediately after the discharge ignition for both electrode polarities, dropping afterwards.
Fig. 9. Al, H and Cu fractions of the plasma composition: (a) Cu point anode and Al point-type cathode; (b) Cu point cathode and Al point-type anode; Square-shaped current pulse, I = 20 A.
Fig. 7. Electron densities obtained from different electric current pulse shapes, Cu point cathode and Al plane-type anode; I = 20 A.
Increasing electron density is detected in discharges performed between copper anode and aluminum cathode point-type electrodes, as highlighted with a round marker in Fig. 8. Moreover, the point-to-plane configuration leads to larger electron densities than the point-to-point one.
Higher content of the cathode material is present in the plasma during the start of the discharge when aluminum is used as cathode, if compared with discharges that apply copper as cathode. Furthermore, a significant fraction of the plasma is composed of hydrogen, which varies between 15% and 30% during the time for both electrode polarities using the point-to-point geometry. The fraction of the copper point-type cathode electrode material in the plasma for discharges made under the pointto-point configuration, presented in Fig. 9b, has a time behavior similar to the point-to-plane one, shown in Fig. 10b. Nonetheless, smaller fraction of copper cathode material is observed under the point-to-plane configuration. Similar results are observed for discharges performed with I = 10 and 40 A under the same working conditions.
Al H Cu
Fig. 10. Al, H and Cu fractions of the plasma composition: (a) Cu point anode and Al plane-type cathode; (b) Cu point cathode and Al plane-type anode; Square-shaped current pulse, I = 20 A [1].
Fig. 8. Plasma electron density observed for different electrode geometries and polarities; Square-shaped current pulses, I = 20 A.
3.4. Plasma composition Emission spectra simulations indicate that DEDM discharges performed between point-type electrodes are dominated by the anode material, as shown in Fig. 9. The
Aluminum and hydrogen fractions of the plasma of discharges performed with point copper cathode and planetype aluminum anode stay stable during the time for I = 10 A; however, the hydrogen fraction rises and aluminum fraction drops for I = 20 and 40 A, as indicated by the arrow in Fig. 11. Furthermore, similar correlation is observed for discharges applying copper as cathode under the point-to-point electrode geometry configuration and electric current I = 20 A. Hydrogen has a high solubility in molten aluminum due to reactions of this metal with moisture of the melting environment, as reported by Lin et al. [16]. Thus, the hydrogen fraction of the plasma is probably originated from the aluminum anode material.
340
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
Fig. 11. Correlation between Al and H fractions of the plasma; Cu point cathode and Al point and plane-type anode; Square-shaped currents.
A correlation between the copper and aluminum fractions of the plasma is also observed in discharges performed with copper anode and aluminum cathode pointtype electrodes. The copper anode fraction of the plasma increases, while the aluminum cathode fraction decreases, as the arrow in Fig. 12 indicates. Moreover, Fig. 12 shows that there is no significant correlation between the copper and aluminum fractions of the plasma for discharges applying copper as cathode and aluminum as anode pointtype electrodes, as well as for both polarities of discharges performed under the point-to-plane electrode configuration.
Fig. 12. Correlation between the Cu and Al fractions of the plasma; Square-shaped currents, I = 20 A.
4. Discussion 4.1. Effects of the electric current The discharge current has effect on several plasma properties, such as fraction of accelerated electrons, electron density, expansion and composition. The fraction of accelerated electrons of discharges performed with electric current I = 10 A is lower than the ones that apply I = 20 and 40 A, as shown in Fig. 5. Since a hot spot diameter Ø 80 μm is assumed for all the emission spectra simulations, the referred lower fraction of accelerated electrons suggest that hot spots of discharges made with electric current I = 10 A have smaller dimensions than with I = 20 and 40 A. An accurate estimation of the hot spot
dimension could be obtained by spatially-resolved optical emission spectroscopy together with the Abel inversion method, as described by Kunze [17]. Fig. 11 presents a correlation between aluminum and hydrogen contents of the plasma. This correlation indicates that the aluminum fraction decreases, while the hydrogen fraction increases during the development of electric discharges performed applying different electric currents with copper point-type cathode electrode. This might be explained by the plasma analysis reported by Kharin [18], who infers that strong anode metal evaporation occurs during the start of a HAVA, followed by an erosion gap enlargement and metal evaporation cooling effect. These phenomena can lead to a drop in the anode spot temperature and metal evaporation, and could be a possible reason for the reduction of the aluminum fraction of the plasma composition during the discharge development; however, additional investigation is still necessary in order to properly clarify those observed plasma composition correlations. 4.2. Effects of the electrodes geometry and polarity Electric discharges performed with point-type cathode are dominated by the anode material, independently of the applied electrode materials and anode electrode geometry. The point-type cathode concentrates the discharge to the anode spot, which becomes an intense source of metal vapor. High fraction of the aluminum point-type cathode is present in the plasma during the beginning of the discharge, which progressively becomes dominated by the copper anode material, as shown in Fig. 12. This can be explained by the characteristics of a HAVA, which starts burning in the cathode material, as described by Boxman et al. [19]. The plasma composition of discharges performed with copper point anode and aluminum plane-type cathode reaches a copper fraction of around 20%. This indicates that an aluminum plane-type cathode does not lead to an anode dominated discharge, since its plasma is mainly formed of components originated from the cathode material, as shown in Fig. 10a. Furthermore, this observation supports the thesis that a point-type cathode is necessary for concentrating the discharge to the anode spot, proposed in Macedo et al. [6]. Application of a point-type aluminum anode leads to discharge plasmas with larger expansion, lower electron density and electron temperature than discharges with a plane-type aluminum anode electrode. The point-type anode allows the discharge to be concentrated to a relatively small region, which increases the anode material erosion and evaporation. Smaller plasma thickness, increasing electron density and temperature, and larger fraction of the cathode material in the plasma of discharges performed with aluminum point-type cathode are observed, when compared with discharges made with copper point-type cathode electrode. These observations agree with the tool wear analysis reported by Tsai et al. [20], who infer that larger fraction of the aluminum point-type cathode is eroded by the discharge
Macedo F. T. B. et al. / Procedia CIRP 68 (2018) 336 – 341
than the copper point-type one, since aluminum has lower melting and evaporation points, and lower electrical and thermal conductivities than copper.
References [1]
Wiessner, M., Macedo, F. T. B., Martendal, C. P., Kuster, F., Wegener, K., 2017, Fundamental investigation of EDM plasmas, part I: a comparison between electric discharges in gaseous and liquid dielectric media, Procedia CIRP, Manuscript submitted for publication/.
[2]
Kanmani Subbu, S., Karthikeyan, G., Ramkumar, J., Dhamodaran, S., 2011, Plasma characterization of dry μ-EDM, The International Journal of Advanced Manufacturing Technology, 56/1-4:187-95.
[3]
Natsu, W., Ojima, S., Kobayashi, T., Kunieda, M., 2004, Temperature distribution measurement in EDM arc plasma using spectroscopy, JSME International Journal Series C, 47/1:384-90.
[4]
Kojima, A., Natsu, W., Kunieda, M., 2008, Spectroscopic measurement of arc plasma diameter in EDM, CIRP Annals Manufacturing Technology, 57/1:203-7.
[5]
Griem, H. R., 1997, Principles of Plasma Spectroscopy: Cambridge University Press.
[6]
Macedo, F. T. B., Wiessner, M., Hollenstein, C., Kuster, F., Wegener, K., 2016, Investigation of the Fundamentals of Tool Electrode Wear in Dry EDM, Procedia CIRP, 46/:55-8.
[7]
Macedo, F. T. B., Wiessner, M., Hollenstein, C., Esteves, P. M. B., Wegener, K., 2017, Fundamental investigation of dry electrical discharge machining (DEDM) by optical emission spectroscopy and its numerical interpretation, The International Journal of Advanced Manufacturing Technology, 90/9:3697-709.
[8]
Klas, M., Š, M., Radjenović, B., Radmilović-Radjenović, M., 2011, Experimental and theoretical studies of the breakdown voltage characteristics at micrometre separations in air, EPL (Europhysics Letters), 95/3:35002.
[9]
MacFarlane, J. J., Golovkin, I. E., Wang, P., Woodruff, P. R., Pereyra, N. A., 2007, SPECT3D – A multi-dimensional collisionalradiative code for generating diagnostic signatures based on hydrodynamics and PIC simulation output, High Energy Density Physics, 3/1–2:181-90.
5. Summary The present work investigates DEDM electric discharges under different processing conditions, analyzing the effect of electric current, electrode materials, geometries and polarities on the properties of the plasma. The plasma diagnostics reported here provide important information for better understanding of different working conditions on the DEDM electric discharge physics, necessary for further development of this new manufacturing technology. DEDM plasma temperature is estimated by emission spectra simulations assuming the formation of an electron beam within the discharges. The electron temperature of discharges performed with copper point cathode and aluminum plane-type anode electrodes just slightly varies during the time, and does not significantly depend on the applied electric current. Differently, the electron density follows the time behavior of the applied electric current shape. Furthermore, smaller fractions of accelerated electrons of electric discharges performed with current I = 10 A suggest the formation of smaller hot anode spots than discharges performed with currents I = 20 and 40 A. The adopted electrode materials and geometry configurations can affect several properties of the DEDM plasma. Increasing electron temperature, electron density and anode material fraction of the plasma are detected when an aluminum point cathode is used, whereas these plasma properties stay fairly constant during the time when a copper point-type cathode is applied under the same electrical parameter set. In addition, electric discharges performed under a point-to-plane configuration have higher electron densities than the ones made under a point-to-point electrode configuration. Analysis of the DEDM plasma composition reveal that the application of a point cathode electrode leads to the formation of anode dominated discharges independently of the used electric current, electrode materials and anode geometry, whereas a plane-type cathode electrode does not concentrate the discharge to the anode spot. Moreover, analysis of discharges performed under different electric currents with copper point cathode and aluminum point and plane-type anode electrodes indicate that the fraction of aluminum content of the plasma drops during the time, while the hydrogen fraction increases. Considerably higher fraction of the cathode material is present in the plasma composition of discharges performed with a point aluminum cathode than the ones made with a point-type copper cathode. These results are supported by tool wear observations reported in the literature. Acknowledgements We would like to thank Dr. Christoph Hollenstein for his great collaboration.
341
[10] Bacon, F. M., Watts, H. A., 1975, Vacuum arc anode plasma 2. Collisional-radiative model and comparison with experiment, Journal of Applied Physics, 46/11:4758-66. [11] Baalrud, S., Longmier, B., Hershkowitz, N., 2009, Equilibrium states of anodic double layers, Plasma Sources Science and Technology, 18/3:035002. [12] Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L., et al., 2012, NIST Cloud Computing Reference Architecture: Recommendations of the National Institute of Standards and Technology (Special Publication 500-292): CreateSpace Independent Publishing Platform. [13] Bacon, F. M., 1975, Vacuum arc anode plasma. I. Spectroscopic investigation, Journal of Applied Physics, 46/11:4750-7. [14] Tsai, Y., Lu, C., 2007, Influence of current impulse on machining characteristics in EDM, Journal of mechanical science and technology, 21/10:1617-21. [15] Gigosos, M. A., Cardeñoso, V., 1996, New plasma diagnosis tables of hydrogen Stark broadening including ion dynamics, Journal of Physics B: Atomic, Molecular and Optical Physics, 29/20:4795. [16] Lin, R., Hoch, M., 1989, The solubility of hydrogen in molten aluminum alloys, Metallurgical and Materials Transactions A, 20/9:1785-91. [17] Kunze, H. J., 2009, Introduction to Plasma Spectroscopy: Springer Berlin Heidelberg. [18] Kharin, S., 1997, Mathematical model of the short arc phenomena at the initial stage, Electrical Contacts, 1997, Proceedings of the FortyThird IEEE Holm Conference on:289-305. [19] Boxman, R. L., Sanders, D. M., Martin, P. J., 1995, Handbook of Vacuum Arc Science and Technology: Fundamentals and Applications: Noyes Publications. [20] Tsai, Y.-Y., Masuzawa, T., 2004, An index to evaluate the wear resistance of the electrode in micro-EDM, Journal of Materials Processing Technology, 149/1:304-9.