Alloy coated EDM wire

Abstract

An electrode wire for use in an electrical discharge machining apparatus includes a metallic core and a layer of gamma phase brass disposed over the metallic core. Particles of beta phase brass are interspersed within the gamma phase brass layer. An oxide layer including zinc is disposed over the gamma phase brass layer.

Claims

1. An electrode wire for use in an electrical discharge machining apparatus, comprising: a metallic core; a layer of gamma phase brass disposed over the metallic core; a layer of beta phase brass between the core and the gamma phase brass layer; particles of beta phase brass separate from the layer of beta phase brass and interspersed within the gamma phase brass layer; and an oxide layer including zinc disposed over the gamma phase brass layer.

2. The electrode wire of claim 1, wherein the beta phase brass layer is continuous.

3. The electrode wire of claim 1, wherein a combined thickness of the gamma and beta phase brass layers is about 12 to 15 μm.

4. The electrode wire of claim 1, wherein the gamma phase brass layer is discontinuous so as to expose the beta phase brass layer.

5. The electrode wire of claim 1, wherein the gamma phase brass layer is discontinuous.

6. The electrode wire of claim 5, wherein portions of the zinc oxide layer also extend into the discontinuities of the gamma phase layer.

7. The electrode wire of claim 1, wherein the zinc oxide layer has a thickness of about 84 nm.

8. The electrode wire of claim 1, wherein the core comprises at least one of copper, a copper zinc alloy, copper clad steel, aluminum clad steel, and a metal and a metal alloy.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1A-1E are schematic illustrations of various stages of forming an EDM wire in accordance with an embodiment of the present invention.

(2) FIG. 2 is a schematic illustration of an EDM wire according to the prior art.

(3) FIG. 3 is an end view of the prior art EDM wire of FIG. 2.

(4) FIG. 4 is a metallographic cross-section of the prior art EDM wire of FIG. 2.

(5) FIG. 5 is an optical photomicrograph of the surface of the prior art EDM wire of FIG. 2 at a diameter of 0.25 mm.

(6) FIG. 6 is a metallographic cross-section of an EDM wire according to the present invention at a diameter of 1.2 mm.

(7) FIG. 7 is a metallographic cross-section of an EDM wire according to the present invention at a diameter of 0.25 mm.

(8) FIG. 8 is an optical photomicrograph of the surface of the EDM wire of FIG. 6.

(9) FIG. 9 is a sketch of a test cut performed to compare the EDM wire of FIG. 1 with the prior art EDM wire of FIG. 2.

DETAILED DESCRIPTION

(10) The present invention relates to electrode wires used for fabricating metal or electrically conducting parts by EDM using an EDM machine tool, and specifically to a process for manufacturing high performance EDM electrode wire utilizing gamma phase brass coatings and an EDM wire produced from the process.

(11) FIGS. 1A-1E illustrate an example wire electrode or EDM wire 10 in accordance with an aspect of the invention. Referring to FIG. 1A, the EDM wire 10 includes a core 12 formed from a metal and/or metal alloy including, for example, copper, a copper zinc alloy, copper clad steel or aluminum clad steel. The core 12 can have a diameter of about 0.8 to 2.0 mm.

(12) A layer 13 of a second metal having a heat of vaporization less than 35 KJ/cm3, e.g., zinc, is coated over the core 12. It will be appreciated, however, that the layer 13 can include additional materials in addition to zinc, such as copper. The layer 14 can be coated on the core 12 in any known manner, such as by electroplating. The layer 13 has a thickness of about 10-12 μm and cooperates with the core 12 to form a composite wire 10.

(13) The composite wire 10 is heated by diffusion annealing, which causes a portion of the layer 13 to be transformed into a brass alloy, such as γ-phase brass, forming a coating layer 14 (see FIG. 1B). In one example, heat treating a core 12 including copper causes the copper to diffuse outwardly into the layer 13, thereby sequentially transforming zinc into a γ-phase brass alloy. The γ-phase brass layer 14 can have a substantially homogenous composition and thickness. In one example, the diffusion annealing can be performed at about 150° C.-160° C. for about 24 hours. The diffusion annealing is performed in an oxygen enriched environment, i.e., in an environment composed of greater than 22% oxygen, causing the outer portion of the layer 13 to oxidize into a thin, zinc oxide layer 16 defining the exterior of the coated wire 10. The zinc oxide layer 16 has a thickness of at least 1 μm. Depending on the extent and/or duration of the heat treatment, a portion of the γ-phase layer 14 adjacent the core 12 can receive additional copper from the core, which transforms the portion into a layer of β-phase brass 18 (see FIGS. 1C-1D).

(14) After the first round of heat treatment, the coated wire 10 can be diffusion annealed again, but this time at a greater temperature and shorter duration than the first round. In one example, the second round of diffusion annealing can be performed at about 275° C. for about 6 hours in an enriched oxygen environment. During the heat treatment, the zinc oxide layer 16 continues to preferentially consume zinc from the underlying γ-phase layer 14 due to reaction kinetics, thereby reducing the zinc content locally within the γ-phase layer at random locations until the local zinc concentration is below the limit for γ-phase existence at these locations. As a result, local precipitation of β-phase particles 20 occurs within the γ-phase layer 14. The β-phase particles 20 are therefore dispersed within the γ-phase layer 14 and can be completely surrounded or enclosed by the γ-phase layer.

(15) Next, the coated wire 10 undergoes a cold drawing process, which deforms the coated wire to attain a desired tensile strength and finish diameter. During the drawing step, the γ-phase layer 14 is redistributed over the circumference of the wire 10, as shown in FIG. 1E. The γ-phase layer 14 is completely brittle and therefore cracks when elongated by the drawing step. As a result, a series of discontinuities or gaps 24 are formed in the layer 14 during drawing. The gaps 24 extend radially inward such that portions of the core and/or β-phase layer 18 are exposed to ambient conditions through the γ-phase layer 14.

(16) At the same time, some of the brittle γ-phase brass particles forming the layer 14 become fractured and embed themselves in the surface of the underlying β-phase layer 18, thereby producing a convoluted topography along the γ-phase layer/β-phase layer interface. Such a configuration can create hydraulic turbulence at the wire 10 surface, thereby enhancing the flushing action of the dielectric.

(17) It is clear from the foregoing that several physical changes occur when the wire 10 is heat treated in an enriched oxygen environment. First, the zinc layer 13 is progressively transformed into a zinc oxide layer 16 and a γ-phase 14 layer and, if desired, an additional β-phase layer 18 radially between the core 12 and the γ-phase layer. Second, the γ-phase 14 layer becomes metallurgically bonded to the layer beneath it, i.e., either the core 12 or the β-phase layer 18, thereby improving adherence between the γ-phase layer and the layer beneath it. Third, the zinc layer 13 continues to form the zinc oxide layer 16 and cannibalizes zinc from the γ-phase layer 14 until β-phase brass particles 20 precipitate out of the γ-phase layer at random locations.

(18) It is also clear from the above that several physical changes occur in the wire 10 during drawing. First, the brittle γ-phase layer 14 fractures and becomes redistributed over the circumference of the wire 10, forming discontinuities or cracks 24 therein. These discontinuities 24 are at least partially filled with portions of the β-phase layer 18, which is extruded outward in the drawing step due to its high ductility. As a result, the β-phase layer 18 is extruded radially outward into the discontinuities 24 to thereby offer a more efficient flushing surface than the core wire 12, which would occupy that space but for the presence of the β-phase layer. Portions of the zinc oxide layer 16 can also extend into the discontinuities 24 following cold drawing. Moreover, the β-phase brass particles 20 can advantageously affect the fracture mechanics of the wire 10 when they are deformed by wire drawing.

Example

(19) A sample (HTCLN) of the EDM wire of the present invention was compared to a sample (SD2) reproduced from the description in Blanc et al. (U.S. Pat. No. 8,378,247). Referring to FIGS. 2-3, the SD2 sample wire 31 included a γ/β double layer construction containing a core 32 made of 63Cu/37Zn overlaid with a continuous β-brass sublayer 33 coating having a thickness E.sub.3. A surface layer 34 overlays the sublayer 33 and has a thickness E.sub.4. The surface layer 34 includes a fractured γ-brass structure 35a revealing β-brass in the fractures. A γ-phase region 35 of the surface layer 35 is bordered by the fractures 35a in the surface layer. β-brass may at least partially fill the fractures 35a in the γ-brass surface layer 34, giving the surface of the wire 31 a certain degree of continuity. An oxide layer 36 overlies the surface layer 34 and has a calculated thickness E.sub.0. The SD2 wire sample 31 has a diameter D.sub.1 of 0.25 mm.

(20) In order to compare the present invention to the current state of the art of γ-brass coated wire electrode technology, it was appropriate to establish a characterization of the metallurgical structure and performance of current γ-brass coated constructions, e.g., the SD2 sample.

(21) That said, FIG. 4 is a lightly etched metallurgical optical cross-section at high magnification of the SD2 sample. FIG. 5 is an optical photomicrograph at high magnification of the surface topography of the SD2 sample. A Selective Dissolution Test as prescribed by Blanc et al. was performed on the sample SD2, with the result being that the oxide layer 36 had a calculated thickness E.sub.o=191 nm (see FIG. 3). This value was consistent with the Blanc et al. preferred result of 100 nm-250 nm.

(22) The sample HTCLN was prepared according to the present invention using the process schedule detailed as follows:

(23) Stage 1. Electroplate 10-12 μm zinc on 1.2 mm diameter 60Cu/40Zn core wire

(24) Stage 2. Heat Treat at 155° C. for 24 hrs in an oxygen atmosphere

(25) Stage 3. Raise Heat Treat temperature to 275° C. and continue for additional 6 hrs

(26) Stage 4. Pickle in concentrated H.sub.2SO.sub.4 solution (10%-15% H.sub.2SO.sub.4/pH=1-2)

(27) Stage 5. Draw to finish diameter of 0.25 mm

(28) The strategy employed in processing the sample HTCLN was to create a sample with a heat treatment known to produce a microstructure similar in elements and dimensions to the sample SD2 while removing any potential excess oxides introduced by the heat treatment responsible for the unique microstructure being evaluated. To this end, pickling the coated wire in Stage 4 removed excessive oxide from the HTCLN sample for purposes for testing the HTCLN sample against the SD2 sample. In use, however, the oxides would not be removed from the coated wire.

(29) As a result, both the SD2 and HTCLN samples were intended to have comparable microstructures except for their γ-phase layer structures, i.e., the presence of β-phase particle precipitates within the γ-phase layer in the sample HTCLN versus the absence of such particles within the γ-phase layer in the sample SD2. This was done for the purpose of establishing that the unique microstructure of the present invention is responsible for the improved performance over the prior art.

(30) FIG. 6 is a photograph of an as polished metallurgical optical cross-section at high magnification of the sample HTCLN at the conclusion of Stage 3. The photograph clearly indicated the presence of β-phase particle 20 precipitates within the γ-phase layer 14 at the intermediate diameter of 1.2 mm prior to wire drawing. This and similar cross-sections of the HTCLN sample following Stage 3 were analyzed with Paxit™ Image Analysis Software and found to average 6.4% aerial content of β-phase particle 20 precipitates, which were previously identified as β-phase particles by EDS analysis on a scanning electron microscope (SEM).

(31) A modified Selective Dissolution Test was performed on the HTCLN sample at the conclusion of Stage 3 where the wire diameter was 1.2 mm. A 120 minute dissolution time was used when performing the tests on the HTCLN sample due to the larger diameter of the HTCLN sample compared to the 0.25 mm diameter SD2 sample. Using this modified test, at the conclusion of Stage 3, the E.sub.o for the HTCLN sample was calculated to be 227 nm. At the conclusion of Stage 4, the E.sub.o was calculated to be 95 nm, which is a significant drop from the as heat treated value.

(32) FIG. 7 is a lightly etched metallurgical optical cross-section at high magnification of the HTCLN sample at the conclusion of Stage 5. The β-phase particle 20 precipitates are clearly shown dispersed within the γ-phase layer 14. It is also clear that the continuous γ-phase layer 14 synthesized during the heat treatment fractured into a discrete but discontinuous layer having a series of cracks or discontinuities 24 extending radially inward towards the core 12. At the same time, the β-phase layer 18 also redistributed around the circumference of the wire 10 circumference but remained a continuous layer due to its greater ductility. To this end, the β-phase layer 18 was extruded radially outward into the discontinuities 24 in and between portions of the γ-phase layer 14. Portions of the zinc oxide layer 16 also extended into the discontinuities 24.

(33) FIG. 8 is an optical photomicrograph at high magnification of the surface topography of the HTCLN sample, which is similar in topography to the SD2 sample. The photomicrograph also illustrates the same surface continuity evidenced in the SD2 sample. A Selective Dissolution Test as prescribed by Blanc et al. was performed on a sample of HTCLN at the finish diameter of 0.25 mm with the result that E.sub.o=84 nm.

(34) Analysis

(35) Considering the above characterizations of the SD2 and HTCLN samples, a comparison of the two constructions is summarized in the table below:

(36) TABLE-US-00001 Wire Construction SD2 Wire Construction HTCLN Nominal γ-layer E.sub.4 Thickness 3-7 μm 5-10 μm Structure of γ-layer Single Phase Field Two Phase Field of γ-brass + of γ-brass β-brass Precipitates Nominal β-layer E.sub.3 Thickness 5-12 μm 4-10 μm Location of β-phase Beneath γ-particles + Same as SD2 + Precipitated Filling Fractures of within γ-particles γ-particles Double Layer Thickness 10-15 μm 12-15 μm Calculated Value of E.sub.0 191 nm 84 nm

(37) Estimating the exact thickness values for E.sub.3 and E.sub.4 was difficult due to 1) fracturing of the γ-phase layer into irregularly shaped particles and groups thereof during drawing, and 2) because the β-phase layer also redistributes itself during drawing. However, estimating the double layer thickness (E.sub.3+E.sub.4) can be more readily and accurately achieved since it can be defined by the outside diameter of the wire and the inner diameter of the β-phase layer.

(38) With these limitations and the structural similarities between the samples SD2 and HTCLN in mind, it is reasonable to conclude that the HTCLN sample has metallurgically significant differences in microstructures from the SD2 sample due to the presence of the β-phase particles precipitated out from the γ-phase layer in the HTCLN sample.

(39) In order to quantify the performance of the SD2 and HTCLN samples, test cuts of a simulated punch were performed on a Model 650 G plus Excetek EDM wire machine tool. The work piece consisted of a 2.0 inch thick plate of hardened (R.sub.c of between 52-56) D2 tool steel surface ground on the top and bottom to create sealed flushing conditions. The geometry of the test cut is illustrated in FIG. 9. The lengths of segments are:

(40) A.sub.0=0.025 inch

(41) A=0.200 inch

(42) B=0.200 inch

(43) C=0.400 inch

(44) D=0.400 inch

(45) E=0.400 inch

(46) F=0.100 inch

(47) G=0.025 inch

(48) The test cut included a timed roughing pass followed by two timed skim cuts performed in sequence. Each pass was initiated at an edge of the plate. The test cuts were spaced out on the plate so that at no time was an edge or cutting path within 0.200 inches of a previous kerf to guarantee the integrity of flushing conditions. Initially multiple cycles of the roughing pass were conducted to establish the most aggressive machine technology that each of the wire constructions could sustain through the complete cycle from A to G without any wire breaks. The Brass Machine Technology provided by the manufacturer was used as a starting point and adjustments made to it until wire breaks occurred. The Excetek machine technology parameters available to the operator are listed below with a brief explanation of their function where appropriate:

(49) TABLE-US-00002 Parameter Range Comment PM (Power) 1-10 OV (Open Voltage) 1-20 ON 0-24 OFF 4-50 AN (arc on) 1-16 AFF (arc off) 4-50 SV (Servo Voltage) 16 V-90 V  WT (Wire Tension) 1-20 10 = 1,200 gms WF (Wire Feed) 1-20 2-21 m/min WA (H.sub.2O Pressure) 1-8  8 = 250 psi FR %  1-500 F (Feed Rate) 0-4  in/min FT (Servo Mode) G95 = servo mode G94 = manual mo SC (Servo Control) 1-99

(50) The machine technologies available and those employed for test cuts of the two SD2 samples and one HTCLN sample are listed in the table below:

(51) TABLE-US-00003 Rough Cuts Brass SD2* SD2 HTCLN Skim 1 Skim2 PM 10 10  10  10 10 6 OV 8 9 8 10 14 12 ON 15 15  15  16 3 2 OFF 8 8 8  8 11 10 AN 8 8 8  8 2 2 AFF 8 8 8  8 11 10 SV 38 43  38  38 38 45 WT 10 10  10  10 13 15 WF 7 7 7  7 7 7 WA 8 8 8  8 1 1 FR % 100 100  100  100  100 100 F 0.150    0.150    0.150    0.150 0.236 0.394 FT G95 G95 G95 G95 G95 G95 SC 10 12  12  14 18 20 Offset 0.008    0.008    0.008    0.008 0.0056 0.0052

(52) In summary the HTCLN sample demonstrated a toughness that allowed it to sustain a more aggressive machine tool technology than the SD2 sample. The parameters most effective at influencing wire performance are identified by underlining. The technology SD2* came the closest of the SD technologies to the HTCLN technology but the test cut with the SD2* technology resulted in five wire breaks in traveling from segments A to G. The same skim technologies 1 and 2 and offsets for all rough cuts and skim passes were used on both Sample SD2 and Sample HTCLN.

(53) The results of the test cuts are summarized in the table below:

(54) TABLE-US-00004 SD2 HTCLN Rough Cut Time 0:17:49 0:15:29 (hrs:mins:sec) Skim 1 Time 0:7:34 0:7:34 Skim 2 Time 0:4:14 0:4:14 Calculated Rough Cut 0.1025 0.1179 Feed Rate (in/min) Surface Finish (Ra μm) 0.795 0.825

(55) The machine tool timed the rough cut starting at the beginning of segment A.sub.0 through the conclusion of segment G where segment A.sub.0 includes a transition where ideal flushing conditions and the servo equilibrium are being established, but the short span of these non-equilibrium conditions has minimal effects on any conclusions drawn from the timing data.

(56) The completed test punches from the test were checked for dimensional stability and both punches were accurate to within a tenth of one mil. The surface finishes of both samples also were acceptably close as measured on a Mitutoyo SJ-410 surface roughness tester. The calculated feed rate (cut length divided by cycle time) of the HTCLN sample was determined to be about 15% faster than that of the SD2 sample. This is clear evidence that the unique microstructure of the HTCLN sample, namely, the presence of β-phase particle precipitates within the γ-phase layer, allowed the HTCLN sample to exhibit improved performance over state of the art EDM wires.

(57) What have been described above are examples of the present invention. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the present invention, but one of ordinary skill in the art will recognize that many further combinations and permutations of the present invention are possible. Accordingly, the present invention is intended to embrace all such alterations, modifications and variations that fall within the spirit and scope of the appended claims.

(58) For example, it will be appreciated that the EDM wire formed according to the present invention can include only the γ-phase layer or both the γ-phase layer and β-phase layer—both constructions including precipitated β-phase particles dispersed/isolated within the γ-phase layer. In the former construction, the discontinuities can extend to the core to expose the core. In the latter construction, the discontinuities expose the β-phase layer.