JP2015160213A - Continuous casting method of slab made of titanium or titanium alloy - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a continuous casting method of a slab made of titanium or titanium alloy, enabling a casting surface of a cast slab to be smoothened even when a slab of any size is cast.SOLUTION: The number, an output and a movement pattern of plasma torches 7 are determined in such a way that: a torch movement period being a time required for moving a plasma torch 7 once in a prescribed movement pattern on the surface of molten metal 12 almost accords with a torch movement period of operation conditions obtained beforehand; and in each of a plurality of portions made by dividing an initial solidification part being a part which is solidified first when the molten metal 12 comes in contact with a casting mold 2 into a plurality of portions in a circumferential direction of the casting mold 2, a slab average input heat quantity being a value obtained by multiplying a correction value by the average input heat quantity to the portion almost accords with the average input heat quantity of operation conditions obtained beforehand Further, an agitation force of electromagnetic agitation is determined in such a way that a molten metal flow time being a time required for causing the molten metal 12 to flow through the width of a torch heating area in a length side direction of the casting mold 2 by the magnetic agitation almost accords with a molten metal flow time of operation conditions obtained beforehand.

Description

  The present invention relates to a method for continuously casting a slab made of titanium or a titanium alloy, which continuously casts a slab made of titanium or a titanium alloy.

  Ingots are continuously cast by injecting a metal melted by vacuum arc melting or electron beam melting into a bottomless mold and solidifying it to draw downward.

  Patent Document 1 discloses an automatic control plasma melting casting method in which titanium or a titanium alloy is melted by plasma arc melting in an argon gas atmosphere and injected into a mold to be solidified. In plasma arc melting performed in an inert gas atmosphere, unlike electron beam melting performed in a vacuum, not only pure titanium but also a titanium alloy can be cast.

Japanese Patent No. 3077387

  By the way, if there are irregularities or scratches on the cast surface of the cast ingot, pretreatment such as cutting the surface before rolling is required, which causes a reduction in yield and an increase in work man-hours. Therefore, it is required to cast an ingot having no irregularities or scratches on the casting surface.

  Here, it is presumed that the heat input / extraction state to the initial solidification part, which is the part where the molten metal first touches the mold and solidifies, greatly affects the properties of the casting surface. Therefore, it is considered that an ingot having a good casting surface can be obtained by appropriately controlling the heat input / extraction state to the initial solidification portion in the vicinity of the molten metal surface.

  However, when a large slab of, for example, about 250 × 1500 mm is continuously cast by plasma arc melting, the heating range of the plasma torch is limited. Therefore, it is necessary to move the plasma torch in the horizontal direction along the mold having a rectangular cross section and perform the drawing while suppressing the growth of the initial solidified portion in the vicinity of the mold.

  However, since the residence time of the plasma torch is long on the long side of the mold, the heat input to the initial solidification portion is large and the solidification shell becomes thin. On the other hand, since the residence time of the plasma torch is short on the short side or corner of the mold, the heat input to the initial solidification part is insufficient and the solidified shell grows (thickens). Thereby, the solidification behavior becomes non-uniform depending on the position of the slab, which leads to deterioration of the casting surface properties.

  Therefore, a flow swirling in the horizontal direction (swirl flow) is generated on at least the molten metal surface by electromagnetic stirring. By this swirl flow, hot molten metal staying at the long side of the mold is transferred to the short side and corner of the mold where the solidified shell is likely to grow. As a result, the temperature rise on the long side of the mold having a long plasma torch residence time and the temperature drop on the short side or corner of the mold having a short plasma torch residence time are alleviated.

  However, the number, output, and movement pattern of the plasma torches necessary for smoothing the casting surface vary depending on the size of the slab to be cast. Moreover, the stirring force of electromagnetic stirring required in order to smooth a casting surface changes with sizes of the slab to cast. Therefore, operation conditions for smoothing the casting surface, such as a heat input pattern and an electromagnetic stirring pattern, must be evaluated for each size of the slab to be cast, which is complicated.

  The objective of this invention is providing the continuous casting method of the slab which consists of titanium or a titanium alloy which can smooth a casting surface, no matter what size slab is cast.

  The present invention continuously casts a slab made of titanium or a titanium alloy by pouring a molten metal in which titanium or a titanium alloy is melted into a bottomless mold having a rectangular cross section and solidifying the molten metal. A casting method for casting a slab of a specific size in advance, obtaining an operating condition capable of smoothing the casting surface of the slab, and providing a plasma torch for heating the molten metal surface in the mold The torch movement period, which is the time required to move once in a predetermined movement pattern on the molten metal surface, substantially coincides with the torch movement period of the operating conditions, and the molten metal first solidifies by touching the mold. In each of a plurality of parts obtained by dividing the initial solidification part as a part into a plurality of circumferential directions of the mold, an average heat input amount obtained by averaging the heat input to the part in the length direction along the mold of the part The slab average heat input, which is a value obtained by multiplying a correction value based on the length of the mold surrounding the torch heating area, which is the area heated by each plasma torch, of the molten metal surface is the average input of the operating conditions. The number of the plasma torches, the output, and the movement pattern are determined so as to substantially match the amount of heat, and the long side of the mold is generated by electromagnetic stirring that causes a flow swirling in the horizontal direction on at least the molten metal surface. Determining the stirring force of the electromagnetic stirring so that the molten metal advancing time, which is the time required for the molten metal to advect the width of the torch heating region in the direction, substantially matches the molten metal advancing time of the operating conditions. Features.

  According to the present invention, the plasma is such that the torch moving period substantially matches the torch moving period of the operating condition obtained in advance, and the slab average heat input substantially matches the average heat input of the operating condition obtained in advance. Determine the number of torches, output, and movement pattern. Here, compared to the heat input pattern obtained in advance under the operating conditions, the heat input pattern in which the temporal change in the amount of heat input to the molten metal surface due to the movement of the plasma torch and the non-uniformity due to spatial fluctuations are almost the same is obtained. Presumed to be a thermal pattern. Therefore, in order to make the non-uniformity of the heat input amount of the molten metal to the molten metal surface due to the movement of the plasma torch almost coincide with the torch movement period of the operating condition obtained in advance. Further, a heat input pattern in which the average heat input amount in each of a plurality of parts obtained by dividing the initial solidification portion into a plurality of parts is a heat input pattern to be obtained, with respect to the heat input pattern of the operation conditions obtained in advance. Presumed. Therefore, the average heat input at each part is made to substantially coincide with the average heat input at each part under the operating conditions obtained in advance. Here, the longer the length of the mold surrounding the torch heating region, the greater the cooling capacity of the mold. Therefore, the slab average heat input obtained by correcting the average heat input with the correction value is made to substantially coincide with the average heat input of the operation conditions obtained in advance. Thus, by determining the number of plasma torches, the output, and the movement pattern, a heat input pattern for smoothing the casting surface can be obtained.

  Further, according to the present invention, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially matches the molten metal advancing time obtained in advance in the operating conditions. Here, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the operation condition obtained in advance is the electromagnetic stirring pattern to be obtained. Therefore, the molten metal advancing time is made to substantially coincide with the molten metal advancing time of the operating conditions obtained in advance. Thus, the electromagnetic stirring pattern for smoothing a casting surface can be obtained by determining the stirring force of electromagnetic stirring.

  Then, by casting the slab using the obtained heat input pattern and electromagnetic stirring pattern, the average surface temperature of the slab in the complete contact region where the mold and the slab are in contact with each other, a good casting surface can be obtained. It can be within a predetermined range. Thereby, even if it casts what size slab, a casting surface can be smoothed.

  Note that “substantially coincident” includes complete coincidence and means that the operation condition obtained in advance and the operation condition to be obtained coincide with each other to such an extent that the torch movement period and the like can be regarded as substantially equal.

It is a perspective view which shows a continuous casting apparatus. It is sectional drawing which shows a continuous casting apparatus. It is explanatory drawing showing the generation | occurrence | production mechanism of a surface defect. It is explanatory drawing showing the generation | occurrence | production mechanism of a surface defect. It is a surface photograph of a slab. It is a surface photograph of a slab. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is a model figure of the perfect contact area | region of a casting_mold | template and a slab. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is a figure showing the time change of slab surface temperature. It is the model figure which looked at the casting_mold | template from the upper direction. It is a figure which shows the average heat input in each site | part which divided | segmented the initial stage solidification part into several site | parts. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the histogram of a heat balance parameter | index. It is a figure which shows the cumulative distribution of a heat balance parameter | index. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the histogram of a heat balance parameter | index. It is a figure which shows the cumulative distribution of a heat balance parameter | index. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is the model figure which looked at the casting_mold | template 2 from upper direction. It is a figure which shows the relationship between a molten metal advancing time and an uneven | corrugated occurrence frequency index.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

(Construction of continuous casting equipment)
The continuous casting method of a slab made of titanium or a titanium alloy according to the present embodiment is a method of injecting a plasma arc melted titanium or titanium alloy into a bottomless mold having a rectangular cross section and drawing it downward while solidifying. This is a continuous casting method for continuously casting a slab made of titanium or a titanium alloy. A slab continuous casting apparatus (continuous casting apparatus) 1 made of titanium or a titanium alloy for carrying out this continuous casting method, as shown in FIG. 1 which is a perspective view and FIG. 2 which is a sectional view, A cold hearth 3, a raw material charging device 4, a plasma torch 5, a starting block 6, and a plasma torch 7 are provided. The continuous casting apparatus 1 is surrounded by an inert gas atmosphere made of argon gas, helium gas, or the like.

  The raw material input device 4 inputs the raw material of titanium or titanium alloy such as sponge titanium and scrap into the cold hearth 3. The plasma torch 5 is provided above the cold hearth 3 and generates a plasma arc to melt the raw material in the cold hearth 3. The cold hearth 3 injects the molten metal 12 in which the raw material is melted into the mold 2 from the pouring part 3a at a predetermined flow rate.

  The casting mold 2 is made of copper, has a bottom and has a rectangular cross-sectional shape, and is cooled by water circulating through at least a part of the rectangular tube-shaped wall portion. The starting block 6 is moved up and down by a drive unit (not shown) and can close the lower opening of the mold 2. The plasma torch 7 is provided above the molten metal 12 in the mold 2. The plasma torch 7 plasmas the molten metal 12 injected into the mold 2 while being horizontally moved on the molten metal 12 by moving means (not shown). Heat with an arc.

  In the above configuration, the molten metal 12 injected into the mold 2 is solidified from the contact surface with the water-cooled mold 2. Then, the starting block 6 that has closed the lower opening of the mold 2 is pulled downward at a predetermined speed, so that the prismatic slab 11 with the molten metal 12 solidified is continuously cast while being drawn downward. Is done.

  Here, in the electron beam melting in a vacuum atmosphere, since a minute component evaporates, it is difficult to cast a titanium alloy. On the other hand, in plasma arc melting in an inert gas atmosphere, not only pure titanium but also a titanium alloy can be cast.

  The continuous casting apparatus 1 may have a flux feeding device that feeds a solid phase or liquid phase flux to the molten metal surface of the molten metal 12 in the mold 2. Here, in the electron beam melting in a vacuum atmosphere, since the flux is scattered, it is difficult to put the flux into the molten metal 12 in the mold 2. In contrast, plasma arc melting in an inert gas atmosphere has the advantage that the flux can be charged into the molten metal 12 in the mold 2.

(Operating conditions)
By the way, when the slab 11 made of titanium or a titanium alloy is continuously cast, if there are irregularities or scratches on the surface (casting surface) of the slab 11, a surface defect occurs in the next rolling process. Therefore, it is necessary to remove irregularities and scratches on the surface of the slab 11 by cutting or the like before rolling, which causes a cost increase such as a decrease in yield and an increase in work processes. Therefore, it is required to cast the slab 11 having no irregularities or scratches on the casting surface.

  Here, as shown in FIGS. 3A and 3B, in continuous casting of the slab 11 made of titanium, the vicinity of the molten metal surface of the molten metal 12 heated by a plasma arc or an electron beam (from the molten metal surface to about 10 mm below the molten metal surface). The mold 2 and the surface of the slab 11 (solidified shell 13) are in contact only in the region. In the deeper region, the slab 11 is thermally contracted to generate an air gap 14 with the mold 2. Then, as shown in FIG. 3A, when the heat input to the initial solidification portion 15 (the portion where the molten metal 12 touches the mold 2 and solidifies first) is excessive, the solidified shell 13 is too thin and the solidified shell due to insufficient strength. A “fracture defect” occurs in which the surface of 13 is torn off. On the other hand, as shown in FIG. 3B, when the heat input to the initial solidified portion 15 is insufficient, the molten metal 12 is covered on the grown (thickened) solidified shell 13 to generate a “hot water covering defect”. FIG. 4A shows a surface photograph of the slab 11 in which the “scratch defect” has occurred, and FIG. 4B shows a photograph of the surface of the slab 11 in which the “water bath defect” has occurred.

  Therefore, it is estimated that the heat input / extraction state to the initial solidified portion 15 in the vicinity of the molten metal surface of the molten metal 12 greatly affects the properties of the casting surface. Therefore, it is considered that a slab 11 having a good casting surface can be obtained by appropriately controlling the heat input / extraction state to the initial solidification portion 15 in the vicinity of the molten metal 12.

  Here, as shown in FIG. 5A, which is a model view of the mold 2 as viewed from above, when the slab 11 having a large size of, for example, 250 × 1500 mm is continuously cast by plasma arc melting, the heating range of the plasma torch 7 There is a limit. Therefore, it is necessary to heat the entire molten metal surface using a plurality of plasma torches 7 having a large output. In FIG. 5A, two plasma torches 7 having a large output are used. Further, since the slab 11 is thick, it is necessary to turn the plasma torch 7 along the mold 2 in order to suppress the growth of the solidified shell 13 on the short side or corner of the mold 2. Here, the arrow shown in FIG. 5A indicates the path along which the plasma torch 7 moves.

  However, since the residence time of the plasma torch 7 is long on the long side of the mold 2, the heat input to the initial solidification part 15 is large and the solidification shell 13 becomes thin. On the other hand, since the residence time of the plasma torch 7 is short on the short side or corner portion of the mold 2, heat input to the initial solidification portion 15 is insufficient and the solidified shell 13 grows (thickens). Thereby, the solidification behavior becomes non-uniform depending on the position of the slab 11, which leads to deterioration of cast surface properties.

  Therefore, as shown in FIG. 5B, which is a model diagram of the mold 2 as viewed from above, at least the molten metal surface of the molten metal 12 in the mold 2 is electromagnetically induced by an electromagnetic stirrer (EMS) (not shown) disposed on the side of the mold 2. To stir. The EMS is obtained by winding an EMS coil around a coil iron core. By electromagnetic stirring by EMS, a flow swirling in the horizontal direction (swirl flow) is generated on the molten metal surface or in the vicinity of the molten metal surface. By this swirl flow, the hot molten metal 12 staying in the long side portion of the mold 2 is transferred to the short side portion and the corner portion of the mold 2 where the solidified shell 13 is likely to grow. As a result, the temperature rise on the long side of the mold 2 with a long residence time of the plasma torch 7 and the temperature drop on the short side or corner portion of the mold 2 with a short residence time of the plasma torch 7 are alleviated.

  In addition, the direction of the swirl flow at least on the molten metal surface of the molten metal 12 may coincide with the swirl direction of the plasma torch 7 or may be in the opposite direction. However, the fluctuation range of the surface temperature of the slab 11 can be reduced by turning at least the molten metal surface of the molten metal 12 in the direction opposite to the turning direction of the plasma torch 7.

  Here, when continuously casting a large slab 11, it is necessary to increase the flow rate of the molten metal 12 with a large stirring force in order to transfer heat to the entire molten metal surface by electromagnetic stirring.

  On the other hand, as shown in FIG. 6A, which is a model view of the mold 2 as viewed from above, when the slab 11 having a small size of, for example, 125 × 375 mm is continuously cast by plasma arc melting, the area of the molten metal surface is small. It is possible to heat the entire molten metal surface with one plasma torch 7 having a small output. Moreover, since the thickness of the slab 11 is thin, it is possible to suppress the growth of the solidified shell 13 on the short side or corner of the mold 2 by reciprocating the plasma torch 7 on the same line. Here, the arrow shown in FIG. 6A indicates the path along which the plasma torch 7 moves.

  Further, as shown in FIG. 6B, which is a model diagram when the mold 2 is viewed from above, when continuously casting a small slab 11, even if the stirring force by electromagnetic stirring is small and the flow rate of the molten metal 12 is slow, It is possible to transfer heat to the entire surface of the hot water.

  Thus, the number, output, and movement pattern of the plasma torches 7 necessary for smoothing the casting surface vary depending on the size of the slab 11 to be cast. Moreover, the stirring force of electromagnetic stirring required in order to smooth a casting surface changes with sizes of the slab 11 to cast. Therefore, operation conditions for smoothing the casting surface such as a heat input pattern and an electromagnetic stirring pattern must be evaluated for each size of the slab 11 to be cast, which is complicated.

  Here, a model diagram of the complete contact region 16 between the mold 2 and the slab 11 is shown in FIG. The complete contact region 16 is a region where the mold 2 and the slab 11 are in contact with each other, which is illustrated by hatching from the molten metal surface to about 10 mm below the molten metal surface. In the complete contact region 16, a passing heat flux q from the surface of the slab 11 to the mold 2 is generated. D is the thickness of the solidified shell 13.

The relationship between the surface temperature T _S of the passing heat flux q and the slab 11 shown in FIG. Here, the passing heat flux q [W / m ² ] and the surface temperature T _S [° C.] of the slab 11 which are heat balance indexes are evaluated by average values in the complete contact region 16. From this relationship diagram, if the average value of the surface temperature T _S of the slab 11 in the complete contact region 16 between the mold 2 and the slab 11 is in the range of 800 ° C. <T _S <1250 ° C., there is no tearing defect or bathing defect. It can be seen that a slab 11 having a good casting surface can be obtained.

Therefore, in order to obtain the slab 11 having a good casting surface state, the operating state in which the casting surface can be smoothed, while the plasma torch 7 moves once in a predetermined movement pattern, in the complete contact region 16. the average surface temperature of the slab 11 may be clamped to the range 800 ℃ _<T S _<of 1250 ° C..

  Therefore, in the present embodiment, a slab 11 having a specific size is cast in advance, and an operation condition capable of smoothing the casting surface of the slab 11 is obtained. And in calculating | requiring the operating condition which can smooth a casting surface in the slab 11 in which this size differs from the slab 11 of this size, the torch moving period and the slab average heat input are the torch moving period and the average of the operating condition obtained in advance. The number, output, and movement pattern of the plasma torches 7 are determined so as to substantially match the heat input.

  In the present embodiment, “substantially coincident” means that the torch movement period and the like are substantially equal to each other, including the complete coincidence, and the operation condition obtained in advance and the operation condition to be obtained. means.

  Here, the torch moving period is a time required to move the plasma torch 7 for heating the molten metal surface of the molten metal 12 in the mold 2 once in a predetermined movement pattern on the molten metal surface. Specifically, the torch moving period is obtained by dividing the moving distance of one plasma torch 7 by the moving speed of the plasma torch 7.

  As shown in FIG. 5A, when casting a large slab 11, the plasma torch 7 is swung on the molten metal surface at a predetermined speed. The time required to make one round of the plasma torch 7 is the torch moving period. As shown in FIG. 6A, when casting a small slab 11, the plasma torch 7 is reciprocated on the molten metal surface at a predetermined speed. The time required to make the plasma torch 7 reciprocate once is the torch moving period.

  Here, when it sees in the same location of the hot_water | molten_metal surface of the molten metal 12, the amount of heat input changes with progress of time because the moving plasma torch 7 approaches or leaves | separates. Further, when viewed from the entire molten metal surface of the molten metal 12, the location near the plasma torch 7 where the amount of heat input is large and the location far from the plasma torch 7 where the amount of heat input is small change as the plasma torch 7 moves. As described above, the movement of the plasma torch 7 causes nonuniformity due to temporal changes and spatial fluctuations in the amount of heat input to the molten metal surface of the molten metal 12.

  Therefore, a heat input pattern in which the temporal change in the amount of heat input to the surface of the molten metal 12 due to the movement of the plasma torch 7 and the non-uniformity due to the spatial variation are substantially equal to the heat input pattern of the operation conditions obtained in advance is obtained. Presumed to be a heat input pattern. Therefore, in order to make the non-uniformity of the heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 substantially coincide with the torch movement period of the operating conditions obtained in advance.

  The average heat input is a plurality of initial solidification portions 15 (portions where the molten metal 12 first solidifies when touching the mold 2) (see FIGS. 3A and 3B) divided into a plurality of portions in the circumferential direction of the mold 2. In each part, the amount of heat input to a certain part is averaged in the length direction along the mold 2 of the part. The slab average heat input is a value obtained by multiplying the average heat input by a correction value. The correction value is a value based on the length that the mold 2 surrounds the torch heating region. The torch heating region is a region where each plasma torch 7 heats the molten metal 12.

  In the present embodiment, as shown in FIG. 9A, which is a model view of the mold 2 as viewed from above, the initial solidified portion 15 is formed along the inner periphery of the mold 2 with corners (1) to (4), long side 1 / 4 (1), (2), long side 1/2 (1), (2), long side 3/4 (1), (2), short side (1), (2), 12 parts in total Dividing into 15a, the average heat input at each part 15a is obtained. Further, as shown in FIG. 9B, which is a model diagram of the mold 2 as viewed from above, representative points are set in each of the twelve portions 15a. That is, in the corners (1) to (4) and the short sides (1) and (2), representative points are respectively set at the center positions in the length direction along the wall surface of the mold 2. Further, in the long side 1/4 (1), (2), the long side 1/2 (1), (2), and the long side 3/4 (1), (2), from the left to the right in the figure. Thus, representative points are set at a position of 1/4 of the long side of the mold 2, a position of 1/2 of the long side of the mold 2, and a position of 3/4 of the long side of the mold 2, respectively.

  As shown in FIG. 10A, which is a model view of the mold 2 as viewed from above, when casting a large slab 11 using two plasma torches 7, half of the molten metal surface of the molten metal 12 is each plasma. This is a torch heating region 17 where the torch 7 heats. On the other hand, when a small slab 11 is cast using one plasma torch 7 as shown in FIG. 10B which is a model view of the mold 2 as viewed from above, the entire molten metal surface of the molten metal 12 is plasma torch. Reference numeral 7 denotes a torch heating region 17 for heating.

  As described above, the growth of the solidified shell 13 near the molten metal surface of the molten metal 12 is greatly affected by the heat input state to the initial solidified portion 15. As shown in FIG. 3A, when the heat input to the initial solidification portion 15 is excessive, a “breakage defect” occurs. On the other hand, as shown in FIG. 3B, when the heat input to the initial solidification portion 15 is insufficient, a “hot water bath defect” occurs.

  Therefore, a heat input pattern in which the heat input state to the initial solidification part 15 substantially matches the heat input pattern of the operating conditions obtained in advance, more specifically, a plurality of parts obtained by dividing the initial solidification part 15 into a plurality of parts. It is presumed that the heat input pattern in which the average heat input amounts in the respective parts 15a substantially coincide is the desired heat input pattern. Therefore, the average heat input at each part 15a is made to substantially coincide with the average heat input at each part under the operating conditions obtained in advance.

  However, the torch heating region 17 shown in FIG. 10A is surrounded on three sides by the mold 2, whereas the torch heating region 17 shown in FIG. And the cooling capability by the casting_mold | template 2 becomes large, so that the casting_mold | template 2 encloses the torch heating area | region 17 long. Therefore, when the cooling capacity of the mold 2 is different between the torch heating region 17 having the operation condition obtained in advance and the torch heating region 17 having the operation condition to be obtained, the average heat input amount of the operation condition to be obtained is determined as the torch heating region. 17 is corrected with a correction value based on the length surrounded by the mold 2. And the slab average heat input which correct | amended the average heat input with the correction value is made to correspond substantially with the average heat input of the operating conditions obtained beforehand.

  In this way, the plasma torch 7 so that the torch moving period substantially coincides with the torch moving period of the operation condition obtained in advance and the slab average heat input substantially coincides with the average heat input of the operation condition obtained in advance. Number, output, and movement pattern are determined. Thereby, the heat input pattern for smoothing a casting surface can be obtained.

  Moreover, in this embodiment, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time by electromagnetic stirring substantially corresponds to the molten metal advancing time of the operating conditions obtained in advance. Here, the molten metal advancing time is a time required for the molten metal 12 to move the width of the torch heating region 17 in the long side direction of the mold 2. Specifically, the molten metal advancing time is a value obtained by dividing the width of the torch heating region 17 (torch effective heating width) in the long side direction of the mold 2 by the flow velocity of the molten metal 12 in the vicinity of the mold 2.

  As shown in FIG. 10A, when casting a large slab 11, the torch heating region 17 of each plasma torch 7 is half the surface of the molten metal 12, so the torch effective heating width is 1 of the long side of the mold 2. The length is / 2. On the other hand, as shown in FIG. 10B, when casting a small slab 11, the torch heating region 17 of the plasma torch 7 is the entire surface of the molten metal 12, so the torch effective heating width is the long side of the mold 2. Full length.

  Here, as shown in FIG. 11, which is a model diagram when the mold 2 is viewed from above, when the plasma torch 7 moves to the left side in the figure of the molten metal surface, the molten metal 12 on the molten metal surface on the right side in the figure away from the plasma torch 7. Temperature drops. Therefore, as indicated by an arrow, the hot molten metal 12 on the left side of the molten metal surface is advected to the right side of the molten metal surface by electromagnetic stirring. Thereby, as shown in FIG. 12, which is a diagram showing the time change of the surface temperature of the slab 11, the temperature drop of the molten metal 12 is alleviated and the slab surface temperature is made uniform as compared with the case where there is no electromagnetic stirring.

  However, when the molten metal advancing time, which is the time required for the molten metal 12 to move through the torch effective heating width, is different, the degree of change in the surface temperature of the slab 11 with the passage of time varies. Specifically, the shorter the molten metal advancing time, the smaller the change in the surface temperature of the slab 11 becomes, and the surface temperature of the slab 11 becomes uniform. Therefore, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the operation condition obtained in advance is the electromagnetic stirring pattern to be obtained. Therefore, the molten metal advancing time is made to substantially coincide with the molten metal advancing time of the operating conditions obtained in advance.

  Thus, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially coincides with the molten metal advancing time of the operating conditions obtained in advance. Thereby, the electromagnetic stirring pattern for smoothing a casting surface can be obtained.

Then, through the cast slab 11 using the obtained heat input pattern and the electromagnetic stirring pattern, the average value of the surface temperature T _S of the slab 11 in a full contact area 16 of contact between the mold 2 and the slab 11, It can be contained in a predetermined range (800 ° C. <T _S <1250 ° C.) where a good casting surface can be obtained. Thereby, even if the slab 11 of any size is cast, the casting surface can be smoothed.

(Flow solidification calculation)
Next, operating conditions for obtaining a slab 11 having a good casting surface over the entire circumference were obtained by fluidized solidification calculation. Specifically, an operating condition capable of smoothing the casting surface of a large slab 11 of 250 × 1500 mm in advance is obtained, and the casting surface of a small slab 11 of 125 × 375 mm is smoothed by fluidized solidification calculation. A possible operating condition was obtained.

  Under the operating conditions obtained in advance, as shown in FIG. 5A, the number of plasma torches 7 is two, and the output of each plasma torch 7 is 750 kW. The movement pattern of each plasma torch 7 is a pattern that turns clockwise with the torch effective heating width set to 750 mm. Each plasma torch 7 is turned about 62.5 mm from the casting wall of the mold 2. The moving speed of the plasma torch 7 is constant at 50 mm / sec. Therefore, the torch moving period of the plasma torch 7 is 30 sec.

  On the other hand, under the operating conditions obtained by the flow solidification calculation, as shown in FIG. 6A, the number of plasma torches 7 is one and the output of the plasma torches 7 is 200 to 250 kW. The movement pattern of the plasma torch 7 is a pattern that reciprocates on the same line with the torch effective heating width set to 375 mm. Therefore, as shown in FIG. 13 which is a model diagram when the mold 2 is viewed from above, the moving speed of the central 210 mm is 30 mm / sec out of 290 mm, which is the one-way distance that the plasma torch 7 moves, on both sides thereof. By changing the transfer speed of 40 mm to 10 mm / sec, the torch moving period of the plasma torch 7 was made to coincide with the torch moving period (30 sec) of the operating conditions obtained in advance.

  Further, the average heat input at each part (see FIG. 9A) obtained by dividing the initial solidification portion 15 into 12 parts 15a was made to substantially coincide with the average heat input under the operating conditions obtained in advance. The average heat input in each part 15a is shown in FIG. Here, the output of the plasma torch 7 is 200 kW. In addition, Case (1) is the average heat input of each part 15a under the operating conditions obtained in advance, and Case (2) is each part 15a substantially matched with the average heat input of Case (1) under the required operating conditions. The average heat input.

  Here, as shown in FIG. 10A, the torch heating region 17 is surrounded on three sides by the mold 2 under the operation conditions obtained in advance. On the other hand, as shown in FIG. 10B, the torch heating region 17 is surrounded on four sides by the mold 2 under the operation conditions to be obtained. Thus, since the cooling condition by the casting mold 2 is larger under the operation condition to be obtained, the average heat input is corrected with the correction value. The correction value was calculated from the following equation (1) using the long side and the short side length (mm) of the mold 2 shown in FIG. 6A.

  (375 + 125 + 375 + 125) / (375 + 125 + 375) = 1.3 Formula (1)

When this correction value is applied to the output value of the plasma torch 7, the output is 250 kW. Further, the slab average heat input, which is a value obtained by correcting the average heat input with this correction value, is shown in FIG. 14 as Case (3). By setting the slab average heat input at each part 15a to 1.0 MW / m ² or more, the growth of the solidified shell 13 in the vicinity of the molten metal surface of the molten metal 12 can be suitably suppressed. Thereby, the slab 11 of a favorable casting surface can be obtained.

  15A to 15C show the relationship between the passing heat flux and the slab surface temperature when electromagnetic stirring is not performed. FIGS. 15A to 15C are plots obtained by extracting heat balance indices (passing heat flux and slab surface temperature) at 12 representative points shown in FIG. 9B while the plasma torch 7 moves once in a predetermined movement pattern. It is a thing. FIG. 15A shows the relationship between the passing heat flux and the slab surface temperature in Case (1). In contrast, FIG. 15B shows the relationship between the passing heat flux and the slab surface temperature in Case (2). The distribution range of the heat balance index in Case (2) is shifted to the low temperature side as a whole with respect to Case (1).

  FIG. 15C shows the relationship between the passing heat flux and the slab surface temperature in Case (3). The distribution range of the heat balance index in Case (3) is equivalent to that in Case (1), and it can be seen that the non-uniformity in the amount of heat input to the molten metal surface of the molten metal 12 is comparable.

  A histogram of the heat balance index is shown in FIG. 16A. Moreover, the cumulative distribution of the heat balance index is shown in FIG. 16B. Case (1) and Case (3) have substantially the same heat balance index distribution, and both have a heat input pattern that can be regarded as similar.

  As described above, the torch moving period substantially coincides with the torch moving period of the operation condition obtained in advance, and the slab average heat input amount substantially coincides with the average heat input of the operation condition obtained in advance at each portion 15a. Furthermore, it can be seen that a heat input pattern for smoothing the casting surface can be obtained by determining the number, output, and movement pattern of the plasma torches 7.

  Next, the stirring force of the electromagnetic stirring was determined so that the molten metal advancing time substantially coincided with the molten metal advancing time of the operating conditions obtained in advance, and electromagnetic stirring was applied according to the electromagnetic stirring pattern. The relationship between the passing heat flux and the slab surface temperature in Case (1) is shown in FIG. 17A, the relationship between the passing heat flux and the slab surface temperature in Case (2) is shown in FIG. 17B, and the passing heat flow in Case (3). FIG. 17C shows the relationship between the bundle and the slab surface temperature. By adding electromagnetic stirring, the variation of the heat balance index is relaxed, and a soaking effect is exhibited. In Case (1) and Case (3), the heat balance index is within the target range. On the other hand, in Case (2), an area below the target range exists. This is presumed to be caused by insufficient heat input due to the large cooling capacity of the mold 2.

  A histogram of the heat balance index is shown in FIG. 18A. Further, FIG. 18B shows the cumulative distribution of the heat balance index. In Case (1) and Case (3), the distribution of the heat balance index is almost the same.

  Thus, the electromagnetic stirring pattern for smoothing the casting surface can be obtained by determining the stirring force of the electromagnetic stirring so that the molten metal advancing time substantially coincides with the molten metal advancing time obtained in advance. I understand that. Then, it is understood that the distribution of the heat balance index is almost the same regardless of the slab size by making the heat input pattern substantially the same and then making the electromagnetic stirring pattern substantially the same.

Here, as shown in FIG. 19 which is a relationship diagram between the passing heat flux and the slab surface temperature, the deviation width of the surface temperature T _S of the slab 11 from the target temperature range (800 ° C. <T _S <1250 ° C.) is defined as a. _{Assuming j} , the unevenness occurrence frequency index D indicating the frequency of occurrence of unevenness on the surface of the slab 11 is expressed by the following equation (2).

D = Σa _j / N (2)
Here, N is the total number of data extraction points. When D is 0, the unevenness of the casting surface in all the locations of the slab 11 is below the target range, indicating that good casting surface properties can be obtained.

  Further, as shown in FIG. 20 which is a model diagram when the mold 2 is viewed from above, the flow velocity (absolute value) in the X direction at a position 10 mm away from the inner surface of the mold 2 is −2L / 5 ≦ x ≦ 2L /. The melt advection time was determined using the average value in the range of 5.

  The relationship between the molten metal advancing time and the unevenness occurrence frequency index D is shown in FIG. In this relationship diagram, the calculation results for each stirring force are plotted while changing the stirring force of electromagnetic stirring. Here, the stronger the stirring force of electromagnetic stirring, the faster the flow rate of the molten metal 12 and the shorter the molten metal advancing time. In addition, the smaller the unevenness occurrence frequency index D, the better the properties of the casting surface.

  From the calculation results of Case (1) and Case (3), if the heat input pattern is controlled so as to be similar to the heat input pattern of the operating conditions obtained in advance, the electromagnetic agitation can be performed according to the melt advection time without depending on the slab size. It can be seen that the relationship between conditions and casting surface properties can be organized. And it turns out that casting surface property improves by making stirring force increase so that molten metal advection time may be 3 sec or less. Thus, the behavior of casting surface improvement by electromagnetic stirring can be quantitatively evaluated by the molten metal advancing time.

  On the other hand, from the calculation result of Case (2), it can be seen that if the heat input pattern is not similar to the heat input pattern of the operation conditions obtained in advance, the casting surface properties are not improved even if the stirring force is increased.

(effect)
As described above, according to the continuous casting method of a slab made of titanium or a titanium alloy according to the present embodiment, the torch moving period substantially coincides with the torch moving period of the operation condition obtained in advance, and the slab average input The number, output, and movement pattern of the plasma torches 7 are determined so that the amount of heat substantially matches the average amount of heat input obtained under the operating conditions obtained in advance. Here, with respect to the heat input pattern of the operating conditions obtained in advance, the heat input pattern in which the temporal change in the amount of heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 and the non-uniformity due to the spatial variation substantially coincided, It is estimated that this is the desired heat input pattern. Therefore, in order to make the non-uniformity of the heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 substantially coincide with the torch movement period of the operating conditions obtained in advance. In addition, a heat input pattern in which the average heat input amount in each of the plurality of portions 15a obtained by dividing the initial solidification portion 15 into a plurality of parts is a heat input pattern to be obtained with respect to the heat input pattern of the operation conditions obtained in advance. Presumed to be. Therefore, the average heat input at each part 15a is made to substantially coincide with the average heat input at each part under the operating conditions obtained in advance. Here, the longer the length that the mold 2 surrounds the torch heating region 17, the greater the cooling capacity of the mold 2. Therefore, the slab average heat input obtained by correcting the average heat input with the correction value is made to substantially coincide with the average heat input of the operation conditions obtained in advance. Thus, the heat input pattern for smoothing a casting surface can be obtained by determining the number of plasma torches 7, an output, and a movement pattern.

  Further, according to the present embodiment, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially matches the molten metal advancing time obtained in advance in the operating conditions. Here, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the operation condition obtained in advance is the electromagnetic stirring pattern to be obtained. Therefore, the molten metal advancing time is made to substantially coincide with the molten metal advancing time of the operating conditions obtained in advance. Thus, the electromagnetic stirring pattern for smoothing a casting surface can be obtained by determining the stirring force of electromagnetic stirring.

Then, through the cast slab 11 using the obtained heat input pattern and the electromagnetic stirring pattern, the average value of the surface temperature T _S of the slab 11 in a full contact area 16 of contact between the mold 2 and the slab 11, It can be contained in a predetermined range (800 ° C. <T _S <1250 ° C.) where a good casting surface can be obtained. Thereby, even if the slab 11 of any size is cast, the casting surface can be smoothed.

Moreover, the growth of the solidified shell 13 in the vicinity of the molten metal surface of the molten metal 12 can be suitably suppressed by setting the slab average heat input amount to 1.0 MW / m ² or more. Thereby, the slab 11 of a favorable casting surface can be obtained.

  Moreover, casting surface property can be improved suitably by increasing the stirring force of electromagnetic stirring and making molten metal advection time 3 sec or less.

(Modification of this embodiment)
The embodiment of the present invention has been described above, but only specific examples are illustrated, and the present invention is not particularly limited, and the specific configuration and the like can be appropriately changed in design. Further, the actions and effects described in the embodiments of the invention only list the most preferable actions and effects resulting from the present invention, and the actions and effects according to the present invention are described in the embodiments of the present invention. It is not limited to what was done.

DESCRIPTION OF SYMBOLS 1 Continuous casting apparatus 2 Mold 3 Cold hearth 3a Pouring part 4 Raw material injection | throwing-in apparatus 5 Plasma torch 6 Starting block 7 Plasma torch 11 Slab 12 Molten metal 13 Solidified shell 14 Air gap 15 Initial solidification part 15a Part 16 Complete contact area 17 Torch heating region

  The present invention relates to a method for continuously casting a slab made of titanium or a titanium alloy, which continuously casts a slab made of titanium or a titanium alloy.

  Ingots are continuously cast by injecting a metal melted by vacuum arc melting or electron beam melting into a bottomless mold and solidifying it to draw downward.

  Patent Document 1 discloses an automatic control plasma melting casting method in which titanium or a titanium alloy is melted by plasma arc melting in an argon gas atmosphere and injected into a mold to be solidified. In plasma arc melting performed in an inert gas atmosphere, unlike electron beam melting performed in a vacuum, not only pure titanium but also a titanium alloy can be cast.

Japanese Patent No. 3077387

  By the way, if there are irregularities or scratches on the cast surface of the cast ingot, pretreatment such as cutting the surface before rolling is required, which causes a reduction in yield and an increase in work man-hours. Therefore, it is required to cast an ingot having no irregularities or scratches on the casting surface.

  Here, it is presumed that the heat input / extraction state to the initial solidification part, which is the part where the molten metal first touches the mold and solidifies, greatly affects the properties of the casting surface. Therefore, it is considered that an ingot having a good casting surface can be obtained by appropriately controlling the heat input / extraction state to the initial solidification portion in the vicinity of the molten metal surface.

  However, when a large slab of, for example, about 250 × 1500 mm is continuously cast by plasma arc melting, the heating range of the plasma torch is limited. Therefore, it is necessary to move the plasma torch in the horizontal direction along the mold having a rectangular cross section and perform the drawing while suppressing the growth of the initial solidified portion in the vicinity of the mold.

  However, since the residence time of the plasma torch is long on the long side of the mold, the heat input to the initial solidification portion is large and the solidification shell becomes thin. On the other hand, since the residence time of the plasma torch is short on the short side or corner of the mold, the heat input to the initial solidification part is insufficient and the solidified shell grows (thickens). Thereby, the solidification behavior becomes non-uniform depending on the position of the slab, which leads to deterioration of the casting surface properties.

  Therefore, a flow swirling in the horizontal direction (swirl flow) is generated on at least the molten metal surface by electromagnetic stirring. By this swirl flow, hot molten metal staying at the long side of the mold is transferred to the short side and corner of the mold where the solidified shell is likely to grow. As a result, the temperature rise on the long side of the mold having a long plasma torch residence time and the temperature drop on the short side or corner of the mold having a short plasma torch residence time are alleviated.

  However, the number, output, and movement pattern of the plasma torches necessary for smoothing the casting surface vary depending on the size of the slab to be cast. Moreover, the stirring force of electromagnetic stirring required in order to smooth a casting surface changes with sizes of the slab to cast. Therefore, operation conditions for smoothing the casting surface, such as a heat input pattern and an electromagnetic stirring pattern, must be evaluated for each size of the slab to be cast, which is complicated.

  The objective of this invention is providing the continuous casting method of the slab which consists of titanium or a titanium alloy which can smooth a casting surface, no matter what size slab is cast.

The present invention continuously casts a slab made of titanium or a titanium alloy by pouring a molten metal in which titanium or a titanium alloy is melted into a bottomless mold having a rectangular cross section and solidifying the molten metal. a casting method, the torch movement period is the time required to move once in a predetermined movement pattern plasma torch to heat the molten metal surface of the melt on the melt surface before Symbol in the mold it is, predetermined operations In each of a plurality of portions formed by dividing the initial solidification portion that is substantially the same as the torch moving period of the condition and is a portion where the molten metal first solidifies by touching the mold in the circumferential direction of the mold, A torch heating region, which is a region heated by each plasma torch on the molten metal surface, is obtained by averaging the heat input to the region in the length direction along the mold of the region. The number of plasma torches, the output, and the movement pattern are determined so that the slab average heat input, which is a value obtained by multiplying a correction value based on the surrounding length, substantially matches the average heat input of the operating conditions. The molten metal transfer time, which is the time required for the molten metal to move through the width of the torch heating region in the long side direction of the mold, by electromagnetic stirring that causes a flow swirling in the horizontal direction on at least the molten metal surface. The stirring force of the electromagnetic stirring is determined so as to substantially coincide with the molten metal advancing time of the operating conditions.

According to the present invention, the plasma torch is configured so that the torch moving period substantially matches the torch moving period of the predetermined operating condition, and the slab average heat input substantially matches the average heat input of the operating condition. The number, output, and movement pattern are determined. Here, compared to the heat input pattern under the above operating conditions, the heat input pattern in which the temporal change in the amount of heat input to the molten metal surface due to the movement of the plasma torch and the non-uniformity due to spatial fluctuations are almost the same is obtained. Presumed to be a pattern. Therefore, in order to substantially match the non-uniformity of the heat input to the molten metal surface due to the movement of the plasma torch, the torch movement period is made to substantially coincide with the torch movement period of the above operating conditions. In addition, with respect to the heat input pattern of the above operating conditions, it is estimated that the heat input pattern in which the average heat input amount in each of the plurality of parts obtained by dividing the initial solidification portion into a plurality of parts is the required heat input pattern. Is done. Therefore, the average heat input at each part is made to substantially coincide with the average heat input at each part under the above operating conditions. Here, the longer the length of the mold surrounding the torch heating region, the greater the cooling capacity of the mold. Therefore, the slab average heat input obtained by correcting the average heat input with the correction value is made to substantially coincide with the average heat input of the above operating conditions. Thus, by determining the number of plasma torches, the output, and the movement pattern, a heat input pattern for smoothing the casting surface can be obtained.

In addition, according to the present invention, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially matches the molten metal advancing time of the above operating conditions. Here, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the above operating conditions is the desired electromagnetic stirring pattern. Therefore, the melt advection time is made to substantially coincide with the melt advection time under the above operating conditions. Thus, the electromagnetic stirring pattern for smoothing a casting surface can be obtained by determining the stirring force of electromagnetic stirring.

  Then, by casting the slab using the obtained heat input pattern and electromagnetic stirring pattern, the average surface temperature of the slab in the complete contact region where the mold and the slab are in contact with each other, a good casting surface can be obtained. It can be within a predetermined range. Thereby, even if it casts what size slab, a casting surface can be smoothed.

Note that “substantially coincident” includes complete coincidence, and means that the above-described operation condition and the required operation condition coincide to such an extent that the torch movement period and the like can be regarded as substantially equal.

It is a perspective view which shows a continuous casting apparatus. It is sectional drawing which shows a continuous casting apparatus. It is explanatory drawing showing the generation | occurrence | production mechanism of a surface defect. It is explanatory drawing showing the generation | occurrence | production mechanism of a surface defect. It is a surface photograph of a slab. It is a surface photograph of a slab. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is a model figure of the perfect contact area | region of a casting_mold | template and a slab. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is the model figure which looked at the casting_mold | template from the upper direction. It is a figure showing the time change of slab surface temperature. It is the model figure which looked at the casting_mold | template from the upper direction. It is a figure which shows the average heat input in each site | part which divided | segmented the initial stage solidification part into several site | parts. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the histogram of a heat balance parameter | index. It is a figure which shows the cumulative distribution of a heat balance parameter | index. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is a figure which shows the histogram of a heat balance parameter | index. It is a figure which shows the cumulative distribution of a heat balance parameter | index. It is a figure which shows the relationship between a passage heat flux and slab surface temperature. It is the model figure which looked at the casting_mold | template 2 from upper direction. It is a figure which shows the relationship between a molten metal advancing time and an uneven | corrugated occurrence frequency index.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

(Construction of continuous casting equipment)
The continuous casting method of a slab made of titanium or a titanium alloy according to the present embodiment is a method of injecting a plasma arc melted titanium or titanium alloy into a bottomless mold having a rectangular cross section and drawing it downward while solidifying. This is a continuous casting method for continuously casting a slab made of titanium or a titanium alloy. A slab continuous casting apparatus (continuous casting apparatus) 1 made of titanium or a titanium alloy for carrying out this continuous casting method, as shown in FIG. 1 which is a perspective view and FIG. 2 which is a sectional view, A cold hearth 3, a raw material charging device 4, a plasma torch 5, a starting block 6, and a plasma torch 7 are provided. The continuous casting apparatus 1 is surrounded by an inert gas atmosphere made of argon gas, helium gas, or the like.

  The raw material input device 4 inputs the raw material of titanium or titanium alloy such as sponge titanium and scrap into the cold hearth 3. The plasma torch 5 is provided above the cold hearth 3 and generates a plasma arc to melt the raw material in the cold hearth 3. The cold hearth 3 injects the molten metal 12 in which the raw material is melted into the mold 2 from the pouring part 3a at a predetermined flow rate.

  The casting mold 2 is made of copper, has a bottom and has a rectangular cross-sectional shape, and is cooled by water circulating through at least a part of the rectangular tube-shaped wall portion. The starting block 6 is moved up and down by a drive unit (not shown) and can close the lower opening of the mold 2. The plasma torch 7 is provided above the molten metal 12 in the mold 2. The plasma torch 7 plasmas the molten metal 12 injected into the mold 2 while being horizontally moved on the molten metal 12 by moving means (not shown). Heat with an arc.

  In the above configuration, the molten metal 12 injected into the mold 2 is solidified from the contact surface with the water-cooled mold 2. Then, the starting block 6 that has closed the lower opening of the mold 2 is pulled downward at a predetermined speed, so that the prismatic slab 11 with the molten metal 12 solidified is continuously cast while being drawn downward. Is done.

  Here, in the electron beam melting in a vacuum atmosphere, since a minute component evaporates, it is difficult to cast a titanium alloy. On the other hand, in plasma arc melting in an inert gas atmosphere, not only pure titanium but also a titanium alloy can be cast.

  The continuous casting apparatus 1 may have a flux feeding device that feeds a solid phase or liquid phase flux to the molten metal surface of the molten metal 12 in the mold 2. Here, in the electron beam melting in a vacuum atmosphere, since the flux is scattered, it is difficult to put the flux into the molten metal 12 in the mold 2. In contrast, plasma arc melting in an inert gas atmosphere has the advantage that the flux can be charged into the molten metal 12 in the mold 2.

(Operating conditions)
By the way, when the slab 11 made of titanium or a titanium alloy is continuously cast, if there are irregularities or scratches on the surface (casting surface) of the slab 11, a surface defect occurs in the next rolling process. Therefore, it is necessary to remove irregularities and scratches on the surface of the slab 11 by cutting or the like before rolling, which causes a cost increase such as a decrease in yield and an increase in work processes. Therefore, it is required to cast the slab 11 having no irregularities or scratches on the casting surface.

  Here, as shown in FIGS. 3A and 3B, in continuous casting of the slab 11 made of titanium, the vicinity of the molten metal surface of the molten metal 12 heated by a plasma arc or an electron beam (from the molten metal surface to about 10 mm below the molten metal surface). The mold 2 and the surface of the slab 11 (solidified shell 13) are in contact only in the region. In the deeper region, the slab 11 is thermally contracted to generate an air gap 14 with the mold 2. Then, as shown in FIG. 3A, when the heat input to the initial solidification portion 15 (the portion where the molten metal 12 touches the mold 2 and solidifies first) is excessive, the solidified shell 13 is too thin and the solidified shell due to insufficient strength. A “fracture defect” occurs in which the surface of 13 is torn off. On the other hand, as shown in FIG. 3B, when the heat input to the initial solidified portion 15 is insufficient, the molten metal 12 is covered on the grown (thickened) solidified shell 13 to generate a “hot water covering defect”. FIG. 4A shows a surface photograph of the slab 11 in which the “scratch defect” has occurred, and FIG. 4B shows a photograph of the surface of the slab 11 in which the “water bath defect” has occurred.

  Therefore, it is estimated that the heat input / extraction state to the initial solidified portion 15 in the vicinity of the molten metal surface of the molten metal 12 greatly affects the properties of the casting surface. Therefore, it is considered that a slab 11 having a good casting surface can be obtained by appropriately controlling the heat input / extraction state to the initial solidification portion 15 in the vicinity of the molten metal 12.

  Here, as shown in FIG. 5A, which is a model view of the mold 2 as viewed from above, when the slab 11 having a large size of, for example, 250 × 1500 mm is continuously cast by plasma arc melting, the heating range of the plasma torch 7 There is a limit. Therefore, it is necessary to heat the entire molten metal surface using a plurality of plasma torches 7 having a large output. In FIG. 5A, two plasma torches 7 having a large output are used. Further, since the slab 11 is thick, it is necessary to turn the plasma torch 7 along the mold 2 in order to suppress the growth of the solidified shell 13 on the short side or corner of the mold 2. Here, the arrow shown in FIG. 5A indicates the path along which the plasma torch 7 moves.

  However, since the residence time of the plasma torch 7 is long on the long side of the mold 2, the heat input to the initial solidification part 15 is large and the solidification shell 13 becomes thin. On the other hand, since the residence time of the plasma torch 7 is short on the short side or corner portion of the mold 2, heat input to the initial solidification portion 15 is insufficient and the solidified shell 13 grows (thickens). Thereby, the solidification behavior becomes non-uniform depending on the position of the slab 11, which leads to deterioration of cast surface properties.

  Therefore, as shown in FIG. 5B, which is a model diagram of the mold 2 as viewed from above, at least the molten metal surface of the molten metal 12 in the mold 2 is electromagnetically induced by an electromagnetic stirrer (EMS) (not shown) disposed on the side of the mold 2. To stir. The EMS is obtained by winding an EMS coil around a coil iron core. By electromagnetic stirring by EMS, a flow swirling in the horizontal direction (swirl flow) is generated on the molten metal surface or in the vicinity of the molten metal surface. By this swirl flow, the hot molten metal 12 staying in the long side portion of the mold 2 is transferred to the short side portion and the corner portion of the mold 2 where the solidified shell 13 is likely to grow. As a result, the temperature rise on the long side of the mold 2 with a long residence time of the plasma torch 7 and the temperature drop on the short side or corner portion of the mold 2 with a short residence time of the plasma torch 7 are alleviated.

  In addition, the direction of the swirl flow at least on the molten metal surface of the molten metal 12 may coincide with the swirl direction of the plasma torch 7 or may be in the opposite direction. However, the fluctuation range of the surface temperature of the slab 11 can be reduced by turning at least the molten metal surface of the molten metal 12 in the direction opposite to the turning direction of the plasma torch 7.

  Here, when continuously casting a large slab 11, it is necessary to increase the flow rate of the molten metal 12 with a large stirring force in order to transfer heat to the entire molten metal surface by electromagnetic stirring.

  On the other hand, as shown in FIG. 6A, which is a model view of the mold 2 as viewed from above, when the slab 11 having a small size of, for example, 125 × 375 mm is continuously cast by plasma arc melting, the area of the molten metal surface is small. It is possible to heat the entire molten metal surface with one plasma torch 7 having a small output. Moreover, since the thickness of the slab 11 is thin, it is possible to suppress the growth of the solidified shell 13 on the short side or corner of the mold 2 by reciprocating the plasma torch 7 on the same line. Here, the arrow shown in FIG. 6A indicates the path along which the plasma torch 7 moves.

  Further, as shown in FIG. 6B, which is a model diagram when the mold 2 is viewed from above, when continuously casting a small slab 11, even if the stirring force by electromagnetic stirring is small and the flow rate of the molten metal 12 is slow, It is possible to transfer heat to the entire surface of the hot water.

  Thus, the number, output, and movement pattern of the plasma torches 7 necessary for smoothing the casting surface vary depending on the size of the slab 11 to be cast. Moreover, the stirring force of electromagnetic stirring required in order to smooth a casting surface changes with sizes of the slab 11 to cast. Therefore, operation conditions for smoothing the casting surface such as a heat input pattern and an electromagnetic stirring pattern must be evaluated for each size of the slab 11 to be cast, which is complicated.

  Here, a model diagram of the complete contact region 16 between the mold 2 and the slab 11 is shown in FIG. The complete contact region 16 is a region where the mold 2 and the slab 11 are in contact with each other, which is illustrated by hatching from the molten metal surface to about 10 mm below the molten metal surface. In the complete contact region 16, a passing heat flux q from the surface of the slab 11 to the mold 2 is generated. D is the thickness of the solidified shell 13.

FIG. 8 shows the relationship between the passing heat flux q and the surface temperature T _S of the slab 11. Here, the passing heat flux q [W / m ² ] and the surface temperature T _S [° C.] of the slab 11 which are heat balance indexes are evaluated by average values in the complete contact region 16. From this relationship diagram, if the average value of the surface temperature T _S of the slab 11 in the complete contact region 16 between the mold 2 and the slab 11 is in the range of 800 ° C. <T _S <1250 ° C., there is no tearing defect or bathing defect. It can be seen that a slab 11 having a good casting surface can be obtained.

Therefore, in order to obtain the slab 11 having a good casting surface state, the operating state in which the casting surface can be smoothed, while the plasma torch 7 moves once in a predetermined movement pattern, in the complete contact region 16. the average surface temperature of the slab 11 may be clamped to the range 800 ℃ <T _S <of 1250 ° C..

  Therefore, in the present embodiment, a slab 11 having a specific size is cast in advance, and an operation condition capable of smoothing the casting surface of the slab 11 is obtained. And in calculating | requiring the operating condition which can smooth a casting surface in the slab 11 in which this size differs from the slab 11 of this size, the torch moving period and the slab average heat input are the torch moving period and the average of the operating condition obtained in advance. The number, output, and movement pattern of the plasma torches 7 are determined so as to substantially match the heat input.

  In the present embodiment, “substantially coincident” means that the torch movement period and the like are substantially equal to each other, including the complete coincidence, and the operation condition obtained in advance and the operation condition to be obtained. means.

  Here, the torch moving period is a time required to move the plasma torch 7 for heating the molten metal surface of the molten metal 12 in the mold 2 once in a predetermined movement pattern on the molten metal surface. Specifically, the torch moving period is obtained by dividing the moving distance of one plasma torch 7 by the moving speed of the plasma torch 7.

  As shown in FIG. 5A, when casting a large slab 11, the plasma torch 7 is swung on the molten metal surface at a predetermined speed. The time required to make one round of the plasma torch 7 is the torch moving period. As shown in FIG. 6A, when casting a small slab 11, the plasma torch 7 is reciprocated on the molten metal surface at a predetermined speed. The time required to make the plasma torch 7 reciprocate once is the torch moving period.

  Here, when it sees in the same location of the hot_water | molten_metal surface of the molten metal 12, the amount of heat input changes with progress of time because the moving plasma torch 7 approaches or leaves | separates. Further, when viewed from the entire molten metal surface of the molten metal 12, the location near the plasma torch 7 where the amount of heat input is large and the location far from the plasma torch 7 where the amount of heat input is small change as the plasma torch 7 moves. As described above, the movement of the plasma torch 7 causes nonuniformity due to temporal changes and spatial fluctuations in the amount of heat input to the molten metal surface of the molten metal 12.

  Therefore, a heat input pattern in which the temporal change in the amount of heat input to the surface of the molten metal 12 due to the movement of the plasma torch 7 and the non-uniformity due to the spatial variation are substantially equal to the heat input pattern of the operation conditions obtained in advance is obtained. Presumed to be a heat input pattern. Therefore, in order to make the non-uniformity of the heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 substantially coincide with the torch movement period of the operating conditions obtained in advance.

  The average heat input is a plurality of initial solidification portions 15 (portions where the molten metal 12 first solidifies when touching the mold 2) (see FIGS. 3A and 3B) divided into a plurality of portions in the circumferential direction of the mold 2. In each part, the amount of heat input to a certain part is averaged in the length direction along the mold 2 of the part. The slab average heat input is a value obtained by multiplying the average heat input by a correction value. The correction value is a value based on the length that the mold 2 surrounds the torch heating region. The torch heating region is a region where each plasma torch 7 heats the molten metal 12.

  In the present embodiment, as shown in FIG. 9A, which is a model view of the mold 2 as viewed from above, the initial solidified portion 15 is formed along the inner periphery of the mold 2 with corners (1) to (4), long side 1 / 4 (1), (2), long side 1/2 (1), (2), long side 3/4 (1), (2), short side (1), (2), 12 parts in total Dividing into 15a, the average heat input at each part 15a is obtained. Further, as shown in FIG. 9B, which is a model diagram of the mold 2 as viewed from above, representative points are set in each of the twelve portions 15a. That is, in the corners (1) to (4) and the short sides (1) and (2), representative points are respectively set at the center positions in the length direction along the wall surface of the mold 2. Further, in the long side 1/4 (1), (2), the long side 1/2 (1), (2), and the long side 3/4 (1), (2), from the left to the right in the figure. Thus, representative points are set at a position of 1/4 of the long side of the mold 2, a position of 1/2 of the long side of the mold 2, and a position of 3/4 of the long side of the mold 2, respectively.

  As shown in FIG. 10A, which is a model view of the mold 2 as viewed from above, when casting a large slab 11 using two plasma torches 7, half of the molten metal surface of the molten metal 12 is each plasma. This is a torch heating region 17 where the torch 7 heats. On the other hand, when a small slab 11 is cast using one plasma torch 7 as shown in FIG. 10B which is a model view of the mold 2 as viewed from above, the entire molten metal surface of the molten metal 12 is plasma torch. Reference numeral 7 denotes a torch heating region 17 for heating.

  As described above, the growth of the solidified shell 13 near the molten metal surface of the molten metal 12 is greatly affected by the heat input state to the initial solidified portion 15. As shown in FIG. 3A, when the heat input to the initial solidification portion 15 is excessive, a “breakage defect” occurs. On the other hand, as shown in FIG. 3B, when the heat input to the initial solidification portion 15 is insufficient, a “hot water bath defect” occurs.

  Therefore, a heat input pattern in which the heat input state to the initial solidification part 15 substantially matches the heat input pattern of the operating conditions obtained in advance, more specifically, a plurality of parts obtained by dividing the initial solidification part 15 into a plurality of parts. It is presumed that the heat input pattern in which the average heat input amounts in the respective parts 15a substantially coincide is the desired heat input pattern. Therefore, the average heat input at each part 15a is made to substantially coincide with the average heat input at each part under the operating conditions obtained in advance.

  However, the torch heating region 17 shown in FIG. 10A is surrounded on three sides by the mold 2, whereas the torch heating region 17 shown in FIG. And the cooling capability by the casting_mold | template 2 becomes large, so that the casting_mold | template 2 encloses the torch heating area | region 17 long. Therefore, when the cooling capacity of the mold 2 is different between the torch heating region 17 having the operation condition obtained in advance and the torch heating region 17 having the operation condition to be obtained, the average heat input amount of the operation condition to be obtained is determined as the torch heating region. 17 is corrected with a correction value based on the length surrounded by the mold 2. And the slab average heat input which correct | amended the average heat input with the correction value is made to correspond substantially with the average heat input of the operating conditions obtained beforehand.

  In this way, the plasma torch 7 so that the torch moving period substantially coincides with the torch moving period of the operation condition obtained in advance and the slab average heat input substantially coincides with the average heat input of the operation condition obtained in advance. Number, output, and movement pattern are determined. Thereby, the heat input pattern for smoothing a casting surface can be obtained.

  Moreover, in this embodiment, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time by electromagnetic stirring substantially corresponds to the molten metal advancing time of the operating conditions obtained in advance. Here, the molten metal advancing time is a time required for the molten metal 12 to move the width of the torch heating region 17 in the long side direction of the mold 2. Specifically, the molten metal advancing time is a value obtained by dividing the width of the torch heating region 17 (torch effective heating width) in the long side direction of the mold 2 by the flow velocity of the molten metal 12 in the vicinity of the mold 2.

  As shown in FIG. 10A, when casting a large slab 11, the torch heating region 17 of each plasma torch 7 is half the surface of the molten metal 12, so the torch effective heating width is 1 of the long side of the mold 2. The length is / 2. On the other hand, as shown in FIG. 10B, when casting a small slab 11, the torch heating region 17 of the plasma torch 7 is the entire surface of the molten metal 12, so the torch effective heating width is the long side of the mold 2. Full length.

  Here, as shown in FIG. 11, which is a model diagram when the mold 2 is viewed from above, when the plasma torch 7 moves to the left side in the figure of the molten metal surface, the molten metal 12 on the molten metal surface on the right side in the figure away from the plasma torch 7. Temperature drops. Therefore, as indicated by an arrow, the hot molten metal 12 on the left side of the molten metal surface is advected to the right side of the molten metal surface by electromagnetic stirring. Thereby, as shown in FIG. 12, which is a diagram showing the time change of the surface temperature of the slab 11, the temperature drop of the molten metal 12 is alleviated and the slab surface temperature is made uniform as compared with the case where there is no electromagnetic stirring.

  However, when the molten metal advancing time, which is the time required for the molten metal 12 to move through the torch effective heating width, is different, the degree of change in the surface temperature of the slab 11 with the passage of time varies. Specifically, the shorter the molten metal advancing time, the smaller the change in the surface temperature of the slab 11 becomes, and the surface temperature of the slab 11 becomes uniform. Therefore, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the operation condition obtained in advance is the electromagnetic stirring pattern to be obtained. Therefore, the molten metal advancing time is made to substantially coincide with the molten metal advancing time of the operating conditions obtained in advance.

  Thus, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially coincides with the molten metal advancing time of the operating conditions obtained in advance. Thereby, the electromagnetic stirring pattern for smoothing a casting surface can be obtained.

Then, by casting the slab 11 using the obtained heat input pattern and electromagnetic stirring pattern, the average value of the surface temperature T _S of the slab 11 in the complete contact region 16 where the mold 2 and the slab 11 are in contact with each other, It can be contained in a predetermined range (800 ° C. <T _S <1250 ° C.) where a good casting surface can be obtained. Thereby, even if the slab 11 of any size is cast, the casting surface can be smoothed.

(Flow solidification calculation)
Next, operating conditions for obtaining a slab 11 having a good casting surface over the entire circumference were obtained by fluidized solidification calculation. Specifically, an operating condition capable of smoothing the casting surface of a large slab 11 of 250 × 1500 mm in advance is obtained, and the casting surface of a small slab 11 of 125 × 375 mm is smoothed by fluidized solidification calculation. A possible operating condition was obtained.

  Under the operating conditions obtained in advance, as shown in FIG. 5A, the number of plasma torches 7 is two, and the output of each plasma torch 7 is 750 kW. The movement pattern of each plasma torch 7 is a pattern that turns clockwise with the torch effective heating width set to 750 mm. Each plasma torch 7 is turned about 62.5 mm from the casting wall of the mold 2. The moving speed of the plasma torch 7 is constant at 50 mm / sec. Therefore, the torch moving period of the plasma torch 7 is 30 sec.

  On the other hand, under the operating conditions obtained by the flow solidification calculation, as shown in FIG. 6A, the number of plasma torches 7 is one and the output of the plasma torches 7 is 200 to 250 kW. The movement pattern of the plasma torch 7 is a pattern that reciprocates on the same line with the torch effective heating width set to 375 mm. Therefore, as shown in FIG. 13 which is a model diagram when the mold 2 is viewed from above, the moving speed of the central 210 mm is 30 mm / sec out of 290 mm, which is the one-way distance that the plasma torch 7 moves, on both sides thereof. By changing the transfer speed of 40 mm to 10 mm / sec, the torch moving period of the plasma torch 7 was made to coincide with the torch moving period (30 sec) of the operating conditions obtained in advance.

  Further, the average heat input at each part (see FIG. 9A) obtained by dividing the initial solidification portion 15 into 12 parts 15a was made to substantially coincide with the average heat input under the operating conditions obtained in advance. The average heat input in each part 15a is shown in FIG. Here, the output of the plasma torch 7 is 200 kW. In addition, Case (1) is the average heat input of each part 15a under the operating conditions obtained in advance, and Case (2) is each part 15a substantially matched with the average heat input of Case (1) under the required operating conditions. The average heat input.

  Here, as shown in FIG. 10A, the torch heating region 17 is surrounded on three sides by the mold 2 under the operation conditions obtained in advance. On the other hand, as shown in FIG. 10B, the torch heating region 17 is surrounded on four sides by the mold 2 under the operation conditions to be obtained. Thus, since the cooling condition by the casting mold 2 is larger under the operation condition to be obtained, the average heat input is corrected with the correction value. The correction value was calculated from the following equation (1) using the long side and the short side length (mm) of the mold 2 shown in FIG. 6A.

  (375 + 125 + 375 + 125) / (375 + 125 + 375) = 1.3 Formula (1)

When this correction value is applied to the output value of the plasma torch 7, the output is 250 kW. Further, the slab average heat input, which is a value obtained by correcting the average heat input with this correction value, is shown in FIG. 14 as Case (3). The growth of the solidified shell 13 in the vicinity of the molten metal surface of the molten metal 12 can be suitably suppressed by setting the average slab heat input at each portion 15a to 1.0 MW / m ² or more. Thereby, the slab 11 of a favorable casting surface can be obtained.

  15A to 15C show the relationship between the passing heat flux and the slab surface temperature when electromagnetic stirring is not performed. FIGS. 15A to 15C are plots obtained by extracting heat balance indices (passing heat flux and slab surface temperature) at 12 representative points shown in FIG. 9B while the plasma torch 7 moves once in a predetermined movement pattern. It is a thing. FIG. 15A shows the relationship between the passing heat flux and the slab surface temperature in Case (1). In contrast, FIG. 15B shows the relationship between the passing heat flux and the slab surface temperature in Case (2). The distribution range of the heat balance index in Case (2) is shifted to the low temperature side as a whole with respect to Case (1).

  FIG. 15C shows the relationship between the passing heat flux and the slab surface temperature in Case (3). The distribution range of the heat balance index in Case (3) is equivalent to that in Case (1), and it can be seen that the non-uniformity in the amount of heat input to the molten metal surface of the molten metal 12 is comparable.

  A histogram of the heat balance index is shown in FIG. 16A. Moreover, the cumulative distribution of the heat balance index is shown in FIG. 16B. Case (1) and Case (3) have substantially the same heat balance index distribution, and both have a heat input pattern that can be regarded as similar.

  As described above, the torch moving period substantially coincides with the torch moving period of the operation condition obtained in advance, and the slab average heat input amount substantially coincides with the average heat input of the operation condition obtained in advance at each portion 15a. Furthermore, it can be seen that a heat input pattern for smoothing the casting surface can be obtained by determining the number, output, and movement pattern of the plasma torches 7.

  Next, the stirring force of the electromagnetic stirring was determined so that the molten metal advancing time substantially coincided with the molten metal advancing time of the operating conditions obtained in advance, and electromagnetic stirring was applied according to the electromagnetic stirring pattern. The relationship between the passing heat flux and the slab surface temperature in Case (1) is shown in FIG. 17A, the relationship between the passing heat flux and the slab surface temperature in Case (2) is shown in FIG. 17B, and the passing heat flow in Case (3). FIG. 17C shows the relationship between the bundle and the slab surface temperature. By adding electromagnetic stirring, the variation of the heat balance index is relaxed, and a soaking effect is exhibited. In Case (1) and Case (3), the heat balance index is within the target range. On the other hand, in Case (2), an area below the target range exists. This is presumed to be caused by insufficient heat input due to the large cooling capacity of the mold 2.

  A histogram of the heat balance index is shown in FIG. 18A. Further, FIG. 18B shows the cumulative distribution of the heat balance index. In Case (1) and Case (3), the distribution of the heat balance index is almost the same.

  Thus, the electromagnetic stirring pattern for smoothing the casting surface can be obtained by determining the stirring force of the electromagnetic stirring so that the molten metal advancing time substantially coincides with the molten metal advancing time obtained in advance. I understand that. Then, it is understood that the distribution of the heat balance index is almost the same regardless of the slab size by making the heat input pattern substantially the same and then making the electromagnetic stirring pattern substantially the same.

Here, as shown in FIG. 19 which is a relationship diagram between the passing heat flux and the slab surface temperature, the deviation width of the surface temperature T _S of the slab 11 from the target temperature range (800 ° C. <T _S <1250 ° C.) is defined as a. _{Assuming j} , the unevenness occurrence frequency index D indicating the frequency of occurrence of unevenness on the surface of the slab 11 is expressed by the following equation (2).

D = Σa _j / N (2)
Here, N is the total number of data extraction points. When D is 0, the unevenness of the casting surface in all the locations of the slab 11 is below the target range, indicating that good casting surface properties can be obtained.

  Further, as shown in FIG. 20 which is a model diagram when the mold 2 is viewed from above, the flow velocity (absolute value) in the X direction at a position 10 mm away from the inner surface of the mold 2 is −2L / 5 ≦ x ≦ 2L /. The melt advection time was determined using the average value in the range of 5.

  The relationship between the molten metal advancing time and the unevenness occurrence frequency index D is shown in FIG. In this relationship diagram, the calculation results for each stirring force are plotted while changing the stirring force of electromagnetic stirring. Here, the stronger the stirring force of electromagnetic stirring, the faster the flow rate of the molten metal 12 and the shorter the molten metal advancing time. In addition, the smaller the unevenness occurrence frequency index D, the better the properties of the casting surface.

  From the calculation results of Case (1) and Case (3), if the heat input pattern is controlled so as to be similar to the heat input pattern of the operating conditions obtained in advance, the electromagnetic agitation can be performed according to the melt advection time without depending on the slab size. It can be seen that the relationship between conditions and casting surface properties can be organized. And it turns out that casting surface property improves by making stirring force increase so that molten metal advection time may be 3 sec or less. Thus, the behavior of casting surface improvement by electromagnetic stirring can be quantitatively evaluated by the molten metal advancing time.

  On the other hand, from the calculation result of Case (2), it can be seen that if the heat input pattern is not similar to the heat input pattern of the operation conditions obtained in advance, the casting surface properties are not improved even if the stirring force is increased.

(effect)
As described above, according to the continuous casting method of a slab made of titanium or a titanium alloy according to the present embodiment, the torch moving period substantially coincides with the torch moving period of a predetermined operating condition, and the slab average heat input amount However, the number, the output, and the movement pattern of the plasma torches 7 are determined so as to substantially match the average heat input amount of the above operating conditions. Here, with respect to the heat input pattern under the above operating conditions, a heat input pattern in which the temporal change in the amount of heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 and the non-uniformity due to spatial fluctuations are almost the same is obtained. Presumed to be a heat input pattern. Therefore, in order to make the non-uniformity of the heat input to the molten metal surface of the molten metal 12 due to the movement of the plasma torch 7 substantially coincide with the torch movement period of the above operating conditions. In addition, a heat input pattern in which the average heat input amount in each of the plurality of portions 15a formed by dividing the initial solidification portion 15 into a plurality of parts is the heat input pattern to be obtained with respect to the heat input pattern of the above operating conditions. It is estimated to be. Therefore, the average heat input at each part 15a is made to substantially coincide with the average heat input at each part under the above operating conditions. Here, the longer the length that the mold 2 surrounds the torch heating region 17, the greater the cooling capacity of the mold 2. Therefore, the slab average heat input obtained by correcting the average heat input with the correction value is made to substantially coincide with the average heat input of the above operating conditions. Thus, the heat input pattern for smoothing a casting surface can be obtained by determining the number of plasma torches 7, an output, and a movement pattern.

Moreover, according to this embodiment, the stirring force of electromagnetic stirring is determined so that the molten metal advancing time substantially matches the molten metal advancing time of the above operating conditions. Here, it is presumed that the electromagnetic stirring pattern in which the molten metal advancing time substantially matches the electromagnetic stirring pattern of the above operating conditions is the desired electromagnetic stirring pattern. Therefore, the melt advection time is made to substantially coincide with the melt advection time under the above operating conditions. Thus, the electromagnetic stirring pattern for smoothing a casting surface can be obtained by determining the stirring force of electromagnetic stirring.

Then, by casting the slab 11 using the obtained heat input pattern and electromagnetic stirring pattern, the average value of the surface temperature T _S of the slab 11 in the complete contact region 16 where the mold 2 and the slab 11 are in contact with each other, It can be contained in a predetermined range (800 ° C. <T _S <1250 ° C.) where a good casting surface can be obtained. Thereby, even if the slab 11 of any size is cast, the casting surface can be smoothed.

Moreover, the growth of the solidified shell 13 in the vicinity of the molten metal surface of the molten metal 12 can be suitably suppressed by setting the average slab heat input to 1.0 MW / m ² or more. Thereby, the slab 11 of a favorable casting surface can be obtained.

  Moreover, casting surface property can be improved suitably by increasing the stirring force of electromagnetic stirring and making molten metal advection time 3 sec or less.

(Modification of this embodiment)
The embodiment of the present invention has been described above, but only specific examples are illustrated, and the present invention is not particularly limited, and the specific configuration and the like can be appropriately changed in design. Further, the actions and effects described in the embodiments of the invention only list the most preferable actions and effects resulting from the present invention, and the actions and effects according to the present invention are described in the embodiments of the present invention. It is not limited to what was done.

DESCRIPTION OF SYMBOLS 1 Continuous casting apparatus 2 Mold 3 Cold hearth 3a Pouring part 4 Raw material injection | throwing-in apparatus 5 Plasma torch 6 Starting block 7 Plasma torch 11 Slab 12 Molten metal 13 Solidified shell 14 Air gap 15 Initial solidification part 15a Part 16 Complete contact area 17 Torch heating region

Claims (3)

This is a continuous casting method in which a slab made of titanium or a titanium alloy is continuously cast by pouring molten metal in which titanium or a titanium alloy is melted into a bottomless mold having a rectangular section and solidifying the molten slab. And
Casting a slab of a specific size in advance, obtaining operating conditions that can smooth the casting surface of this slab,
The torch moving period, which is the time required to move the plasma torch for heating the molten metal surface in the mold once in a predetermined movement pattern on the molten metal surface, is approximately equal to the torch moving period of the operating conditions. In each of a plurality of portions obtained by dividing the initial solidified portion, which is a portion that the molten metal first contacts the mold and solidifies, into a plurality of circumferential directions of the mold, the amount of heat input to the portion is A correction value based on the length of the mold surrounding the torch heating region, which is a region heated by each plasma torch, of the molten metal surface is averaged to the average heat input amount in the length direction along the mold of the part. The number of the plasma torches, the output, and the movement pattern are determined so that the slab average heat input which is the multiplied value substantially matches the average heat input of the operation conditions,
The molten metal advancing time, which is the time required for the molten metal to move through the width of the torch heating region in the long side direction of the mold, by electromagnetic stirring that causes a flow swirling in the horizontal direction on at least the molten metal surface, A continuous casting method of a slab made of titanium or a titanium alloy, wherein the stirring force of the electromagnetic stirring is determined so as to substantially coincide with the molten metal advancing time of the operating conditions. The said slab average heat input is 1.0 MW / m < ² > or more, The continuous casting method of the slab which consists of titanium or a titanium alloy of Claim 1 characterized by the above-mentioned. The continuous casting method of a slab made of titanium or a titanium alloy according to claim 1 or 2, wherein the molten metal advection time is 3 sec or less.