US3713479A

US3713479A - Direct chill casting of ingots

Info

Publication number: US3713479A
Application number: US00110190A
Authority: US
Inventors: N Bryson
Original assignee: Alcan Research and Development Ltd
Current assignee: Alcan Research and Development Ltd
Priority date: 1971-01-27
Filing date: 1971-01-27
Publication date: 1973-01-30
Anticipated expiration: 1990-01-30
Also published as: AT330387B; SE384639B; IT946979B; AU470900B2; DE2203256A1; GB1381166A; BE778509A; CH556704A; CA966974A; ES399244A1; DK136886C; NO140132C; FR2126189A1; NL149398B; JPS548611B1; FR2126189B1; CS209837B2; DD99522A5; ATA57872A; AU3831872A

Abstract

In a direct chill casting of metal (e.g. aluminum) ingots, wherein an externally solidified ingot having an initially molten core is progressively withdrawn from a shallow, cooled, openended mold to which molten metal is progressively supplied, the ingot emerging from the mold passes successively through a first cooling zone extending from the mold for a predetermined distance along the path of ingot advance, and a second cooling zone located at that predetermined distance from the mold. Separate supplies of coolant fluid are respectively directed onto the ingot surface in the two zones, in such manner that the coefficient of heat transfer from the ingot to the coolant is substantially greater in the second zone than in the first. Specifically, the restricted intensity of cooling provided in the first zone is selected to maintain the outer portion of the ingot in solid state but preferably without completely solidifying the ingot core as the ingot traverses the first zone, while the greater intensity of cooling in the second zone effects complete solidification of the ingot core and simultaneously provides a high rate of cooling of the ingot periphery.

Description

United States Patent [1 1 Bryson 1 1 3,713,479 1 Jan. 30, 1973 1541 DIRECT CHILL CASTING OF INGOTS [75] Inventor: Neil Burton Bryson,

Ontario, Canada Kingston,

[73] Assignee: Alcan Research and Developmen t Ilimited, Montreal, Quebec, Canada [22] Filed: Jan. 27, I971 211 Appl. No.: 110,190

FOREIGN PATENTS OR APPLICATIONS 877,185 5/1953 Germany ..164/283 OTHER PUBLICATIONS Metal Industry, 10 October 1963. T8200. M586. Page 527.

Primary Examiner-R. Spencer Annear Attorney- Christopher C. Dunham, P. E. Henninger, Lester W. Clark, Robert S. Dunham, Gerald W. Griffin, Howard J. Churchill, R. Bradlee Boal, Robert Scobey and Henry T. Burke [57] ABSTRACT In a direct chill casting of metal (e.g. aluminum) ingots, wherein an externally solidified ingot having an initially molten core is progressively withdrawn from a shallow, cooled, open-ended mold to which molten metal is progressively supplied, the ingot emerging from the mold passes successively through a first cooling zone extending from the mold for a predetermined distance along the path of ingot advance, and a second cooling zone located at that predetermined distance from the mold. Separate supplies of coolant fluid are respectively directed onto the ingot surface in the two zones, in'such manner that the coefficient of heat transfer from the ingot to the coolant is substantially greater in the second zone than in the first. Specifically, the restricted intensity of cooling provided in the first zone is selected to maintain the outer portion of the ingot in solid state but preferably without completely solidifying the ingot core as the ingot traverses the first zone, while the greater intensity of cooling in the second zone effects complete solidification of the ingot core and simultaneously provides a high rate of cooling of the ingot periphery.

19 Claims, 5 Drawing Figures MoLTfN' Al PATENTED JAN 30 I975 3.713 479 sum 2 [IF 4 PATENTEU JAN 30 I973 SHEET 3 OF 4 will com

com

omw

PATENTEU JAN 30 I975 SHEET 0F 4 MIC.

com

omw

h ov wimp.

DIRECT CHILL CASTING OF INGOTS BACKGROUND OF THE INVENTION This invention relates to direct chill casting procedures for continuous or semicontinuous production of metal ingots. In an important specific aspect, the invention is directed to procedures for direct chill casting of aluminum metal and alloys thereof, herein generically termed aluminum.

Although the procedures of the invention are broadly applicable to the casting of any metal that can be satisfactorily cast in continuous or semicontinuous manner by direct chill techniques, the invention will be described herein for purposes of illustration with specific reference to the continuous casting of aluminum ingots.

In a typical example of present-day practice, the continuous direct chill casting of an aluminum ingot is effected in a shallow, open-ended, axially vertical mold which is initially closed at its lower end by a downwardly movable platform or stool. The mold is surrounded by a cooling jacket through which a coolant fluid such as water is continuously circulated to provide external chilling of the mold wall. Molten aluminum is introduced to the upper end of the chilled mold, and as this molten metal solidifies in a region adjacent the periphery of the mold, the platform is moved downwardly. With effectively continuous downward movement of the platform and correspondingly continuous supply of molten aluminum to the mold, there is produced an ingot of desired length.

The ingot emerging from the lower end of the mold is externally solid but is still molten in its central portion; in other words, the pool of molten aluminum within the mold extends downwardly into the central portion of the downwardly moving ingot for some distance below the mold, as a sump of molten metal having a progressively decreasing cross section as the ingot solidifies inwardly until its core portion becomes completely solid.

As an important feature of the direct chill casting process, continuously supplied coolant fluid such as water is brought into direct contact with the outer surface of the advancing ingot below the mold. This direct chilling of the ingot surface serves both to maintain the peripheral portion of the ingot in solid state and to promote internal solidification of the ingot.

Conventional direct chill casting procedure involves provision of a single direct chill cooling zone below the mold. Typically, the cooling action in this zone is effected by directing a substantial continuous flow of water (discharged, for example, from the lower end of the mold cooling jacket, and distributed substantially uniformly around the periphery of the ingot) onto the ingot surface immediately below the mold, just as the ingot emerges from the mold, in such manner that the water impinges with considerable force on the ingot surface at a substantial angle thereto, and flows downwardly over the ingot surface with continuing but diminishing cooling effect. Thus the greatest intensity of cooling is provided immediately below the mold outlet end, a locality which is ordinarily some distance above the level in the path of ingot advance at which the ingot core solidifies completely.

At the described locality of greatest cooling intensity, the coefficient of heat transfer from the ingot to the cooling fluid is typically about 0.5 cals./cm/sec./C, far

larger than the average value of the coefficient of heat transfer from the ingot to the mold (which is commonly about 0.05 cals./cm/sec./C), and also higher than the value of the ingot-to-coolant heat transfer coefficient at any lower level in the path of ingot advance. Typically, also, the thickness of the solidified ingot shell at this level of maximum cooling immediately below the mold is less than one fourth the maximum horizontal dimension of the ingot.

A problem encountered in conventional direct chill casting, especially (but not exclusively) in casting cylindrical ingots, is the tendency of the ingots to develop serious structural defects produced by socalled hot cracking, i.e., longitudinal center cracks formed incident to solidification and cooling of the ingot. These defects render the ingots unacceptable for many purposes. In the conventional practice described above, avoidance of hot cracking requires that the depth (vertical extent) of the molten metal sump below the lower edge of the mold be maintained at a value not greater than the minimum transverse dimension of the ingot, and indeed very comm only at a value no greater than two thirds of the minimum transverse dimension of the ingot.

For given cooling conditions, ingot dimensions and alloy composition, the sump depth is determined by the casting speed, i.e., the rate of downward advance of the ingot from the mold. Since conventional direct chill casting systems are not designed to permit appreciable control of cooling intensity, regulation of sump depth must be achieved by appropriate restriction of the casting speed. Specifically, the constraint imposed on sump depth by the necessity of avoiding hot cracking has generally limited casting speeds to values between about 1 and 7 inches per minute, depending on alloy composition and ingot size and shape. Such limitation of casting speed is undesirable from the standpoint of productivity of the casting operation; it would be very advantageous, as enabling more rapid, efficient and economical'ingot production, to cast ingots at speeds substantially greater than those now attainable without producing hot cracking.

Efforts previously made to eliminate hot cracking have been predicated on the assumption that hot cracking and cold cracking might be prevented in the same way; but expedients used to prevent cold cracking have not proven successful in preventing hot cracking, and have indeed sometimes increased hot cracking susceptibility. In particular, it has heretofore been proposed to attempt to prevent hot cracking by reducing the intensity of cooling in all or part of the direct chill cooling zone, on the theory that hot cracking was caused by residual tensile stresses in the cast ingot, and that such stresses would be minimized if the ingot surface were kept at a higher than normal temperature to reduce the temperature differential between the ingot core and surface through and beyond the region of core solidification. Expedients suggested for this purpose have included reduction in volume and cooling efficiency of the supplied direct-chill coolant, with use of a fog spray or pulsed water supply to the ingot surface rather than a steady, impinging stream of water, or alternatively, removal of the cooling water from the ingot surface after initial impingement on the surface, by a so-called wipe-off operation. It has been found, however, that these expedients, with the possible exception of fog spray, do not enable any material increase in casting speed, i.e., without production of center cracking, and in at least some instances reduced cooling or wipe-off techniques appear even to enhance the susceptibility of the ingots to hot cracking.

While the foregoing considerations have been discussed with reference to aluminum casting operations in which the ingot advances along a downward vertical path, it will be understood that they are also applicable to other continuous or semicontinuous casting operations (such as operations wherein the casting path does not have a vertical orientation), and in varying degree, to the direct chill casting of metals other than aluminum.

SUMMARY OF THE INVENTION The present invention broadly embraces the discovery that advantageously superior freedom from center cracking in direct chill cast ingots, and very significantly enhanced casting speeds in direct chill casting operations, can be achieved by procedure including subjecting an ingot (advancing from a continuous casting mold) to the action of two separate direct chill cooling zones positioned in succession along the path of ingot advance, wherein the first zone, extending from the mold for a predetermined distance along the path of ingot advance, provides a relatively restricted cooling intensity (as represented by magnitude of ingot-coolant heat transfer coefficient) and the second zone, located at that predetermined distance from the mold, provides a substantially greater intensity of cooling. The direct chill cooling is effected by respectively directing separate supplies of coolant fluid (e.g., water) onto the ingot surface in the two zones, in such manner as to provide the specified cooling intensities in the two zones. The first zone cooling is of such intensity as to maintain the peripheral portion of the ingot in solid state but preferably without completely solidifying the ingot core, while the cooling in the second zone effects complete solidification of the core and at the same time provides a high rate of cooling of the peripheral portion of the ingot. Thus, in this procedure, the ingot encounters the greatest intensity of cooling at or adjacent the locality (in its path of advance) at which the core becomes fully solidified, in contrast with present conventional practice, wherein the ingot encounters the greatest intensity of cooling immediately beyond the outlet end of the mold, and ordinarily at least somewhat ahead of the locality of core solidification.

In the practice of the invention, a suitable range of positions for the locality of application of the second zone coolant is that in which the distance between the point of complete solidification of the ingot core and the point of initial impingement of the second-zone coolant on the ingot (as measured along the path of ingot advance) is not more than about one fourth the minimum transverse dimension of the ingot. Preferably, especially when a liquid such as water is used as the second zone coolant, the locality of application of the second-zone coolant is ahead of the point of core solidification, as it takes some time for water or like coolant to begin to cool the ingot at the rate required at the core solidification point. However, the locality of impingement of second-zone coolant may be disposed beyond the point of core solidification (i.e., within the stated range of positions) if the coolant used extracts heat from the ingot sufficiently quickly so that the desired cooling rate can be achieved at the core solidification point, ahead of the point of coolant application.

It is presently preferred, in use of a coolant such as water, to apply the coolant at a locality which is spaced ahead of the point of complete core solidification, along the path of ingot advance, by a distance equal to about one sixth the minimum transverse dimension of the ingot. This may be contrasted with conventional direct chill casting operations, utilizing a single direct chill cooling zone beyond the mold, wherein the point of application of the direct chill coolant (water) is usually spaced ahead of the point of complete core solidification by a distance (measured along the path of ingot advance) equal to about one half to two thirds the minimum transverse dimension of the ingot.

Initial cooling of the ingot in the mold is performed in such manner as to maintain (within the mold) an average coefficient of heat transfer from the ingot to the mold sufficient to produce a thin solid ingot shell at the mold outlet end having a thickness adequate to withstand frictional stresses between the mold and the ingot. Further particular features of the invention, in specific aspects thereof, reside in the provision of an average coefficient of heat transfer from ingot to coolant in the first zone equal to between about one and about six times the average heat transfer coefficient in the mold and preferably equal to at least about twice the average heat transfer coefficient in the mold, and the provision of an ingot-coolant heat transfer coefficient in the second zone equal to at least about one and one half times (preferably at least about five times) the average heat transfer coefficient in the first zone.

While broadly applicable to the casting of a wide variety of metals, the procedures of the invention afford special advantages for the casting of aluminum ingots, in overcoming the particularly serious center cracking problems that have heretofore limited casting speed in production of such ingots, and the invention in one specific sense is directed particularly to aluminum casting procedures. In such procedures, the average coefficient of heat transfer from the aluminum to the mold is typically about 0.05 calories/cmlsecond/T; preferably, the average coefficient of heat transfer from the ingot surface to the coolant liquid in the first direct chill cooling zone is between about 0.1 and about 0.2 caloriesIcm /secondPC; and also preferably, the coefficient of heat transfer from the ingot surface to the coolant liquid in the second zone is at least about 0.5 calories/cm /second/C.

it is found that the present invention enables production of sound, crack-free ingots even in casting operations wherein the sump depth (distance of the core solidification locality from the mold outlet end) is substantially greater than the minimum transverse dimension of the ingot being cast; i.e., the invention overcomes the limitation as to sump depth heretofore considered essential for avoidance of center cracking. Thus the invention permits use of casting speeds far in excess of conventional ranges, with maintained freedom from hot cracking. This is so despite the fact that for any given set of casting conditions (ingot dimensions, alloy composition and casting speed), the sump depth in an ingot being cast by the present process will ordinarily be somewhat greater than that-in an ingot being cast by conventional procedure, owing to the reduced intensity of cooling (as compared with the intensity of cooling encountered by the emerging ingot in conventional practice) in the first direct chill cooling zone.

In the present process as in prior practice, the sump depth (for given conditions of ingot dimensions, alloy composition and cooling intensities) is directly-related to the casting speed; as the casting speed increases, so does the sump depth. Accordingly, the distance from the mold to the second cooling zone is selected, with reference to the contemplated casting speed, so as to position the second cooling zone within the abovedefined range of positions in relation to the locality in the path of ingot advance at which the sump terminates (i.e., the point at which the core of the ingot becomes completely solid), thereby to provide the desired increase in cooling intensity at that locality. Again in contrast to conventional casting operations, it is found that the procedure of the present invention affords superior flexibility of operation especially with respect to casting speed; if it is desired to increase or decrease the casting speed, the second cooling zone is positioned farther from or closer to the mold to accommodate the corresponding change in sump depth, and the advantages of the invention in preventing center cracking are again realized at the new casting speed.

Without limitation of the invention by any particular theory, it is at present believed that the advantages of the invention are attributable in particular to the effect of the described direct chill cooling steps and condi tions on the cooling rate of the peripheral portion of the ingot at the locality (in the path of ingot advance) at which the ingot core becomes completely solidified.

It is further believed at present that hot center cracking of ingots in direct chill casting operations is a consequence of excessive tensile stresses developed within the ingot at the locality at which core solidification becomes complete. The tensile strength of the metal is at a minimum within a few degrees of the solidus point, and hence the core immediately after solidification is particularly susceptible to tensile stresses. Specifically, it is believed that crack-producing tensile stresses may be created at the locality of core solidification in the path of ingot advance by an excessive disparity between the cooling rates (and hence the rates of contraction) of the ingot core and periphery at that locality. The core metal, at the point of solidification, undergoes rapid cooling, and concomitantly rapid contraction; if the cooling and contraction rates of the peripheral portion of the ingot at the critical locality are too low in relation to the cooling and contraction rates of the core, center cracking (according to the present theory) results.

In conventional direct chill casting practice, with the greatest intensity of cooling applied immediately beyond the outlet end of the mold and usually substantially ahead of the locality at which the core of the advancing ingot solidifies, the peripheral portion of the ingot is very rapidly reduced in temperature as it emerges from the mold, and is thereafter cooled at a progressively diminishing rate as the ingot proceeds along its path of advance from the mold. Thus at the locality of core solidification, the cooling rate of the ingot periphery may be very low in relation to the core cooling rate, especially as the casting speed is increased, since increase in casting speed displaces the locality of core solidification progressively farther from the locality of greatest cooling intensity. Accordingly, the present theory would indicate (as is in fact the case) that increase in casting speed, in conventional direct chill casting operations, enhances the likelihood of center cracking. Reduced cooling and wipe-off techniques, heretofore proposed for avoidance of center cracking, do not diminish and may even aggravate the disparity between ingot core and peripheral cooling rates at the locality of core solidification.

In the present procedure, in contrast, the cooling rate of the peripheral portion of the ingot is significantly higher (i.e., closer to the cooling rate of the core) at the locality of complete solidification of the core, than in conventional practice, for any given casting conditions and given casting speed. Owing to the relatively reduced cooling intensity in the first direct chill cooling zone of the invention, the temperature of the ingot periphery remains relatively high (in comparison to conventional practice) as the ingot approaches the locality of complete core solidification. The maintained high peripheral temperature of the ingot facilitates attainment of a high peripheral cooling rate at the critical locality, since the cooling rate is dependent on the temperature differential between the ingot periphery and the applied coolant; and this high cooling rate is then achieved by the application of intense cooling in the second zone. Thus the direct chill cooling steps and conditions of the invention cooperate to produce a substantially higher ratio of ingot periphery cooling rate to ingot core cooling rate at the locality of core solidification, resulting in a closer match between-core and peripheral contraction rates at that locality, than in conventional practice.

Further features and advantages of the invention will be apparent from the detailed description hereinbelow set forth, together with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic elevational sectional view showing the practice of the present procedure in an illustrative embodiment;

FIG. 2 is an elevational sectional view of one specific form of direct chill casting apparatus arranged to provide direct chill cooling in the first zone of the present procedure;

FIG. 3 is a similar view of an alternative form of apparatus for providing the first zone direct chill cooling;

FIG. 4 is a graph in which the temperature of various points within an aluminum ingot cast in accordance with the present procedure is plotted as a function of time of advance of such points beyond the casting mold; and

FIG; 5 is a graph similar to FIG. 2, showing the temperature of comparable points in an aluminum ingot cast by conventional procedure plotted as a function of time of advance of such points beyond the casting mold.

DETAILED DESCRIPTION Referring to the drawings, FIG. 1 illustrates in simplified schematic view one form of apparatus for continuously casting aluminum ingots in accordance with the present invention. This apparatus (which is arranged for so-called vertical casting, i.e., for casting operations wherein the ingot descends vertically from the mold as it is cast) includes an axially vertical annular mold 10 (open at its lower end) to which molten aluminum metal 11 is supplied for casting an ingot 12. The mold 10, fabricated of a metal suitably resistant to deterioration under conditions for casting aluminum, has a vertical inner wall 14 which defines an axially vertical casting zone 14a of desired horizontal cross section, it being understood that the mold wall configuration determines the cross-sectional shape of the produced ingot; by way of specific illustration,

reference will be made herein to an annular mold wall 14 which is cylindrical, i.e., circular in cross section, for producing a cylindrical ingot.

Surrounding the outer surface of the mold wall 14 is a cooling jacket 15, shown for simplicity as formed of further wall portions cooperating with the wall 14 to define and enclose an annular chamber 15a completely laterally surrounding the casting zone. This enclosed chamber is supplied as through a pipe 15b under control of a valve 15c with a cooling fluid such as water for chilling the mold wall 14, and is preferably kept continuously filled with a flowing or circulating body of the fluid, designated 16.

Located within the cooling chamber 15a is an annular baffle 17 disposed in concentric outwardly spaced relation to the mold wall 14 and extending vertically upward from the floor of the cooling chamber for a distance somewhat less than the height of the chamber. This baffle directs the circulating flow of water within the chamber 15a in such manner as to afford desired efficacy of cooling of the mold wall.

At the start of the casting operation, the lower end of the casting zone 14a is closed by a stool or platform 18 which is supported on a hydraulic ram 20. As molten aluminum in the casting zone solidifies around the lower portion of the periphery of that zone, the stool 18 is drawn vertically downward by operation of the ram 20. The solidifying base of the ingot being cast, resting on the stool, then begins to emerge from the lower end of the casting zone.

The mold apparatus is arranged to direct a spray of cooling fluid onto the emerging solidified ingot surface immediately below the casting zone. Thus, as shown, an annular slit 22 (or a plurality of slits or openings disposed in annular array) may be provided in the lower wall of the cooling jacket 15, extending entirely around the mold periphery and oriented to direct water from the chamber 15a of the cooling jacket onto the surface of the emerging ingot with substantially uniform distribution of water around the periphery'of the ingot. This spray of water, impinging on the ingot surface, acts to enhance the cooling of the ingot as it moves downwardly away from the mold. In the arrangement shown, water is continuously supplied to the chamber 15a and is continuously discharged through the slit 22 onto the ingot surface, so that there is a continuous flow of coolant fluid for removing heat from the solidifying metal.

Molten aluminum metal is continuously supplied to the upper end of the casting zone 14a as through a dip tube 24 that opens downwardly into the upper portion of the casting zone, so as to maintain the pool of molten metal in the casting zone at a substantially constant level as the solidifying ingot is progressively withdrawn from the mold, i.e., as the stool is drawn downwardly.

During the described continuous casting operation, molten metal within the casting zone 14a solidifies around the periphery of the mold wall 14 as it is cooled by heat transfer to the externally chilled mold surface. This solidification progresses sufficiently far inward toward the center of the mold so that the ingot emerging from the lower end of the mold has an externally solid and self-sustaining shell 25 even though the central portion or core 26 of the emerging ingot is still molten. With an effectively continuous supply of molten metal to the mold, and correspondingly continuous downward advance of the cast ingot from the mold, the molten central portion 26 of the ingot emerging from the mold extends downwardly as a molten metal sump (constituting the lower end of the molten metal pool in the mold) of progressively decreasing cross section in a downward direction; in other words, aided by the effect of the cooling spray supplied through slit 22, the emerging ingot progressively solidifies toward its center until, at a level 27 located at some distance below the lower end of the mold, the core of the ingot becomes entirely solid.

The structures and procedures thus far described are, broadly speaking, conventional in present-day commercial direct chill continuous casting of aluminum ingots. Important features of the present invention reside in the provision of special direct chill cooling conditions below the mold, and more particularly in the provision of two direct chill cooling regions successively traversed by the descending ingot below the mold, as will now be explained.

The first of the direct chill cooling zones provided in accordance with the invention extends from the outlet end of the mold for a predetermined distance along the path of advance of the ingot below the mold, and is designated 28. In this zone, a first supply of coolant fluid, e.g., water, is directed onto the surface of the ingot for cooling the ingot, in a manner providing an average coefficient of heat transfer from the ingot to the coolant fluid in the first zone having a value effective to maintain the ingot shell 25 in solid state while maintaining the core 26 of the ingot in molten state throughout the first zone.

The second direct chill cooling zone, designated 30, is located at the last-mentioned predetermined distance from the mold outlet end. In this second zone, a second supply of coolant fluid (e.g., water), separate from the coolant fluid supplied in the first zone 28, is directed onto the ingot surface for cooling the ingot in a manner providing a coefircient of heat transfer from the ingot surface to the coolant fluid in the second zone substantially greater than the average coefficient of heat transfer in the first zone and effective to produce complete solidification of the ingot.

In the arrangement shown in FIG. 1, the supply of coolant fluid in the first direct chill zone 28 is the stream of water discharged onto the ingot surface from the cooling jacket 15 through the slit or slits 22. In a conventional direct chill mold for vertical casting, the slit 22 is oriented to direct the spray against the ingot surface at a substantial angle (e.g., 30 to 45) to the vertical, so that despite vaporization of the water (effected by heat from the ingot shell surface) water in liquid state and in substantial volume comes into direct contact with the ingot surface immediately below the ingot mold. In accordance with the embodiment of the invention illustrated in FIG. 1, however, the slit 22 is oriented to direct a spray of water against the ingot at a substantially flatter angle (e.g., an angle of about to the vertical), and the volume of water thus discharged is also substantially reduced, as compared with conventional practice. For example, the volume of water discharged through slit 22 may be approximately half that conventionally so discharged in the casting of an ingot of given size, configuration and composition; this reduction in volume of discharged water may be accomplished by appropriate control of volume of water introduced to the cooling jacket.

In comparison with conventional vertical casting practice, the reduced angle of impingement of the water spray from slit 22, together with the reduced volume of sprayed water, greatly decreases the contact of liquid water with the ingot surface immediately below the mold, especially since at the flattened angle of impingement of the spray there is a greater extent of vaporization of the water by the heat of the ingot before the water can reach the ingot surface. Consequently, throughout the first direct chill cooling zone, the cooling intensity (i.e., the average coefficient of heat transfer from the ingot surface to the applied coolant over the extent of the first cooling zone) is very materially less than the cooling intensity immediately below the mold in conventional direct chill casting operations. It will be understood that this effect of reduction in impingement angle is found to occur in vertical casting operations, but in horizontal casting (wherein the ingot being formed advances along a horizontal path from the mold) a small angle of impingement tends to increase cooling efficiency. Thus, in use of the invention in horizontal casting operations, the first-zone cooling is controlled in other ways, e.g., such as the alternatives hereinafter described with reference to FIGS. 2 and 3.

To provide the second direct chill cooling zone of the invention, in the embodiment of FIG. 1, a sub-mold cooling jacket or water ring 32 is positioned at the lower end of the first cooling zone 28, in surrounding relation to the descending ingot. The water ring 32 comprises an annular enclosed chamber, to which water is continuously supplied through an inlet pipe 34 controlled by a valve 35, and surrounds the ingot concentrically. The inner wall 36 of the ring 32 is spaced outwardly from the surface of the ingot for a sufficient distance to provide clearance for descent of the stool l8 and the ingot.

A vertically spaced annular slit 37 (or annular array of openings), surrounding the entire ingot and communicating with the interior of the water ring 32, is provided in the ring inner wall 36 for directing a stream of water from the ring onto the ingot surface. The slit 37 is oriented to direct the last-mentioned stream of water onto the ingot surface at a substantially greater angle to the vertical than the spray directed by the slit 22 described above; for example, the slit 37 may be oriented to direct water onto the ingot surface at an angle of 30 to 45 to the vertical, corresponding to the angle at which water is usually directed onto an ingot surface from the lower end of a direct chill casting mold in conventional casting operations as heretofore practiced. In addition, the volume of water directed through the slit 37 to the ingot surface (determined by the volume of water supplied through pipe 34 to the ring) is substantially greater than that discharged through slit 22, being (for example) approximately equal to the volume of water customarily directed onto an ingot surface at the lower end of the mold in a conventional direct chill casting operation. Owing to the fact that water is discharged from the ring 32 onto the ingot surface in greater volume and at a greater angle of impingement than the water discharged through slit 22, liquid water in substantial volume comes into direct contact with the ingot surface in the second cooling zone 30. Thus, in the second cooling zone, there is provided a substantially greater cooling intensity (coefficient of heat transfer from the ingot surface to the applied coolant) than in the first zone 28.

As will now be understood, the practice of the present method in the apparatus of FIG. 1, for continuously casting an aluminum ingot, includes the steps of supplying molten aluminum through the dip tube 24 to the inlet end of the mold 10, while cooling the mold with water in the cooling jacket 15) for solidifying the peripheral portion of the aluminum therein to form an ingot 12 having an externally solid shell 25 and an initially molten core 26, and while continuously advancing the ingot through and beyond the outlet end of the mold (by effecting continuous downward movement of the stool l8), and while cooling the ingot beyond the mold for progressively solidifying the molten core to produce a completely solidified ingot, the molten core extending as a molten metal sump within the ingot for a predetermined distance beyond the mold outlet end along the path of advance of the ingot. Specifically, in the method of the invention, the ingot-cooling step comprises successively advancing the ingot through the first and second direct chill cooling zones defined above while providing in these respective zones the above-described cooling conditions. This procedure is continued until an ingot of desired length has been cast.

As stated, the cooling conditions provided in the regions successively traversed by the descending ingot constitute especially important features of the invention. Initial cooling of the supplied molten metal occurs within the mold, wherein the cooling conditions are such as to provide an average coefficient of heat transfer from the metal to the mold sufficient to produce at the mold outlet end a thin solid ingot shell having a thickness adequate to withstand frictional stresses between the mold and the ingot. In the method of the invention, the cooling conditions within the mold may be essentially comparable to those heretofore conventionally employed and may, for example, provide an average coefficient of heat transfer from the aluminum to the mold of about 0.05 calories/cmlsecondPC.

The average coefficient of heat transfer from the ingot to the coolant fluid in the first direct chill cooling zone 28, over the extent of that zone, is maintained (by appropriate control of the volume and/or manner of supply of coolant fluid) equal to between about one and about six times the value of the aforementioned average coefficient of heat transfer in the mold, and is preferably equal to at least about two times the value of the average coefficient of heat transfer in the mold. A presently preferred range of values for the average coefficient of heat transfer from the ingot surface to coolant liquid in the first zone 28, for casting of aluminum ingots, is between about 0.1 and about 0.2 calories/cm /second/C. As stated, the cooling intensity thus provided in the first zone is such as to maintain the ingot shell 25 in solid state and to solidify the major portion of the ingot cross section while maintaining at least the central core portion of the ingot in molten state throughout the extent of the first zone.

In the second direct chill cooling zone 30, the coefficient of heat transfer from the ingot surface to the coolant fluid is maintained (again by appropriate control of the volume and/or manner of supply of coolant fluid to the ingot surface in the zone 30) equal to at least about one and one half times the value of the aforementioned average coefficient of heat transfer in the first zone 28, and preferably equal to at least about five times the value of the average coefficient of heat transfer in the first zone. In presently preferred practice for the casting of aluminum ingots, the coefficient of heat transfer from the ingot surface to coolant liquid in the second zone 30 is at least about 0.5 calories/cm /second/C, i.e., about equal to or greater than the ingot-coolant heat transfer coefficient provided immediately below the mold in conventional operations for direct chill casting of aluminum ingots. The second zone cooling intensity is, as stated, effective to produce complete solidification of the ingot, the second zone having a vertical extent (in the embodiment of FIG. 1) sufficient to effect such complete solidification, i.e., to solidify the central core that remains molten throughout the first zone.

As already indicated, in the process of the invention the second direct chill cooling zone 30 is positioned adjacent the level 27 at which the core of the advancing ingot becomes completely solidified. Specifically, the second zone coolant water from slit 37 impinges on the surface of the descending ingot at a level which (in the illustrated embodiment of the invention) is spaced above the level of the sump extremity 27 by a distance equal to about one sixth the minimum transverse dimension (in this case, the diameter) of the ingot.

Stated more generally, the point of impingement of the second zone coolant on the ingot surface should be spaced from the extremity of the molten metal sump by a distance (along the path of ingot advance) equal to not more than about one fourth the minimum transverse dimension of the ingot. In use of a liquid such as water to provide the second zone coolant, it is preferred that the second zone coolant impinge on the ingot ahead of the sump extremity, and it is specifically preferred that the spacing between this locality of impingement and the sump extremity along the path of ingot advance be equal to about one sixth the minimum transverse dimension of the ingot, for attainment of the desired cooling effect at the locality of complete solidification of the ingot core. The predetermined distance for which the first direct chill cooling zone 28 extends below the mold is that between the lower end of the mold and the level at which the second zone coolant impinges on the ingot. Thus the ingot encounters a relatively abrupt increase in cooling intensity (i.e., incident to passing from the first cooling zone 28 to the second cooling zone 30) at or adjacent the level at which the core becomes completely solid.

For an ingot of given dimensions and compositions, the depth of the molten sump 26 below the mold is dependent on the cooling conditions encountered by the advancing ingot and on the speed of ingot advance. Owing to the reduced intensity of cooling in the first zone 28 (as compared with the cooling intensity encountered by the ingot immediately below the mold in conventional practice), the sump depth is greater for any given casting speed in the procedure of the present invention than in conventional direct chill casting operations. The casting speed (i.e., rate of ingot advance from the mold) may also be greater in the present procedure than has heretofore been possible, providing still further increase in sump depth.

In thearrangement of FIG. 1, the water ring 32, which constitutes the lower terminus of the first cooling zone and provides the second cooling zone, is positioned adjacent the level in the path of ingot advance at which the core becomes completely solid, as already explained. As will be understood from the foregoing discussion, this position of the water ring is determined inter alia by the desired casting speed. With increase in casting speed, and concomitant increase in sump depth, the water ring is positioned further below the mold so as to maintain the desired positional relationship between the second cooling zone and the lower end of the sump (level 27).

To summarize, comparing the procedure of the present invention with conventional practice, it will be noted that the cooling conditions within the mold may be essentially the same as in present-day conventional operations. However, the cooling intensity in the first direct chill zone (encountered by the ingot as it emerges from the mold) is very substantially lower than the intensity of cooling encountered by the emerging ingot in conventional practice. The intensity of cooling applied to the ingot in the second zone, at the predetermined distance from the mold adjacent the locality at which the core becomes completely solidified, may be comparable to (or greater than) the cooling intensity encountered by the ingot as it emerges from the mold in conventional practice; but this region of most intense cooling is, as stated, spaced away from the mold by the aforementioned predetermined distance, rather than (as in conventional operations) being located immediately adjacent the mold and substantially ahead of I the locality of complete core solidification.

The procedure of the present invention enables production of sound, crack-free ingots at casting speeds very substantially greater than those attainable in prior practice without center cracking. This advantage is believed attributable to the fact that the present procedure provides a substantially higher cooling rate of the ingot periphery, at the level of core solidification, than is achieved in conventional operations. The high peripheral cooling rate at such level is attained by the application of a high intensity of cooling in the second cooling zone, and also by the relatively low intensity of cooling in the first cooling zone, which maintains the ingot periphery at a comparatively high temperature so as to enhance the rate of cooling of the periphery achieved by the cooling in the second zone. Provision of a high peripheral cooling rate at the level of core solidification reduces the difference in contraction rates between the strongly cooling core and the ingot periphery at such level and hence minimizes the tensile stresses between the core and the periphery which (as is now believed) have heretofore tended to cause center cracking in direct chill cast aluminum ingots. In particular, the casting speed in the present procedure is not limited by the conventional requirement that the sump depth below the outlet end of the mold be no greater than the minimum transverse dimension of the ingot, for avoidance of center cracking.

While the above-described use of reduced volume of water directed onto the ingot at a flattened angle of impingement (as compared with conventional direct chill casting practice) represents one convenient way of providing the desired relatively low intensity cooling in the first direct chill cooling zone, such cooling may be accomplished in other ways e.g., similarly providing significant reduction in contact of the ingot surface by liquid water immediately below the mold. For example, a fog spray or mist comprising droplets of water entrained in a flow of air or other gas may be directed onto the ingot surface below the mold to provide the first zone cooling, or a pulsed (i.e., intermittent) fiow of water may be directed onto the ingot surface at that locality, in place of the conventional continuous flow.

Referring now to FIG. 2, there is shown an alternative form of mold construction arranged to provide the relatively low-intensity first zone direct chill cooling in the process of the present invention. This apparatus includes an axially vertical annular mold wall 40 to which is secured structure defining a water box or cooling jacket 42 externally surrounding the mold wall. Water supplied by suitable means (not shown) in continuous flow to the jacket 42 chills the mold wall to effect cooling of molten aluminum contained within the mold.

Mounted within the jacket 42 is an annular baffle 44, concentrically surrounding the mold wall in adjacent but outwardly spaced relation to the external surface of the wall and projecting upwardly from the floor of the jacket to a locality adjacent but somewhat below the top of the jacket. Water supplied to the jacket flows over the top of the baffle and down through the restricted annular space 45 defined between the baffle and the mold wall to effect desired chilling of the mold. At its lower end, the space 45 opens into an annular chamber 46, from which the water is discharged, in a manner hereinafter described, after descending through the space 45.

Mounted within the chamber 46 is a ring 48, which concentrically surrounds the mold wall 40 and cooperates with the floor of the cooling jacket to define a second annular chamber 50. The major flow of the water entering chamber 46 from the space 45 passes over the upper surface of the ring 48 and is discharged from the chamber 46 through a plurality of relatively large outlet passages 52. However, a minor flow of the water in chamber 46 enters the second chamber 50 through a plurality of holes 54 which are formed in the ring 48 and are individually very substantially smaller in diameter than the outlet passages 52. Fluid from the chamber 50 is discharged downwardly toward the surface of the ingot emerging from the mold through an annular slit 56 (or an annular array of slits or openings) formed at the lower end of the mold wall; the slit 56 is oriented to conduct fluid from chamber 50 and to direct it as a jet or spray at an acute angle to the emerging ingot surface.

An annular air manifold 58 is mounted directly beneath the chamber 50 and communicates therewith through a plurality of holes 60 equal in diameter to, and positioned in register with, the holes 54 which admit water through ring 48 to the chamber 50. Air, forced into the manifold by suitable means (not shown) through a plurality of passages 61, flows upwardly into the chamber 50 through the holes 60, opposing the downward flow of water into the chamber through the holes 54. This air mixes with the water in chamber 50. The air-water mixture is expelled from the latter chamber toward the ingot through the slit 56 as a fog or mist, comprising fine droplets of water entrained in the forced air flow; these droplets are vaporized by the heat of the ingot, providing a layer of steam around the ingot which affords some cooling of the ingot but at a substantially reduced cooling intensity as compared to a steady stream of water impinging on the ingot in liquid state. The cooling intensity in this embodiment can readily be controlled, to provide desired cooling conditions in the first direct chill cooling zone of the present process, by adjustment of the supply of air to the manifold 58.

By way of example, in one illustrative structure embodying the features shown in FIG. 2, used with a mold having a vertical depth of 4 "/8 inches for casting a 6- inch diameter aluminum ingot, the radial dimension of the space 45 was as inch. Six water outlet passages 52 were provided, spaced 60 apart around the circumference of the ingot, and six air inlet passages 61 were similarly spaced 60 apart around the mold periphery. The holes 54 and 60, each 1/16 inch in diameter, were spaced inch apart around the mold periphery. The annular slit 56 had a width in a range between 0.030 inch and 0.060 inch.

The described arrangement for providing the first zone direct chill cooling in the process of the invention may be substituted for the mold arrangement shown in FIG. 1. One advantage of the system of FIG. 2 is the ease with which cooling conditions can be varied over a wide range of cooling intensities, even while a casting operation is in progress, by adjustment of air supply to the manifold, which changes the air-to-water ratio. Also, the mold may be used for conventional casting operations, if desired, without structural modification; i.e., the mold water may simply be discharged in continuous flow (without aeration) through the slit 56, in the same manner as in a conventional mold.

A further alternative arrangement of mold apparatus for providing the first zone cooling in the present process is illustrated in FIG. 3. This arrangement is adapted to provide a pulsed discharge of water from the lower end of the mold onto the surface of the emerging ingot. As in the structures described above, the apparatus of FIG. 3 (which would replace the mold l0 and cooling jacket 15 in the system of FIG. 1) includes an axially vertical annular mold wall 64 adapted to receive molten aluminum for continuous casting of an ingot, and structure providing a cooling jacket 66 laterally surrounding the mold wall. Water supplied by suitable means (not shown) to the cooling jacket in continuous flow circulates through the jacket and chills the mold wall externally. Also as in the abovedescribed mold structures, an annular baffle 68, mounted within the cooling jacket in spaced but closely surrounding relation to the mold wall, defines an annular space 70 between the baffle and mold wall,.open at its upper end; the cooling jacket water flows over the top of the baffle and downwardly through the space 70, chilling the mold wall.

The inner surface 72 of the baffle 68 extends vertically downward from the upper end of space 70, in facing parallel relation to the outer surface of the mold wall. In the lower portion of the cooling jacket, the baffle surface 72 slopes downwardly and inwardly toward the mold wall as indicated at 73, then again extends vertically downwardly as indicated at 74, and finally slopes downwardly and outwardly away from the mold wall as indicated at 75 to the lower extremity of the cooling jacket. This surface 72 75 provides the normal path for flow of water through and beyond the space 70 and owing to the outward slope of the lowermost portion 75 of the described surface, water following such normal path is discharged from the space 70 in a direction away from the surface of the ingot 12.

The outer surface 77 of the mold wall 64 extends vertically downward to a level slightly below the inwardly sloping portion 73 of the baffle surface 72. As will be seen from FIG. 3, the portion of space 70 defined between the lowermost extent of this vertical portion of the mold outer surface and the portion 74 of the mold inner surface is restricted in width (as compared to the upper portion of space 70) owing to the inward slope of the baffle surface at 73.

At a locality opposite the baffle surface portion 74, a shoulder 78 is formed in the mold wall surface 77, and a further short vertical portion 80 of surface 77, offset inwardly with respect to the major extent of surface 77, extends downwardly from the shoulder 78. Below surface portion 80, the mold outer surface slopes downwardly and inwardly toward the ingot, as indicated at 81, to the lower end of the mold.

Means are provided in the structure of FIG. 3 for controllably diverting the flow of water from the baffle surface portion 75 to the mold wall surface portion 81, which directs the water inwardly toward and against the ingot surface. Specifically, outwardly of the baffle surface portion 74 there is provided an annular chamber 83, concentrically surrounding the mold wall and communicating with the lower portion of the space 70 through plural axially horizontal holes 84 which open into the space 70, as shown, at a level slightly below the shoulder 78 of the mold wall surface 77. These holes 84 are relatively small in diameter, e.g., ll 16 inch.

A fluid such as water or air is supplied to the chamber 83 through means illustrated schematically as a conduit 86 under control of a valve 87, which may be electrically operated and itself controlled by a suitable timing device 88 for effecting intermittent supply of fluid to the chamber 83. When the valve 87 is open, the pressure developed in chamber 83 by the supply of fluid through the conduit 86 forces the supplied fluid through the holes 84. This secondary fluid flow reacts with the main flow of water descending through the space past the holes 84 in such manner as to deflect that main flow of water against the mold wall surface portion 81. Upon closing of the valve 87, the secondary fluid flow through holes 84 ceases, and the main water flow returns to the baffle surface portion 75. The recess formed by shoulder 78 in the mold wall outer surface is vented to the atmosphere through small passages 90 formed in the mold wall, in order to ensure that the main water flow will not remain attached to the surface portion 81 (by the so-called Coanda effect) after the deflecting force of the secondary fluid flow is shut off.

The operation of the apparatus of FIG. 3 to provide the desired first zone cooling in the process of the present invention may now be readily understood. With continuous supply of cooling water to the jacket 66 and correspondingly continuous flow of the water downwardly through space 70 and thence out of the lower end of the mold, the timing device 88 is operated to cause regular intermittent opening and closing of the valve 87 and thereby to cause regular, intermittent supply of fluid through the conduit 86 to the chamber 83. Accordingly, the flow of water descending past the holes 84 in the space 70 is intermittently subjected to the deflecting action of the secondary fluid flow through the holes 84. Each time that secondary flow is applied, the main water flow is diverted to surface portion 81 and is thus directed against the surface of the ingot emerging from the mold. Each time the secondary flow is interrupted, the main water flow returns to the surface portion and is diverted away from the ingot. Thus the ingot is subjected to a pulsed or intermittent stream of water in the first cooling zone rather than to a continuous stream. The frequency and duration of the pulses, and hence the supply of water per unit time to the ingot surface (which determines cooling intensity) are readily controlled by means of the timing device 88.

As one further example of the process of the invention, the casting mold may be positioned immediately above a pit filled with water through which the emerging ingot descends, the pit and mold being so arranged that the ingot enters the water as it emerges from the mold. The heat of the ingot vaporizes water to form a jacket or barrier of steam thatsurrounds the descending ingot and inhibits contact of liquid water with the ingot surface. Within the pit, and spaced below the outlet end of the mold, there is provided awater ring (generally similar to the ring 32 of FIG. 1) which directs jets of water against the ingot surface; these jets penetrate the steam barrier to provide direct contact of the ingot surface with liquid water. In such arrangement, the first cooling zone is the portion of the ingot path of advance between the outlet end of the mold and the water ring; in this zone, the steam barrier provides a relatively low intensity of cooling. The second cooling zone is provided by the water ring, which effects cooling of higher intensity by causing contact of the ingot surface with liquid water. The relative positions of the first and second zones are as defined above the with reference to FIG. 1. Also, the arrangement of apparatus elements may be essentially as shown in FIG. 1, with the mold water discharge slit 22 omitted, and with coolant to the ingot in the first direct chill zone at plural localities spaced along the zone, to insure maintenance of a solid ingot shell throughout the zone. It will be understood that reference herein to the step of directing a first supply of coolant onto the ingot in the first direct chill zone embraces operations wherein that coolant supply is directed onto the ingot from plural sources and/or at plural localities along the path of ingot advance, within the first zone; and it will be further understood that where the first zone coolant is supplied from plural sources and/or at plural localities, such supply is controlled to provide throughout the first zone the cooling conditions described above, i.e., the specified conditions of average heat transfer coefficient.

The effect of the present process in reducing the disparity in cooling rates between the core and periphery of an ingot at the locality of core solidification is illustrated in FIGS. 4 and 5, which show graphically the temperature of various points spaced radially outward from the core of an ingot, as a function of time of advance of such points from a casting mold, in ingots cast respectively by the present process and by conventional direct chill casting procedure. Both ingots were 6-inch diameter cylindrical ingots cast from the aluminum alloy identified by the Aluminum Association designation AA6063, at a casting speed of nine inches per minute. The ingot temperatures were measured by thermocouples implanted in the ingot in a common horizontal plane and at various distances from the core.

In each of FIGS. 4 and 5, curves A, B, C, D and E respectively represent temperatures measured by thermocouples respectively located in a common horizon- 'tal plane at distances of approximately inch, 1 inch, 1 /2 inches, 2 inches and 3 inches from the outer surface of the ingot. Cooling rates at each of these different calities are compared, in each of FIGS. 4 and 5, for the interval during which the temperature of the core of the ingot (curve E) decreased from 650 to 600C, i.e., the range of temperatures just below the temperature at which the core solidifies.

As stated, the ingot of FIG. 4 was cast in accordance with the present procedure, utilizing an arrangement of the type shown in FIG. 1, with the water ring 32 positioned three inches below the outlet end of the mold, and with water discharged from the mold slit 22 at a rate of 6 k imperial gallons per minute and from the water ring at a rate of 35 imperial gallons per minute. As shown in FIG. 4, when the core of this ingot (curve E) began to cool through the 650-600C temperature range, the temperature of the periphery of the ingot as measured by the outermost thermocouple located about $4 inch from the ingot surface (curveA) was about 300C, and the temperature measured by the latter thermocouple decreased by 25 while the core was cooling through 50 from 650 to 600C. Thus, the

ratio of peripheral cooling rate to core cooling rate during the last-mentioned period of core cooling was about 0.5.

The ingot represented by FIG. 5 was, as stated, cast by conventional procedure utilizing a single direct chill cooling zone below the mold with the maximum intensity of cooling applied immediately below the mold outlet end and substantially above the locality of core solidification. In this case, when the core began to cool through the 650 600C temperature range (curve E), the periphery of the ingot (again as measured by a thermocouple located about 56 inch inwardly of the ingot surface, and represented by curve A in FIG. 5) was at a temperature below 250C; and while the ingot core cooled from 650 to 600C, the ingot periphery cooled through only 10C. Hence, in this case the ratio of ingot periphery cooling rate to core cooling rate was 0.2.

In short, the present procedure greatly reduced the disparity in cooling rates between the core and periphery of an ingot at the locality at which the ingot core had just become completely solid and was cooling through the 650 600C temperature range. This result may be attributed both to the heightened intensity of cooling at that locality provided by the present invention and by the higher temperature of the ingot periphery at the point of core solidification, attained by the present invention as a result of the reduced intensity of cooling in the first direct chill cooling zone. The conventionally cast ingot of FIG. 5 exhibited severe center cracking, while the ingot of FIG. 4 cast by the present procedure was sound and crack-free.

By way of further illustration of the procedure of the present invention, reference may be had to the following specific examples of casting of 6-inch diameter cylindrical ingots of grain-refined AA6063 alloy. In each example, the ingot was cast in an axially vertical mold cooled by circulation of water through a surrounding cooling jacket, and the second direct chill cooling zone of the invention was provided by a water ring of the type schematically shown at 32 in FIG. 1, spaced below the mold.

EXAMPLE I An ingot was cast at a speed of nine inches per minute with flow of water through the mold cooling jacket at a rate of 20 imperial gallons per minute and a water flow rate of 15 imperial gallons per minute through the sub-mold water ring which was located 3 inches below the lower end of the mold. The first zone cooling was provided by directing pulsed streams of water (1 second on, 2 seconds off) onto the ingot surface immediately below the mold. The ingot was found to be free of center cracking, although it had isolated surface cracks.

Another ingot was cast by the same procedure except that the first zone cooling was provided by directing aerated water (approximately 5 imperial gallons per minute of water, in mixture with air) onto the ingot surface immediately below the mold, and the water ring was located 3 5: inches below the lower end of the mold. The resultant ingot was entirely crackfree.

EXAMPLE II ing provided by discharge of water from the mold at a rate of 6 imperial gallons per minute through a slit oriented to direct the water onto the ingot surface at an angle of to the vertical. The water ring providing the second zone cooling was located 4 inches below the mold. The ingot was entirely free of cracks.

Another ingot free of cracks was cast at the same speed in the same apparatus, with the water ring positioned 3 inches below the mold, and with water flowing at a rate of 8 imperial gallons per minute through the slit providing the first zone cooling, and at a rate of 35 imperial gallons per minute through the water ring.

EXAMPLE Ill Using the apparatus of Example II, but with the water ring located 7 inches below the mold, an ingot was cast at a speed of 12 inches per minute. The water for the first zone cooling was discharged through the 10 slit at a rate of 10 imperial gallons per minute, and the water from the water ring was discharged at a rate of 35 imperial gallons per minute. The ingot was entirely free of cracks.

In contrast with the foregoing examples, ingots of the same alloy, dimensions and configuration cast by conventional direct chill procedures (i.e., utilizing a single direct chill cooling zone below the mold, with maximum intensity of cooling immediately below the mold) exhibited severe center cracking at casting speeds of 9 and 12 inches per minute, although crack-free ingots were produced in this conventional procedure at a casting speed of 6 inches per minute. Similar ingots cast with pulsed water cooling and with wipe-off of direct chill coolant two inches below the mold, but without use of a high intensity second direct chill cooling zone, also exhibited severe center cracking at a casting speed of 9 inches per minute.

It is to be understood that the invention is not limited to the features and embodiments hereinabove specifically set forth but may be carried out in other ways without departure from its spirit.

I claim:

1. Procedure for continuously casting an ingot, including the steps of a. supplying molten metal to the inlet end of a casting mold having an open outlet end, while b. cooling the mold for solidifying the peripheral portion of the metal therein to form an ingot having an externally solid shell and an initially molten core, and while c. continuously advancing said ingot through and beyond said outlet end of said mold, and while d. cooling said ingot beyond said mold for progressively solidifying said molten core to produce a completely solidified ingot, said molten core extending as a molten metal sump within said ingot to an extremity of the sump beyond said mold outlet end along the path of advance of said ingot; wherein the improvement comprises:

e. the ingot-cooling step comprising i. in a first cooling zone extending from said mold outlet end for a predetermined distance along the path of advance of said ingot, directing a first supply of coolant fluid onto the surface of said ingot for cooling said ingot, in a manner providing an average coefficient of heat transfer from said ingot to the coolant fluid in said first zone having a value effective to maintain said ingot shell in solid state while maintaining the core of said ingot in molten state substantially throughout said first zone; and in a second cooling zone, located at said predetermined distance from said mold outlet end, separately directing a second supply of coolant fluid onto the ingot surface for cooling said ingot, in a manner providing a coefficient of heat transfer from the ingot surface to the coolant fluid in said second zone substantially greater than said average coefficient of heat transfer in said first zone, and effective to produce complete solidification of said ingot, thereby to provide a substantial increase in heat transfer coefficient as aforesaid at about said predetermined distance from said mold outlet end; and

f. the ingot-advancing step comprising advancing said ingot at a rate such that said molten sump terminates, by solidification of the center of said ingot, at a locality adjacent said predetermined distance beyond said mold outlet end, for maintaining the extremity of said sump at a locality adjacent the locality of substantial increase in heat transfer coefficient as aforesaid.

2. Procedure according to claim 1, wherein said average coefficient of heat transfer in said first zone is substantially higher than the coefficient of heat transfer from said metal to said mold within said mold.

3. Procedure according to claim 1, wherein said path of ingot advance is oriented vertically downward; and wherein the step of directing a first supply of coolant fluid onto the surface of said ingot comprises directing a first continuous stream of liquid onto the ingot surface at a preselected acute angle of impingement and wherein the step of directing a second supply of coolant fluid onto the ingot surface comprises directing a second continuous stream of liquid onto the ingot surface at an angle of impingement substantially greater than the angle of impingement of said first stream.

4. Procedure according to claim 3, wherein the volume of liquid per unit time directed onto the ingot surface in said second zone is substantially greater than the volume of liquid per unit time directed onto the ingot surface in said first zone.

5. Procedure according to claim 1, wherein the step of directing a first supply of coolant fluid onto the surface of said ingot for cooling said ingot comprises mixing a flow of gas with a flow of liquid for entraining the liquid in the gas and directing the mixture of gas and entrained liquid onto the ingot surface.

6. Procedure according to claim 1, wherein the step of directing a first supply of coolant fluid onto the surface of said ingot comprises intermittently directing a stream of liquid onto the ingot surface.

7. Procedure according to claim 6, wherein the step of intermittently directing a stream of liquid onto the ingot surface comprises establishing and maintaining a continuous flow of liquid in the direction of advance of the ingot and alternately directing said flow against and away from the ingot surface.

8. Procedure according to claim 1, wherein the step of directing a first supply of coolant fluid onto the surface of said ingot comprises establishing and maintaining a body of liquid in surrounding relation to said ingot at least substantially throughout the extent of said flrst zone, said ingot advancing through said body of liquid in said first zone, and heat from said ingot vaporizing liquid of said body adjacent the surface of said ingot, and wherein the step of directing a second supply of coolant fluid onto the ingot surface comprises directing a stream of said liquid against the ingot surface at a substantial angle of impingement in said second zone.

9. Procedure according to claim 1, wherein said molten metal is aluminum.

10. Procedure according to claim 9, wherein the average coefficient of heat transfer from said aluminum to said mold within said mold is not more than about 0.05 calories/cm /secondPC; wherein said average coefficient of heat transfer from said ingot surface to the coolant fluid in said first zone is between about 0.1 and about 0.2 calories/cm /second/C; and wherein said coefficient of heat transfer from saidingot surface to the coolant fluid in said second zone isat least about 0.5 calories/cm /second/C.

11. Procedure for continuously casting an ingot, including the steps of a. supplying molten metal to the inlet end of a casting mold having an open outlet end, while cooling the mold for solidifying the peripheral portion of the metal therein to form an ingot having an initially molten core, the mold-cooling. step providing an average coefficient of heat transfer from the metal to the mold sufficient to produce at the mold outlet end a thin solid ingot shell having a thickness adequate to withstand frictional stresses between the mold and the ingot, and while 0. continuously advancing said ingot through and beyond said outlet end of said mold, and while d. cooling said ingot beyond said mold for progressively solidifying said molten core to produce a completely solidified ingot, said molten core extending as a molten metal sump within said ingot to an extremity of thesump beyond said mold outlet end along the path of advance of said ingot; wherein the improvement comprises:

e. the ingot-cooling step comprising i. in a first cooling zone extending from said mold outlet end for a predetermined distance along the path of advance of said ingot, directing a first supply of coolant fluid onto the surface of said ingot for cooling said ingot, in a manner providing an average coefficient of heat transfer from said ingot to the coolant fluid in said first zone over said predetermined distance equal to between about one and about six times the value of said average coefficient of heat transfer in said mold, said average coefficient of heat transfer in said first zone being sufficient to maintain said ingot shell in solid state and to solidify the major portion of the ingot cross section; and

. in a second cooling zone, located at said predetermined distance from said mold outlet end, separately directing a second supply of coolant fluid onto the ingot surface for cooling said ingot, in a manner providing a coefficient of heat transfer from the ingot surface to the coo lant fluid in said second zone equal to at least about one and one half times the value of said average coefiicient of heat transfer in said first zone, said predetermined distance being of such value that said second supply of coolant fluid impinges on the ingot surface at a locality adjacent the extremity of said molten metal sump within said ingot beyond said mold outlet end along the path of ingot advance, and said second zone extending along the path of advance of said ingot for a distance sufficient to effect complete solidification of said ingot.

12. Procedure according to claim 11, wherein said average coefficient of heat transfer in said first zone is equal to at least about two times the value of the average coefficient of heat transfer in the mold.

13.. Procedure according to claim 12, wherein said coefficient of heat transfer in said second zone is equal to at least about five times the value of the average coefficient of heat transfer in said first zone.

14. In apparatus for continuously casting an ingot, in combination,

a. an annular mold adapted to receive and contain a supply of molten metal for casting into an ingot, said mold having an open outlet end arranged for advance of the ingot through and beyond said mold outlet end along a defined path as the peripheral portion of said ingot solidifies within said mold;

. a cooling jacket laterally surrounding the mold for receiving and conducting a continuous flow of coolant liquid for cooling the mold;

. an annular chamber surrounding the outlet end of the mold, said chamber having plural apertures communicating with the jacket for admission of liquid from the jacket to the chamber and said chamber further including passage-defining means for discharging fluid from the chamber toward the surface of said ingot beyond the mold outlet end;

d. means independent of said chamber for discharging liquid from said cooling jacket away from contact with said ingot; and

. means for controllably supplying gas to said chamber for opposing introduction of liquid to said chamber through said apertures and for mixture with liquid within said chamber such that the fluid discharged from said chamber through said passage-defining means toward said ingot comprises a mixture of gas and liquid. I

15. In apparatus for continuously casting an ingot, in

combination,

. means defining a passage for conducting a stream of the liquid within thev cooling jacket in a direction substantially parallel to the direction of advance of said ingot, said passage having an outlet end opening in the direction of advance of said distance equal to not more than about one fourth the ingot adjacent the mold outlet end, the outlet end minimum transverse dimension of said ingot.

of said passage being defined by diverging surfaces 17. Procedure according to claim 16, wherein said respectively sloping toward and away from said insecond supply of coolant fluid impinges on the ingot got, and said passage being further arranged s surface ahead of the sump extremity in the path of adthat liquid of said stream follows and is directed by Vance of Said g one of said divergent surfaces as the stream is A meihod according to claim 17, wherein said discharged through the passage outlet end;and second supply of coolant fluid impinges on the ingot d. means for controllably applying to the stream, ad- Surface at a locality spaced from the p extremiiy jacent the passage outlet end and in a direction along the P Ofingot advance by a distance equal to transverse to the path f fl f the Stream, a fl not more than about one sixth the minimum transverse of fluid effective to deflect liquid of said stream dimension of Said ingotfrom said one diverging surface to the other of said procedul'e according to claim 11, wherein the diverging surfaces. distance between said mold outlet end and said sump 16. Procedure according to claim 11, wherein said extremity along P of f? of Said ingot, is second supply of coolant fluid impinges against the surf f f than mm'mum transverse face of said ingot at a locality spaced from the sump exdlmens'on of sald mgot' tremity along the path of advance of said ingot by a

Claims

1. Procedure for continuously casting an ingot, including the steps of a. supplying molten metal to the inlet end of a casting mold having an open outlet end, while b. cooling the mold for solidifying the peripheral portion of the metal therein to form an ingot having an externally solid shell and an initially molten core, and while c. continuously advancing said ingot through and beyond said outlet end of said mold, and while d. cooling said ingot beyond said mold for progressively solidifying said molten core to produce a completely solidified ingot, said molten core extending as a molten metal sump within said ingot to an extremity of the sump beyond said mold outlet end along the path of advance of said ingot; wherein the improvement comprises: e. the ingot-cooling step comprising i. in a first cooling zone extending from said mold outlet end for a predetermined distance along the path of advance of said ingot, directing a first supply of coolant fluid onto the surface of said ingot for cooling said ingot, in a manner providing an average coefficient of heat transfer from said ingot to the coolant fluid in said first zone having a value effective to maintain said ingot shell in solid state while maintaining the core of said ingot in molten state substantially throughout said first zone; and ii. in a second cooling zone, located at said predetermined distance from said mold outlet end, separately directing a second supply of coolant fluid onto the ingot surface for cooling said ingot, in a manner providing a coefficient of heat transfer from the ingot surface to the coolant fluid in said second zone substantially greater than said average coefficient of heat transfer in said first zone, and effective to produce complete solidification of said ingot, thereby to provide a substantial increase in heat transfer coefficient as aforesaid at about said predetermined distance from said mold outlet end; and f. the ingot-advancing step comprising advancing said ingot at a rate such that said molten sump terminates, by solidification of the center of said ingot, at a locality adjacent said predetermined dIstance beyond said mold outlet end, for maintaining the extremity of said sump at a locality adjacent the locality of substantial increase in heat transfer coefficient as aforesaid.

8. Procedure according to claim 1, wherein the step of directing a first supply of coolant fluid onto the surface of said ingot comprises establishing and maintaining a body of liquid in surrounding relation to said ingot at least substantially throughout the extent of said first zone, said ingot advancing through said body of liquid in said first zone, and heat from said ingot vaporizing liquid of said body adjacent the surface of said ingot, and wherein the step of directing a second supply of coolant fluid onto the ingot surface comprises directing a stream of said liquid against the ingot surface at a substantial angle of impingement in said second zone.

9. Procedure according to claim 1, wherein said molten metal is aluminum.

10. Procedure according to claim 9, wherein the average coefficient of heat transfer from said aluminum to said mold within said mold is not more than about 0.05 calories/cm2/second/*C; wherein said average coefficient of heat transfer from said ingot surface to the coolant fluid in said first zone is between about 0.1 and about 0.2 calories/cm2/second/*C; and wherein said coefficient of heat transfer from said ingot surface to the coolant fluid in said second zone is at least about 0.5 calories/cm2/second/*C.

11. Procedure for continuously casting an ingot, including the steps of a. supplying molten metal to the inlet end of a casting mold having an open outlet end, while b. cooling the mold for solidifying the peripheral portion of the metal therein to form an ingot having an initially molten core, the mold-cooling step providing an average coefficient of heat transfer from the metal to the mold sufficient to produce at the mold outlet end a thin solid ingot shell having a thickness adequate to withstand frictional stresses between the mold and the ingot, and while c. continuously advancing said ingot through And beyond said outlet end of said mold, and while d. cooling said ingot beyond said mold for progressively solidifying said molten core to produce a completely solidified ingot, said molten core extending as a molten metal sump within said ingot to an extremity of the sump beyond said mold outlet end along the path of advance of said ingot; wherein the improvement comprises: e. the ingot-cooling step comprising i. in a first cooling zone extending from said mold outlet end for a predetermined distance along the path of advance of said ingot, directing a first supply of coolant fluid onto the surface of said ingot for cooling said ingot, in a manner providing an average coefficient of heat transfer from said ingot to the coolant fluid in said first zone over said predetermined distance equal to between about one and about six times the value of said average coefficient of heat transfer in said mold, said average coefficient of heat transfer in said first zone being sufficient to maintain said ingot shell in solid state and to solidify the major portion of the ingot cross section; and ii. in a second cooling zone, located at said predetermined distance from said mold outlet end, separately directing a second supply of coolant fluid onto the ingot surface for cooling said ingot, in a manner providing a coefficient of heat transfer from the ingot surface to the coolant fluid in said second zone equal to at least about one and one half times the value of said average coefficient of heat transfer in said first zone, said predetermined distance being of such value that said second supply of coolant fluid impinges on the ingot surface at a locality adjacent the extremity of said molten metal sump within said ingot beyond said mold outlet end along the path of ingot advance, and said second zone extending along the path of advance of said ingot for a distance sufficient to effect complete solidification of said ingot.

13. Procedure according to claim 12, wherein said coefficient of heat transfer in said second zone is equal to at least about five times the value of the average coefficient of heat transfer in said first zone.

14. In apparatus for continuously casting an ingot, in combination, a. an annular mold adapted to receive and contain a supply of molten metal for casting into an ingot, said mold having an open outlet end arranged for advance of the ingot through and beyond said mold outlet end along a defined path as the peripheral portion of said ingot solidifies within said mold; b. a cooling jacket laterally surrounding the mold for receiving and conducting a continuous flow of coolant liquid for cooling the mold; c. an annular chamber surrounding the outlet end of the mold, said chamber having plural apertures communicating with the jacket for admission of liquid from the jacket to the chamber and said chamber further including passage-defining means for discharging fluid from the chamber toward the surface of said ingot beyond the mold outlet end; d. means independent of said chamber for discharging liquid from said cooling jacket away from contact with said ingot; and e. means for controllably supplying gas to said chamber for opposing introduction of liquid to said chamber through said apertures and for mixture with liquid within said chamber such that the fluid discharged from said chamber through said passage-defining means toward said ingot comprises a mixture of gas and liquid.

15. In apparatus for continuously casting an ingot, in combination, a. an annular mold adapted to receive and contain a supply of molten metal for casting into an ingot, said mold having an open outlet end arranged for advance of the ingot through and beyond said mold outlet end along a defined path as the peripheral porTion of said ingot solidifies within said mold; b. a cooling jacket laterally surrounding the mold for receiving and conducting a continuous flow of coolant liquid for cooling the mold; c. means defining a passage for conducting a stream of the liquid within the cooling jacket in a direction substantially parallel to the direction of advance of said ingot, said passage having an outlet end opening in the direction of advance of said ingot adjacent the mold outlet end, the outlet end of said passage being defined by diverging surfaces respectively sloping toward and away from said ingot, and said passage being further arranged so that liquid of said stream follows and is directed by one of said divergent surfaces as the stream is discharged through the passage outlet end; and d. means for controllably applying to the stream, adjacent the passage outlet end and in a direction transverse to the path of flow of the stream, a flow of fluid effective to deflect liquid of said stream from said one diverging surface to the other of said diverging surfaces.

16. Procedure according to claim 11, wherein said second supply of coolant fluid impinges against the surface of said ingot at a locality spaced from the sump extremity along the path of advance of said ingot by a distance equal to not more than about one fourth the minimum transverse dimension of said ingot.

17. Procedure according to claim 16, wherein said second supply of coolant fluid impinges on the ingot surface ahead of the sump extremity in the path of advance of said ingot.

18. A method according to claim 17, wherein said second supply of coolant fluid impinges on the ingot surface at a locality spaced from the sump extremity along the path of ingot advance by a distance equal to not more than about one sixth the minimum transverse dimension of said ingot.