System and Method for Depositing a Metal to Form a Three-Dimensional Part

Abstract

A system and method depositing metal to form a three-dimensional (3D) part on a substrate. A wire is moved relative to a location on the substrate while a laser heats a proximal end of the wire at the location using a laser beam. The laser causes the wire and substrate to reach a melting point of the wire to fuse the wire at the location on the substrate. The wire can be preheated by passing a current through the wire.

Claims

1. A system for depositing a metal to form a three-dimensional (3D) part, comprising: an electrically conductive substrate; a wire made of the metal; means for moving the wire relative to a location on the substrate; means for providing an electrical current through the wire into the substrate; and a laser beam, wherein the wire and substrate are heated to reach a melting point of the wire to fuse the wire at the location on the substrate.

2. The system of claim 1, wherein the means for moving the wire and the means for providing the electrical current to the wire are combined into an electrically conductive tube.

3. The system of claim 2, further comprising: means for heating the wire using an electrical current applied to the means for moving the wire.

4. The system of claim 3, wherein the means for moving, the laser beam, and the electric current are controlled independently.

5. The system of claim 4, wherein the means for moving, the laser beam, and the electric current are in coordination to cause the wire and the substrate to reach a melting point of the metal to fuse the wire at the location of the substrate.

6. The system of claim 5, wherein the coordination is controlled to optimize the fusion of the wire to the base substrate and prior deposited metal while minimizing vaporization and spattering of melted metal.

7. The system of claim 1, wherein the electrical current open circuit voltage is below the arc initiation and arc sustain voltages of the wire to substrate junction.

8. The system of claim 1, wherein a diameter of the wire is in a range of 0.003 to 0.010 inch.

9. The system of claim 1, wherein the electric current is less than 10 amperes.

10. The system of claim 1, wherein a rate of the depositing of the metal is a nominal 1 Kg/hour.

11. The system of claim 1, wherein the means for moving includes means for moving the substrate.

12. The system of claim 1, wherein the means for moving includes means for moving the wire.

13. The system of claim 12, wherein the means for moving the wire includes motor-driven rollers.

14. The system of claim 11, wherein the moving of the substrate includes means for translating and means for rotating the substrate.

15. The system of claim 12, wherein the moving of the wire includes means for extending and retracting the wire.

16. The system of claim 1, wherein the means for moving includes means for moving the wire, the substrate and the laser beam.

17. The system of claim 16, wherein relative motions of the wire, the laser beam, and laser beam power are coordinated to optimize fusion of the wire to the substrate, and to minimize vaporization of the wire and substrate, and spatter.

18. A method for depositing a metal to form a three-dimensional (3D) part, comprising steps: moving a wire relative to a substrate; and heating a proximal end of the wire at a location on the substrate using a laser beam and an electrical current to cause the wire and substrate to reach a melting point of the wire to fuse the wire at the location on the substrate.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0023] FIG. 1 is elevated side view of a system for printing a three-dimensional metallic part according to embodiments of the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] FIG. 1 is an elevated side view of the system according to embodiments of the invention. Feedstock wire 100 is moved by drive rollers 105a and 105b. The feedstock wire 100 passes through the wire guide and electrical contact tube 110. A power supply 115 supplies a controlled amount of current through the contact tube 110, the feedstock wire 100, through a workpiece 140, with the current returning to the power supply 115.

[0025] The required amount of current for a wire 0.006 inches (about 150 microns) is very moderate—only about six to nine amperes, at a nominal 1 Kg/hour deposition rate for steel, and far, far below the tens to hundreds of amperes that would be used for processes such as MIG welding or MIG cladding.

[0026] The ohmic heating supplied in section 120 of the feedwire 100 is intentionally insufficient to melt the wire. Instead, the wire 100 continues to a laser beam supplied by a laser 130, focused at the intersection of the hot wire 100 and the base substrate (prior layer) 140, at focus point 135. That is, only the proximal end of the wire in contact with the substrate is heated to a melting point by the laser beam. Note that the means for moving the wire, the substrate, the laser beam, controlling the electric current and controlling the laser power can be controlled independently, as well as controlled in a coordinated manner.

[0027] Both the current from power supply 115 and the energy of laser 130 raise the bulk temperature of the wire and the surface temperature of the substrate at focus point 135 to well above the melting point of the feedstock wire 100, and the feedstock wire 100 welds to the surface of the substrate 140. The form of the weld surface is determined by the surface tension of the melted metal, which then solidifies into cooling deposit 145 very rapidly. Repeated passes of the process produce a built up area of solid metal 150 to construct a final desired 3D metal part.

[0028] If feedstock wire 100 is an inert metal (such as gold, platinum, tantalum, etc.), then the process can be carried out in open air. However, if the feedstock wire 100 is a more common alloy, such as carbon or alloy steel, aluminum, titanium, etc., then the heated region 120, molten region 135, and cooling deposit 145 can kept under an inert atmosphere, or alternatively in a vacuum chamber. Since the temperatures are relatively low compared to an arc process, the inert atmosphere can be argon, carbon dioxide, or even very inexpensive nitrogen.

[0029] Note that the proximal end of the wire in focus region 135 of the laser beam is completely molten and subject only to gravity and surface tension to determine the physical shape; it retains no solid structure. Thus, feed rollers 105a and 105b can be used to stop the process (by retraction) and initiate the process (by extrusion) with high positional accuracy.

[0030] Although the metal in region 135 is molten, it is still a continuous electrical and uninterrupted electrical path, and no arc ever forms. This is an important contrast versus the MIG welding processes. Typical MIG welders operate at 100 or more amperes of current. The system according to the embodiments uses current levels 1/10 that. Because no arc is formed, and oxygen is excluded by either vacuum or inert gas shielding, there is no spatter or slag formation.

[0031] To contrast further with MIG-based 3D printing, in a MIG welding or cladding process, a constant-voltage (and thus dynamically-variable current) power supply supplies a wire electrode fed by a motor. In the MIG processes, the metal is melted by an arc (an electrically induced plasma); the process is dynamically stable as shortening the arc results in a great increase in current because of the constant-voltage characteristic of a MIG welding system, this causes a higher heat input into the metal and a more rapid melt-back. On the occasion that the MIG electrode wire shorts the arc, the output current increases rapidly and ohmic resistance causes the MIG electrode wire to melt rapidly.

[0032] The MIG process is not absolutely stable however because the melting process produces droplets of molten metal on the tip of the wire electrode, which alternately form and drip into the weld pool on the workpiece, so the current varies tens of times per second in a sawtooth as droplets form (shortening the arc and increasing the current) and detach (lengthening the arc, and decreasing the current). The deposition of these individual droplets and the dynamically changing heat contribute to the surface roughness of MIG-deposited 3D metal printing. In contrast, the invention uses no arc, the current is constant, and therefore the ohmic heating constant for any particular set of production parameters.

[0033] In particular, avoidance of arcing is part of a preferred embodiment of the invention to avoid arcing, vaporization, and spatter. Preferably, the power supply 115 is turned off during initial contact of wire 100 with the substrate 140 at the start of a section of deposit 145, and the current turned on after contact is first made by extending the wire 100 with feed rollers 105a and 105b. Similarly, it is preferred to decrease the current to zero at the end of a section of deposit 145 before wire retraction, again to avoid arcing, vaporization, and spattering. For this reason, in a preferred embodiment of the invention, the open-circuit voltage of the power supply should be insufficient to strike or maintain an arc; for air, argon, nitrogen, or carbon dioxide at normal atmospheric pressure, the voltage required to create or maintain an arc is typically 18 volts to 24 volts. For this reason, a power supply with an open-circuit voltage below 18 volts, such as the range from three to six volts is preferred. Additionally, as minimizing sparking also minimizes spatter, a power supply with minimal to no output capacitance is preferred.

[0034] One may question why the ohmic heating supplied by power supply 115 is useful or necessary. The reason is the thermal gradient imposed by the laser 130. If the laser 130 is the sole heat source for the molten region 135, then two phenomena prevent economical part production: reflection, and vaporization.

[0035] Reflection is a lesser problem. For a CO.sub.2 laser operating at 10 micron nominal wavelength, most steel alloys (including matte-finish stainless steel “as-drawn” wire) have a surface absorption of about 0.16 (that is, 84% of the laser power is reflected away and only 16% goes into actual heating). This makes the laser energy “expensive” in terms of both the physical laser which must be ˜6 times larger than absolute heating requires, and of the electrical power input to that laser.

[0036] Vaporization is a bigger problem. If laser 130 is to supply sufficient heat to fully melt feedstock wire 100 from top to bottom (or side to side, as may be applicable according to the angle of laser 130), then one finds that for the useful range in diameter of feedstock wire 100 (diameters from roughly 0.003 to 0.010 inch, 75 to 250 micron), and a target feed rate of 1 Kg/hour of metal deposited, that with a laser power of roughly 100 watts (necessary to raise 1 Kg/hour of steel to about 2700 deg C.) the face of the feedstock wire 100 closest to the laser approaches the boiling point of steel (approximately 2900 deg C.) while the back of the feedstock wire has not yet achieved the melting point of steel (approximately 1500 deg C.). Thus, the face of the wire starts boiling before the back side of the wire fully melts.

[0037] Worse, the surface of substrate 140 stays well below 1000 deg C. In this situation, a good fusion weld between the feedstock wire 100 and the substrate 140 does not occur and the deposit produced 150 has flaws and fusion defects, if indeed that any bond occurs at all.

[0038] In this case, increasing the laser 130 power level does not help; the face of feedstock wire 100 simply ablates and carries away the increased energy as vaporized metal. Almost none of the additional energy couples into the substrate 140.

[0039] Applying the laser 130 at a more inclined angle or targeted more toward substrate 140 rather than feedstock wire 100 does heat the substrate 140 more adequately, but the substrate 140 can be presumed to be of the same material (and thus same reflectivity) as feedstock wire 100. In that case, as described above, 84% of the energy of the laser beam reflects from the substrate surface; reasonable geometric optics predict that because of the great disparity in heat-sinking capacity between the substrate 140 and the feedstock wire 100 that the feedstock wire 100 begins to boil away before the substrate 140 reaches the melting point.

[0040] One solution to this problem is to increase the diameter of the feedstock wire and increase the focused spot diameter of the laser so that the dwell time from when the wire enters the laser spot to the time the wire exits the laser spot becomes long in comparison to the time it takes for the heat of the laser to conduct from the laser-facing side of the wire to the shadowed side of the wire. In this case of a large wire and large laser spot, the overall deposition rate in kilograms of metal per hour is maintained or even increases. On the downside, the relatively large wire diameter and large laser spot size cause the minimum feature size to be quite large, each layer of deposition to be quite thick, and so the final surface finish to be rather rough.

[0041] Another solution to this problem is to deposit much more slowly; decreasing the production target from 1 Kg of useful part per hour to 1/10th that (100 g/hour) indeed allows sufficient time for laser 130 to heat the feedstock wire 100 and substrate 140 in melt region 135 to adequately bond the deposit 145 to substrate 140.

[0042] However, this method is undesirable in a production environment because it drastically slows the rate of production. A cost analysis indicates that if one ignores the cost of the feedstock, then parts that are produced at 1/10 the speed are usually parts that are 10 times more expensive. Of course, if circumstances are such that speed of production is not important, then one preferred embodiment of the invention is a system that simply omits ohmic heating supply 115. Another preferred embodiment of the invention uses ohmic heating plus laser heating during longer sections of depositing wire 100 onto substrate 140, and uses laser 130 alone to establish and terminate the fusion at the start and end of each section of wire deposit.

[0043] This speed of production issue is the reason for including the ohmic heating power supply 115. First, the ohmic heating supply preheats the feedstock wire 100 in bulk and with 100% uniformity; there is no “wire facing the laser” vs. “wire facing away from the laser” differential, nor any reflection of energy away, unlike the 10 micron infrared laser reflection. Second, the same ohmic heating effect that heats feedstock wire 100 also heats the substrate 140 in the melt region 135, again with 100% efficiency.

[0044] A second-order effect improves the heating ratio between feedstock wire 100 and substrate 140: the effect of convection versus conduction. The feedstock wire 100 is exposed to natural (and if necessary, forced) convection cooling into the atmosphere in area 120, while substrate 140 in the melt region 135 is not cooled convectively.

[0045] Although intuitively it seems that convective losses in the wire heating region 120 are wasted energy and slows the process, the actual result is that ohmic heating power supply 115 heating in substrate 140 (which is only conductively cooled) actually couples energy preferentially into the substrate 140 rather than into the feedstock wire 100 because although the wire 100 has a smaller cross-section, the wire 100 can dissipate the heat much more readily by convection, versus the substrate 140. As a rule of thumb, in metals, free convection is 10 times more efficient at heat transfer than pure conduction, and forced convection/ventilation is 100 to 1000 times more effective at cooling than simple conduction.

[0046] Application to 3D Printing

[0047] The invention as described above and shown in FIG. 1 is amenable for direct use in a 3D printer to print metal. The invention can be mounted in a conventional XYZ gantry if it is acceptable to have all of the laydown deposits 150 parallel in all layers. In the situation where surface finish specifications do not allow for the initiation and termination of laydown deposits 150 at the surface, then a simple turntable supporting substrate 140 with a slip-ring to pass the current from power supply 115 will permit the laydown of deposit 150 in any relative direction in any layer as desired.

[0048] Although the invention has been described by way of examples of preferred embodiments, it is to be understood that various other adaptations and modifications can be made within the spirit and scope of the invention. Therefore, it is the object of the appended claims to cover all such variations and modifications as come within the true spirit and scope of the invention.