This invention relates to a system and method of melting raw materials, such as reactive metals, e.g. titanium, zirconium, nickel, cobalt, and their alloys. The molten material can subsequently be used to form ingots or castings. The invention is presently considered especially useful for forming small cross-sectional ingots, and/or ingots or castings that will later be converted into powder, where homogeneity of each granule of powder is of particular concern.
Small cross-sectional bars and castings of these metals are used throughout the aerospace, automotive, energy, and medical industries. They can be machined or forged into any number of shapes. They may be used as the feedstock to be drawn into wire.
Such bars are typically made from larger ingots which are incrementally heated to high temperatures and then forged down into the desired size. The forging process can lead to considerable yield loss—a 60-70% yield of usable metal is typical. This is mainly due to deformation of the ends of the ingot after a number of forging steps. In addition, it can take months for an ingot to await its turn in the queue to be forged. Still further, due to the relatively small surface area to volume ratio of the large ingots and associated cooling rates, the grain size of the finished product may be larger than desired.
For all these reasons, it is desirable to cast the ingots nearer to their desired final cross-sectional size, a feat which has heretofore not been accomplished for small cross-sectional ingots.
It is also desirable to ensure that the ingots are as homogeneous as possible, for reasons that will be apparent to those of ordinary skill in the art.
Furthermore, parts made from powdered metals are increasingly common. The powder is usually formed by grinding, or by remelting and atomizing, an ingot or casting that has been cast from a molten material. The parts can then be produced by consolidating the powder either directly into a final shape, or into a preform that is then machined. In most uses, it is usually very important that each powder particle be of the same composition. This can only be achieved by ensuring that the metal ingot or casting from which the powder is formed is homogeneous, which can in turn only be achieved if the molten metal from which the ingot or casting is made is homogeneous.
The most common method of ensuring homogeneity in the molten metal is to stir the molten metal. Another method, which is mentioned in U.S. Pat. No. 6,006,821 to Haun et al., dated Dec. 28, 1999, and assigned to the Applicant herein, uses an induction coil. It should be noted that the induction coil disclosed therein is powered separately from the plasma arc torch using an additional power source. U.S. Pat. No. 6,006,821 to Haun et al. is hereby incorporated by reference.
Metals such as titanium, zirconium, nickel, cobalt, and their alloys can be contaminated by the oxide refractories used to make induction furnaces. Therefore, these metals are typically melted in segmented water-cooled copper vessels, with an associated induction coil and its separate power source. However, this melting technique is only about 25% efficient thermally.
Other methods of melting metals to thereby form ingots are known in the art.
Raw material is fed into an electrically conductive vessel. A plasma arc torch melts at least some of the raw material within the vessel to thereby create a molten material. An inductor, physically disposed adjacent the vessel, and electrically disposed in series with the vessel in operation, effects electromagnetic stirring of the molten material by interacting with the current of the plasma arc torch.
In some embodiments, the inductor is not connected to any additional power source.
A switch may further be provided, and be operable to switch the system between a first configuration in which the inductor is in series with the vessel, and a second configuration in which the inductor is electrically bypassed and is not in series with the vessel.
The power supply may be a direct current power supply, an alternating current power supply, or both. The plasma arc torch may use direct current, alternating current, or a simultaneous combination of both to melt the material.
An actuator may be provided to tilt the vessel between a first position, for receiving the raw material and having it melted therein, and a second position, for pouring at least some of the molten material out of the vessel. The molten material may be poured into a receptacle such as a mold. The mold may include a heated, upper portion, where the molten material is maintained in molten form, and a lower portion maintained at a temperature at which the molten material solidifies to thereby form an ingot.
The raw material may be fed into the vessel in batches, and poured out of the vessel in additional batches. The raw material may be fed into the vessel with a feeder, such as a bar feeder, a bulk feeder, a hopper, or a canister.
The raw material may be a reactive metal such as titanium, zirconium, nickel, cobalt, or combinations or alloys thereof, and may be in the form of compacted disks, cylinders, blocks, loose material wrapped in foil, unwrapped loose material, or scrap pieces of the raw material.
Exemplary embodiments will be described in more detail with reference to the accompanying drawings, in which:
Exemplary embodiments of the present invention provide a system and method for producing a homogeneous melt from raw material in solid form. The raw material is fed into a vessel. A plasma arc torch melts at least some of the raw material within the vessel to thereby create at least a portion which is molten. An induction coil, provided around or below the vessel, is in series with the plasma arc, thereby providing electromagnetic stirring of the molten metal without the need for a separate power source. This stirring leads to superior homogeneity over that of comparable known systems.
Exemplary embodiments of the present invention also provide a system and method for producing ingots or castings, such as small cross-sectional area ingots, from raw material in solid form. In one exemplary embodiment, this is accomplished first by melting the material as described above, then pouring the molten material into any desired receptacle, such as a mold. The pouring may take place in any desired manner.
In another exemplary embodiment, the raw material is fed into a tiltable vessel in a substantially upright position. A plasma arc torch melts at least some of the raw material within the vessel to thereby create a portion which is molten, while an induction coil, provided around of below the vessel, provides electromagnetic stirring of the metal without the need for a separate power source. The vessel is then tilted to pour some of the molten material into a receptacle such as a mold to thereby form a casting or an ingot.
The invention is especially considered particularly suitable for titanium, zirconium, nickel, cobalt, and combinations and alloys thereof.
Referring to
Once in the vessel 10, the raw material is melted by a stationary or movable plasma arc torch 12, shown schematically as creating a plasma arc 14, and powered by a power source 16.
It will be appreciated that metals such as titanium, zirconium, nickel, cobalt, and their alloys cannot be melted in ceramic lined vessels. The molten material would react with the ceramic and become contaminated to the point of being unusable. Therefore, in one exemplary embodiment, the vessel 10 is made of copper, which is considered more suitable as a melting receptacle for melting these metals. However, because of the relative melting point of copper compared to some of the metals which may be used with the invention, it may be advantageous to cool the copper vessel. Therefore, in one exemplary embodiment, the vessel 10 is a water-cooled (or other fluid-cooled) copper vessel. Typically, the bottom surface and the sides of the vessel are water-cooled. Thus, while the top portion of the material within the vessel is molten, some amount of the material may re-solidify (or, in some cases, not melt to begin with) to form a solid skull at the bottom of the vessel. The skull may be considered undesirable, but for large quantities of material, it constitutes a small fraction of the overall processed material. As reported by the inventors herein, when melting metals such as titanium, zirconium, nickel, cobalt, and their alloys, not all of the material can be maintained molten, and this can sometimes be advantageous despite the inherent efficiency losses. Any appropriately sized and shaped vessel may be used, depending on the constraints of the system 100.
In those instances in which an alloy ingot or other casting is desired, correct melting and mixing of the raw material is crucial. The volume of the vessel 10 should thus be large enough to hold the discrete pieces of raw material while melting, as well as to effectively pre-mix the alloy and even out any small compositional variations inherent to the raw material from one piece to the next. This may be further achieved by purposely emptying the vessel on a regular basis, leaving a minimal amount of skull to avoid the build-up of higher melting point elements, components, or alloys. In presently preferred embodiments, the vessel is not used to refine the alloy, so relatively long residence times are not required.
Also shown schematically in
While the term “inductor” is often used to refer to an inductor within an AC circuit, this term is not intended to be so limited. In a presently preferred embodiment, the power supply 16 is a DC power supply and the plasma arc torch 12 is configured to use direct current to melt the material. In this respect, the inductors 18 may be termed “DC coils” rather than inductors. These terms should be considered interchangeable. In other embodiments, the power supply 16 is an AC power supply or comprises both an AC and a DC power supply. A suitable plasma arc torch 12 may be selected based on the power supply 16 that is to be used, but the inventors have conceived that the inductors 18 may be used interchangeably in any AC, DC, or combination system.
A switch 20 may also be provided to turn the electromagnetic stirring on and off. In its simplest form, as shown, the switch 20 may be a single pole, single throw switch. When the switch is open, as illustrated, the induction coils 18 are effectively in series with the vessel 10 and plasma arc torch 12. When the switch 20 is closed, the vessel 10 is connected directly to ground, and the stirring is turned off.
Referring to
Turning now to
In those instances in which an alloy ingot is desired, correct melting and mixing of the raw material is crucial. The volume of the vessel should thus be large enough to hold the discrete pieces of raw material while melting, as well as to effectively pre-mix the alloy and even out any small compositional variations inherent to the raw material from one piece to the next. This may be further achieved by purposely emptying the vessel on a regular basis, leaving a minimal amount of skull to avoid the build-up of higher melting point elements, components, or alloys. The vessel is not used to refine the alloy, so relatively long residence times are not required. The tilt-pouring of the vessel itself enables the rapid turnover of raw material, thereby creating a nearly homogeneous liquid, which is then delivered to the mold.
Turning now to the mold, the mold may have many different possible shapes depending upon the articles desired. Any suitable closed- or open-bottom mold may be used.
The mold may be shaped to create a specific part or parts or any preformed shape which can be converted into a part or parts. In this case, the mold may have an open top and closed bottom. Alternatively, the mold may be shaped for semi-continuous ingot production. In this case, the mold may have an open top and bottom. Any number of molds may be moved into and out of the casting position in a semi-continuous fashion.
One exemplary open-bottom mold 26 will be described. As was described above, the molten material is fed into the mold in discrete amounts or batches. Referring to
The mold may have a segmented temperature control system, i.e. be cooled at the bottom and heated at the top, where the molten material is fed in. This maintains a certain depth of molten material above the portion of material that is in the process of solidifying at any given time. The pressure created by this molten head ensures the formation of an ingot which is free from porosity and other defects, such as solidification shrinkage voids. In addition, the constant mixing created by the heater ensures a chemically homogeneous molten pool, thereby ensuring chemical homogeneity throughout the length of the ingot. Some of the solidified material may also be re-melted by the molten head and mixed in with it, further adding to the homogeneity. The cooling within the mold may be, e.g, water cooling, and the heater may be, e.g, an induction heater. An exemplary material for the mold is copper.
The mold may be a small cross-sectional area mold. For example, for metals such as titanium, zirconium, nickel, cobalt, or combinations or alloys thereof, it has heretofore been very difficult to create ingots with cross-sectional areas of about 7.1 square inches or less (e.g. circular cross-sectional ingots with diameters of about 3.0 inches or less). For molds of this size, if a plasma arc torch were used to heat the material in the top portion of the mold, the diameter of the plasma arc would be large enough to destroy the mold itself. Therefore, an alternative heat source such as an induction coil may be used to maintain the top portion of the material within the mold in its molten state.
Additionally or alternatively, the term “small cross-sectional area” can refer to a mold of any appropriate size to accomplish any one or more of the following effects:
For example, the mold may have a cross-sectional area of about 7.1 square inches or less. An exemplary ingot size is 2⅛ inches diameter by 120 inches or more long. This may be very close to the desired final size, and require only a small amount of machining to remove undesirable as-cast features related to the way the ingot solidifies and cools. Furthermore, because of the higher surface area to volume ratio and associated cooling, and because of the temperature gradients established in the ingot by the segmented (heated/cooled) mold, a typical as-cast grain size for a titanium alloy ingot is 100 micrometers or less.
However, the mold is not limited to a circular cross-section, but may have a cross-section that is polygonal, polygonal with rounded corners, or any other desired shape. Still further, the mold is not limited to a constant cross-sectional size or shape. The mold may be tapered or have other non-constant cross-sectional shapes. In such embodiments, a “small cross-sectional area” mold may be considered a mold with a cross-sectional area of about 7.1 square inches or less across any cross-section, or alternatively, a mold with a cross-sectional area of about 7.1 square inches or less across some cross-sections, and larger cross-sectional areas across others.
The resulting ingot was approximately 50 cm long. Transverse cuts were made across the ingot near the top and bottom as well as the middle of the ingot. See
Representative photomicrographs were taken showing a fine grain microstructure. See
Casting of titanium-aluminide (TiAl) alloys is an emerging industry with the potential to be more cost effective than forging methods. Steel mold casting has been suggested as a possible technique to process TiAl alloys into an ingot that can be machined or forged as-cast, or can be ground or remelted and atomized into powder form. This process would allow for good surface finish, higher dimensional tolerances, and a reusable mold. Testing was conducted with a titanium aluminide alloy. High temperature gradients can cause the ingot or skull to shatter violently. This thermal sensitivity makes processing and extraction of the TiAl alloy more difficult. In addition, safety precautions must be taken to prevent damage and injury. Helium (He) gas was used for the plasma torch.
Experimental Procedure
Around 6.5 kg of a TiAl alloy was loaded into the tiltable hearth 10 of the plasma arc melting system 100 to be melted by the plasma torch 12 using helium gas. DC stirring coils 18 were used to mix the elements of the molten metal in the hearth 10 to ensure homogeneous composition (See
The stirring function is able to be turned on and off without halting the melting process using the switching cylinder 20, which creates a short that bypasses the stirring coils 18. After the melt was mixed, the hearth 10 was placed into tilting position and the torch 14 was pointed at the rear of the hearth 10 and at a far enough distance to avoid being hit by the tilting hearth 10. The hearth 10 was tilted until all liquid contents were poured out. A single pour filled the steel mold completely and then the process ended and the ingot was extracted after cooling.
Note that
The steel mold (not illustrated in
The extracted ingots were cut longitudinally using a powered saw. After sectioning, the surface finish of the ingot and the longitudinal cross-sections were inspected and photographed. The longitudinal sections of the ingots reveal shrinkage porosity along the center of the ingots (
As will be understood by those skilled in the art, the present invention may be embodied in other specific forms without departing from the essential characteristics thereof. Many other embodiments are possible without deviating from the spirit and scope of the invention. These other embodiments are intended to be included within the scope of the present invention, which is set forth in the following claims.
This application claims priority to, and the benefit of, U.S. Provisional application 61/702,726, filed Sep. 18, 2012, the disclosure of which is hereby incorporated by reference.
Number | Name | Date | Kind |
---|---|---|---|
3173981 | Myers | Mar 1965 | A |
3554522 | Herrmann | Jan 1971 | A |
3944412 | Liu | Mar 1976 | A |
4940486 | Sommerville | Jul 1990 | A |
6006821 | Haun et al. | Dec 1999 | A |
20030150374 | Sasatani | Aug 2003 | A1 |
20040056394 | Jackson | Mar 2004 | A1 |
20050175063 | Roberts et al. | Aug 2005 | A1 |
20080298425 | Jackson | Dec 2008 | A1 |
Number | Date | Country |
---|---|---|
0273975 | Jul 1988 | EP |
2001-019593 | Jan 2001 | JP |
2007-024396 | Feb 2007 | JP |
10-2012-0031421 | Apr 2012 | KR |
2346221 | Feb 2009 | RU |
Entry |
---|
Moskvichev et al. RU 2346221 C1 published Feb. 2009. Machine translation. |
SU 432320 A published Apr. 1975. Machine translation. |
International Search Report and Written Opinion mailed Dec. 10, 2013 in Int'l Patent Application No. PCT/US2013/060477, 13 pages. |
Extended EP Search Report, received by the European Patent Office in application No. 13838299.9; 8 pages. |
Number | Date | Country | |
---|---|---|---|
20140182416 A1 | Jul 2014 | US |
Number | Date | Country | |
---|---|---|---|
61702726 | Sep 2012 | US |