US2804663A

US2804663A - Control system for continuous casting

Info

Publication number: US2804663A
Application number: US364416A
Authority: US
Inventors: Jr Isaac Harter; Temple W Ratcliffe; Robert R Mcclain
Original assignee: Babcock and Wilcox Co
Current assignee: Babcock and Wilcox Co
Priority date: 1953-06-26
Filing date: 1953-06-26
Publication date: 1957-09-03
Anticipated expiration: 1974-09-03
Also published as: LU32963A1; GB762030A; BE529941A; FR1107669A

Description

Sept. 3, 1957 RTER, JR ETAL 2,804,663

CONTROL SYSTEM FOR CONTINUOUS CASTING 5 Sheefs-Sheet 1 Filed June 26, 1953 FlG.l

INVENTORS ASA/4C HAETEELJK. TEMPLE WPATCZ/FFE H A my a BY ROBE/97R. 4490mm TTORNEY Sept. 3, 1957 HARTER, JR, ET AL 2,804,663

- CONTROL SYSTEM FOR CONTINUOUS CASTING Filed June 26, 1953 5 Sheets-Sheet 2 F IG. 6

INVENTORS lSAAC HARTER,J, 75/14/ 15 WRATCL/FFE ATTORNEY Sept. 3, 1957 l. HARTEI Q, JR., ET AL 2,804,663

CONTROL SYSTEM FOR CONTINUOUS CASTING 5 Sheets-Sheet 3 Filed June 26, 1953 SYII- Rmam Y T E T AM N NM R E W 0 w n AL aw A 2. Y B 4 G F l/ w l. HARTER, JR., ETAL 2,804,663

CONTROL SYSTEM FOR CONTINUOUS CASTING 5 Sheets-Sheet 4 F- e F w M a T R MM M N 3&9: wmw e n W I A A QT W w [W NR Y B 236: vi g km 3 El F fil lfik & :5 v

Sept. 3, 1957 Filed June 26, 1953 Sept. 3, 1957 I. HARTER, JR., ET AL 2,804,663

CONTROL SYSTEM FOR CONTINUOUS CASTING 5 Sheets-Sheet 5 Filed June 26, 1953 228% 9 25% zQwsfi $1M sRu Ric Y 0 E T N N R m o VCPCR T NAZF. T IAPB A 6M United States Patent O CONTROL SYSTEM non CONTINUOUS cAs'riNo Isaac Harter, Jr., and Temple W. Ratclitfe, Beaver, and Robert R. McClain, Beaver Falls, Pa., assignors to The Babcock & Wilcox Company, New York, N. Y., a cerporation of New .lersey Application June 26, 1953, Serial No. 364,416

2. Claims. (Cl. 22-571) The present invention relates to electrical control sy tems, and more particularly to control systems for regulating the operation of continuous casting units for the production of metallic ingots.

In the continuous casting of ferrous metals, for example, molten metal is delivered to the continuous casting mold at a substantially uniform temperature. The molten metal is cooled within the casting mold by heat exchange with a heat exchange fluid so that a skin or shell of solidified metal is formed, and an embryo casting may be withdrawn from the mold. Advantageously, the molten metal level within the mold is varied to obtain a high rate of heat exchange between the metal being cast and the fluid cooled mold wall. The rate of heat transfer to the mold wall is advantageously increased with changes in the molten metal level by reason of the heat storage capacity of the mold walls and the increased conformity and heat exchange contact between the metal being cast and the mold wall. The molten metal level changes within the mold also have a tendency to move surface accumulations of oxides and slag to the periphery of the mold so that the frozen slag will form on the skin of the ingot with a minimum efiect on the quality of the product.

In the embodiment of the invention hereinafter described, the change in the molten metal level is obtained by changing the withdrawal rate of the embryo casting from the mold, and maintaining the delivery rate of molten metal to the mold substantially uniform. The regulation of the withdrawal mechanism is accomplished by an automatic control device which is actuated by the molten metal level rising to an upper position in the mold whereby the withdrawal mechanism is operated at a speed such that the embryo casting is withdrawn at a rate greater than the rate of molten metal delivery to the mold. This causes the molten metal level to be lowered in the mold. The high speed of the casting withdrawal is continued for a selected length of time which is coordinated with the withdrawal rate and the molten metal pour rate to cause the molten metal level within the mold to fall to a lower position. Thereafter, the withdrawal mechanism is stopped, or operated at a reduced speed, so that the molten metal level will again rise in the mold. When the molten metal rises to the desired upper position in the mold, the pattern of withdrawal mechanism operation is repeated.

With the rise and fall of the molten metal within the mold controlled by operation of the withdrawal mechanism, it is both possible and practical to regulate the rate of molten metal delivery to the mold in coordination therewith. As hereinafter described, the rate of delivery of molten metal to the mold is regulated by an anticipating control system of the general type disclosed and claimed in a copending application of S. 0. Evans and I. Harter, Jr., Serial No. 152,404, filed March 28, 1950, now Patent 2,709,284, issued May 31, 1955.

In the present invention, the continuous casting control system includes means for operating the casting withdrawal mechanism which is actuated by heat sensitive means, such as one or more thermocouples embedded in the wall of the mold. Means are also provided in the withdrawal mechanism control system whereby a visual and/or audible signal is utilized to inform the operators of operating conditions. In addition, a thermocouple is positioned in the mold wall at a position adjacent the upper edge of the mold. When the molten metal rises in the mold to the position of this couple, an automatic means reverses the tilting direction of the molten metal sources to quickly stop the delivery of molten metal to the mold. The rate of molten metal delivery to the mold is regulated in accordance with the deviation between the actual rate of molten metal delivery to the mold and a standard rate of delivery, Where the deviation determined in one cycle of operation is used to compensate for the deviation in a succeeding cycle of operation.

The various features of novelty which characterize our invention are pointed out with particularity in the claims annexed to and forming a part of this specification. For a better understanding of the invention, its operating advantages and specific objects attained by its use, reference should be had to the accompanying drawings and descriptive matter in which we have illustrated and described an embodiment of the invention.

Fig. 1 is an elevation, partly in section of a continuous casting apparatus operated in accordance with the control system of the present invention;

Fig. 2 is an enlarged elevation, partly in section, of the upper portion of the apparatus shown in Fig. l, as viewed from the opposite side;

Fig. 3 is a further enlarged section of the mold portion of the apparatus shown in Fig. 2;

Fig. 4 is an enlarged elevation, partly in section taken on the line 4-4 of Fig. 5, of the screw drive mechanism of the arc furnace shown in Fig. 2;

Fig. 5 is a plan, partly in section, taken on line 5-5 of Fig-4;

Fig. 6 is an elevation, partly in section, taken on line 6-6 of Fig. 5;

Fig. 7 is a wiring diagram of the control system applied to the apparatus shown in Fig. 1, with the circuit shown in the conventional dead board condition; and

Fig. 8 is a graphical illustration showing the operation of the molten metal pouring control system.

The invention is illustrated in the drawings as applied to a continuous casting unit of the general type disclosed and claimed in the copending application of I. Harter, Jr., et al., Serial No. 316,446, filed October 23, 1952. As shown, the casting apparatus includes metal pouring means arranged to deliver molten metal at a predetermined substantially uniform temperature into the upper end of a vertically arranged stationary, liquid cooled casting mold. The molten metal is cooled within the mold by heat exchange with a cooling liquid so as to solidify a rim or shell of solidified metal around an embryo casting. Leaving the mold, the shell of the casting has sufficient thickness and strength to be generally self-supporting and is withdrawn from the mold by a withdrawal mechanism. Beneath the mechanism the casting may be cut into lengths, coiled or otherwise handled as desired.

As shown in Figs. 1 and 2, the molten metal is prepared in a tilting arc furnace 20 with the stream of molten metal discharged from the furnace 20 through a tun dish 21. The tun dish is constructed and arranged for the delivery of a substantially slag free stream of molten metal to the open upper end of the continuous casting mold 22. As shown, the molten metal stream leaving the furnace is protected against oxidation by the use of shields, 23 and 24 as disclosed in said copending application. Preferably, the space within the shields is filled by an inert gas to displace oxygen, or by a gas which burns to consume oxygen.

As shown in Fig. 2, th arc furnace is tiltable about trunnions mounted in trunnion bearings 26, with the axis of tilt formed thereby movable in a horizontal direction on the pedestals 27. Furnace tilting is accomplished by a motor driven screw 28 operable within a nut assembly 30 secured to the furnace by a fixed arm 31. The tun dish is rotatably positioned about a horizontal axis 29 by a motor driven gear reducer (not shown), and is horizontally positioned relative to the mold 22 by a mechanism such as disclosed in said copending application. The operation of the furnace and tun dish tilting .mechanism is hereinafter described in more detail.

The molten metal delivered to the mold 22 is cooled by heat transfer with a cooling medium which is passed at high velocity through a cooling passageway surrounding the mold. As shown in Figs. 1, 2 and 3, the cooling fluid consists of water which is introduced to the mold through a main conduit 32 which is provided with branch connections 33 opening to opposite sides of a distributing chamber 34 positioned at the upper end of the mold. The water leaving the mold 22 is deflected away from the casting by a deflector 35 to enter a housing 36 which encloses the lower end of the mold, and an after-cooling,

guiding and supporting section 37 for the casting 38.

The casting emerging from the lower end of the mold 22 and moving downwardly into the after cooling section 37 is subjected to spray cooling from a plurality of cooling water jets impinging directly upon the casting surface. The stream and vapors produced by contact between the cooling water spray and the hot casting are discharged from the housing 36 through a vent conduit 40. The casting passing through the after-cooling section 37 is restrained from outward expansion by a plurality of spring loaded guide rolls 41.

Cooling water discharged from the mold 22 and water from the cooling spray in the after-cooling section 37 accumulates in the lower portion of the housing 36 from which some is withdrawn by a pump 42, and some is continuously withdrawn through the valved overflow pipe 43. The pump discharges the water through a pair of filters 44 and the pipe 45 for reintroduction through vertical manifolds 46 and valved branch connections 47 to the spray nozzles which are positioned in upright water distributing pipes 48.

As shown in Fig. l, the casting withdrawal mechanism indicated generally at 50 includes two pairs of rolls, with a pair of rolls engaging opposite sides of the casting 38, and the roll pairs vertically spaced. The rolls are driven in concert at a controlled speed by a variable speed motor 49. It is important for the casting to be supported in the guide rolls of the after cooling section 37 and in the withdrawal rolls so that the axis of the casting 38 is coaxial with the mold 22. As hereinafter described, the pinch rolls of the withdrawal mechanism 50 are operated in a cycle of changing speeds so as to vary the molten metal level within the mold 22.

In the illustrated embodiment of the invention, the withdrawal mechanism 50 is operated in an intermittent cyclic pattern including a run portion during which the withdrawal mechanism withdraws the casting at a weight rate greater than the pouring weight rate of molten metal to the mold. The operation of the withdrawal mechanism for the run portion of the cyclic operation is initiated by the molten metal level within the mold rising to a predetermined position within the mold 22. When the molten metal reaches the upper position, the presence of the molten metal is detected by heat sensitive means associated with the wall of the mold. The duration of the run period is determined by a timer mechanism which is coordinated with the withdrawal rate so that the molten metal level within the mold will drop a predictable amount, providing the-pour rate is substantially as desired. At-the end of the timed period the withdrawal mechanism is stopped by the operation of the timer and the moldten metal is permitted to rise during the dwell portion of the cycle until the molten metal level again actuates the heat sensitive means to repeat the cycle.

As shown particularly in Fig. 3, the heat sensitive means includes a thermocouple 51 which is inserted in a bore 52 drilled longitudinally of the mold wall. The thermocouple is in heat exchange contact with the inner surface of the mold wall 54 so that any increase in temperature of the wall at this position, as caused .by the presence of molten metal at this level, results in a measurable change in the electromotive forces produced by the thermocouple. The upper level of molten metal within the mold is indicated at 55, while the lower level is indicated at 56. Preferably, at least two thermocouples are imbedded in the wall 54 on the opposite sides of the mold with their effective contact on the wall at substantially the same horizontal level.

In the embodiment shown, one insulated wire of the thermocouple is extended into the bore 52, with the other side of the thermocouple fastened to the wall of the mold as shown at 57. The thermocouples are connected to separate amplifiers (not shown) which in turn are connected with and deflect millimeters of the high-low limit contact type, so that changes in the temperature detected by the thermocouples causes deflection of the ammeter arms to contact either the high or low limit contacts. The described amplifiers and deflecting millimeters are well known, and are indicated at 58. The high contact limits of the ammeters are in effect, connected in parallel, and when the molten metal reaches the upper level 55, either orboth ammeter contacts are closed to activate a circuit which initiates the operation of the withdrawal mechanism '50.

The wiring diagram of the withdrawal mechanism control system isshown on the left side of Fig. 7. As shown, closing contact DD (as accomplished by the thermocouple circuit) energizes relay coil 10 to close contact 10A and energize relay coils WRA and WS. Simultaneously, a reversing or latching relay LL is energized by the closing of contact 103 so that the withdrawal timer motor W will run a preset period of time. The withdrawal timer will operate in-one direction, i. e. clockwise or counterclockwise until an arm on the timer adjustable limit microswitch contacts a BM or A-M. During the clockwise operation of the timer W, relay 1 will be energized and contact 1A closed to furnish power to relay WR thereby closing contact WRl in the drive circuit of the withdrawal mechanism 50. The withdrawal mechanism will operatesimultaneously with and for the same period of time as-the withdrawal timer W.

The withdrawal timer W is a reversible, R. P. M., non-coasting synchronous motor driving a screw. A nut equipped with ball-bearings travels longitudinally along the screw. Extension arms on the nut actuate one fixed microswitch A-M at the clockwise limit of motor rotation and a microswitch BM, whose lateral position is adjustable, atthe counter-clockwise limit of motor rotation. By adjusting the spacing between microswitches AM and BM the withdrawal timer may be preset for a desired withdrawal time. The direction of rotation of the withdrawal timer motor depends upon the position of latching relay LL. The motor is so wired in the control circuit (as shown) that it must complete a run in one direction and actuate one of the limit microswitches .to energize relay A or B, before it can run in the opposite direction. Contacts A2 and B2 are so arranged that they prevent latching relay LL from receiving a multiple signal (due to thermocouples lagging each other) which would cause it to reverse, and reverse the direction of rotation of the withdrawal timer motor before it had completed a timed run. To assure that the withdrawal timer-motor will start in the desired direction, contact 1-3 :or '-2B opens when the motor is started, respectively, clockwise -or counter-clockwise to prevent relays l or 2 from being shunted across whichever motor winding is not energized.

When the withdrawal timer motor is running clockwise, relay 1 is energized; when it is running counter-clockwise, relay 2 is energized. Relay WR is thus energized through either contact 1A or 2A. (depending upon direction of rotation of the withdrawal timer motor) and remains energized closing contact WR, to signal the withdrawal mechanism 50 drive circuit of the motor 49 to operate the withdrawal rolls as long as the withdrawal timer is completing a timed run.

It will be understood that it is possible for unusual occurrences in the mold to cause false indications of the molten metal level within the mold 22. Such false indications may be caused by the presence of a thin shell of solidified metal temporarily positioned within the mold, masking the thermocouples position and in effect insulating the thermocouple 51 from the temperature efiect of the molten metal. It is also possible to have a molten metal delivery rate to the mold which is high enough to maintain a molten metal level generally at the level of the thermocouples 51 even though the withdrawal mechanism 56 is operating.

If operating conditions are such that the molten metal level remains at the position of the thermocouple, the high rate contact switch DD shown in Fig. 7 will remain closed to hold the relay if in an energized condition. With relay it energized, contact A will be closed, holding relay WRA energized whereby the contacts WRA-l in the variable speed drive of motor 49 continues to operate the withdrawal mechanism 56. This condition will occur regardless of the preset time cycle ordinarily used to regulate the operation of the run portion of the withdrawal cycle. This condition will continue until the molten metal level falls below the level of the thermocouple position i. e. below level 55 (Fig. 3).

A safety warning system is incorporated in the molten metal level indicator assembly whereby a signal is given if the molten metal level does not fall from the thermocouple level after a preset length of time, as determined by time delay relay HHL. HHL coil is energized through either contact 1A or 2A when the Withdrawal timer motor W is operating and contact HHL-l will close after the interval of time required for the molten metal level to fall from the thermocouple 51 level during a normal withdrawal. If at this time the molten metal level is still high (in the vicinity of level 55), and contact W82 is closed due to the fact that relay WS is energized through contact 10A of relay 1%) which is in turn energized when high limit contact DD is closed, power is furnished to the high level warning signal light WL.

Another signal is given if a condition occurs where a portion of a solidified rim or shell of the freezing ingot sticks to the upper regions of the mold wall 54 after the bar proper has been withdrawn, maintaining the signal from the thermocouple 51 at a sufficiently high amplitude to register in the molten metal level indicator so that low limit contact E of the millimeter in the control 58 is closed, but of insulficient amplitude to close high limit DD of the contact millimeter so that relay it? is not energized and relay WS is tie-energized maintaining contact WSll closed. When contact HHL-l of time delay relay HHL now closes after the previously specified time interval, power is furnished to a warning signal HB which may be visual or audible.

As shown in Figs. 3, 4, 5 and 6, means are provided to avoid overflow of molten metal over the top of the mold 22. This is accomplished by the installation of a thermocouple 60 in the mold wall 54 at a level atjacent the upper end 61 of the mold (see Fig. 3). For example, the thermocouple 69 may be positioned 3 to 6 inches from the top of the mold. When the molten metal rises within the mold to the position of the thermocouple 60, the electromotive force produced by the couple is amplified and utilized to actuate a contact in a milliampere meter of the type described in connection with couple 51. Closing the contact energizes a coil in control 58 which in turn stops the motor 62. (Fig. 2) driving the tilting mechanism of the pouring vessel 20, reverses the wiring circuit of the motor and restarts the motor in the opposite direction, and rotates the tun dish 21 about its axis 29 to stop flow of metal to the mold 22.. Simultaneously with the motor reversal the coil actuates a power piston which shifts the position of a pair of mechanical clutches in the gear drive of the vessel 26 tilting mechanism to increase the rotational movement of the screw 28 so that the vessel is quickly moved to a nonpouring position.

The tun dish 21 is advantageously provided with a discharge weir 21' in its wall adjacent the vessel 20, so that excess metal in the tun dish can be discharged into a fixed mold instead of the continuous casting mold 22 when the thermocouple 60 is actuated by a high molten metal level in the mold 22.

While the withdrawal mechanism will ordinarily continue in operation as long as the molten metal level remains above the position of thermocouple 51, actuation of the thermocouple closes a contact 51 (see Fig. 7) in the operating circuit of the motor 49 to insure operation of the withdrawal mechanism during the period the molten metal level is at or above the position of the thermocouple 60. When the molten metal level falls below the level of the thermocouple 60, the couple is de-energized and the molten metal supply means returns to normal operation.

The tilting mechanism of the furnace 26 is shown generally in Fig. 2 and includes the variable speed motor 62 which rotates the screw 28 through a suitable speed reducer and gear train 63. Referring to Figs. 4, 5 and 6, the input shaft 64 of the speed reducer 65 is connected with the motor 62 and drives a gear train through its output shaft 66. As shown, the output shaft of the speed reducer 65 is connected to the hub of a spur gear 67 which engages a pinion 68 on a jack shaft 70. The jack shaft is mounted in

bearings

71 and 72, and is spaced to one side of the output shaft 66 of the speed reducer and is parallel thereto. The pinion gear is journaled for rotation on a sleeve 73 on the jack shaft 70 and is connected thereto through a jaw clutch 74 which is selectively engaged as hereinafter described. The clutch i4 is slidably keyed to the jack shaft 70. A spur gear 75 is keyed to the jack shaft adjacent the bearing 72 with the gear in turn meshing with a pinion 76 which is keyed to a shaft 77 in axial alignment to the output shaft 66 of the speed reducer 65. The shaft 77 is provided with an outboard bearing (not shown) beneath the lower end of the screw 28 drive mechanism 78, and with an inboard bearing 80 which is supported in the hub of the spur gear 67. A jaw type clutch 81 is slidingly mounted on the shaft 77 to engage teeth on the hub of the spur gear 67. The pinion 76 on the shaft 77 also meshes With a gear 82 which is offset to the opposite side of the shaft 77 from the jack shaft 70, and is directly connected with the screw drive mechanism 78.

The clutches .74 and 81 are interconnected by a horizontally disposed rod 83 which is supported intermediate the length thereof by a pivot 34. A stub arm 85 extending normal to the rod 83 is pivotally connected at 86 with a drive rod 87 extending from a hydraulic pistonhtl. With the linkage described, movement of the clutches is accomplished through actuation of the double'aeting hydraulic piston whereby one of the clutches engages while the other clutch is disengaged. With this arrangement the rotational motion of the speed reducer 65 may be ransmitted through the spur gear 67, pinion 68, gear 75 and pinion 76 to the spur gear 82 of the screw drive mechanism 78. Alternately, the output shaft 66 is directly connected to shaft 77 by the clutch 81. The direct drive through the shafts 66 and 77 will result in a lower speed of the vessel tilting mechanism, at the same '7 motor speed than when the drive isconnected from shaft 66 to thejack shaft 70 and thence to the screw drive mechanism 78. The slow speed alwaysraises the furnace and the high speed always lowers the furnace.

In the normal operation of the continuous casting unit the withdrawal mechanism is controlled in alternate run and dwell cycles with infrequent use of the safety elements described. The duration of the desired dwell time is determined by a dwell timer incorporated in the control circuit shown in Fig. 7. The desired dwell timer D, like the withdrawal timer, always operates for a specific length of time. The dwell timer always starts when the withdrawal timer W stops, upon completion of a timed withdrawal. During a cycle in which the actual dwell time is longer than the desired dwell time (i. e. the pouring rate is too slow), the dwell timer stops before the withdrawal timer starts. During a cycle in which the actual dwell time is shorter than the desired dwell time (i. e. the pouring rate is too rapid) the withdrawal timer starts before the dwell timer stops.

With the withdrawal mechanism operated in an intermittent cycle as hereinbefore described, any variation in the actual period of time required to raise the molten metal level within the mold as compared with the desired period of time, as determined by the desired dwell timer will be indicative of an error in the rate of molten metal delivery to the mold. As shown in Fig. 8, the comparison between the actual and desired dwell time is utilized to regulate the pouring rate of the vessel. This is accomplished as shown in Fig. 7 by a controlled adjustment of a vernier rheostat positioned in the power circuit of the vessel tilting motor, as supplemented (when desired) by the controlledadjustment of a master control rheostat likewise inserted in the power circuit of the vessel tilting motor.

The vernier controlconsists in a pair of drive motors LV and MV operated in response to the deviation between the desired and actual dwell times in the continuous casting apparatus. The output shafts of motors LV and MV are connected into a differential gear DA with the output shaft therefrom connected with a second differential gear DB. The second differential gear DB receives the rotational output from motor NV while the output shaft of the differential gear DB is connected through a clutch CL-l to the Vernier rheostat.

The motor NV starts at the end of the desired time (dwell timer) except in-a short dwell when it starts on the end of the actual dwell time (as actuated by the thermocouple 51). The motors LV and MV always start at the beginning of the dwell (whenthe withdrawal mechanism stops) (or at the end of run timer W operation). LV and MV run in opposite directions and alternate in direction of rotation in successive cycles. Also one of the two (alternately) has a memory or deviation from the preceding cycle. Thus the operation of the two motors LV and MV gives a net of present actual deviation as compensatedby the previous deviation. The

net is fed into the differential. at the end of the shorter of the two deviations, whether actual or memory. L and M timer motors always run in opposite directions; if one of the motor shafts is rotating clockwise the other must rotate counter-clockwise. Both L and M timers start when the dwell timer starts. Whichever timer motor is driving its nut toward the clockwise limit (EM or FM for L timer or M timer respectively) it always runs until it opens this limit. The other timer motor runs counterclockwise forra period of time which is equal to the actual dwell time. a i e V The two synchronous motors LV differential motor and MV differential motor are of.,the same specifications as the timer motors, and operatein parallel with L timer motor and M timer motor respectively.- .These motors drive differentialDA in such a manner that the output of the driven shaftof differentialDA is zero if ',both

of'them'run for'the same length of time. If 'either'LV' differential motor or MV differential motor continues to run after the other has stopped, there is an output from the driven shaft. The output is such as to increase the rate of pour if the differential motor which continues to run is operating parallel to whichever timer motor is running counter-clockwise. Conversely, the output is such as to decrease the rate of pour if the differential motor which continues to run is operating parallel to whichever timer motor is running clockwise.

Another synchronous motor, NV differential motor, of the same specifications as the timer motors, drives one spur gear of differential DB. The driven shaft of differential DA drives the other spur gear of differential DB. The output from the driven shaft of differential DB is always the sum of the output of differential DA and the output of NV differential motor, multiplied by an appropriate constant which depends upon the gear ratio employed. For example, if NV differential motor and L or M differential motors are both running in a direction to either increase or decrease the rate of pour, and both run for the same length of time, the output from the driven shaft of differential DB is twice that of either one (multiplied by a suitable factor for gear ratio); if one of the two is running in a direction to increase the rate of pour, and the other to decrease the rate of pour, and both run for the same length of time, the output of differential DB is zero.

NV differential motor runs so as to decrease the rate of pour when both the withdrawal timer W and the dwell timer D are operating. It is evident that this condition (i. e. the overlapping of operation of the withdrawal timer and the dwell timer) occurs when the pouring rate is too high, and high metal level causes the withdrawal portion of the cycle to commence before the desired dwell time has elapsed. NV differential motor runs so as to increase the rate of pour when neither the withdrawal timer W nor the dwell timer D is operating. This condition (i. e. a lag between the operation of the Withdrawal timer and the dwell timer) occurs when the pouring rate is too low and the metal level has not reached the level of the thermocouple 51 to cause the withdrawal portion of the cycle to commence after the desired time has elapsed. NV differential motor is idle when the withdrawal portion of the cycle begins at the instant the desired dwell time has elapsed. The output of differential DB drives (through a suitable series of gears) a rheostat in the furnace tilt variable speed motor 62 drive circuit.

A synchronous motor, PM master motor, of the same specifications as the timer motors drives a master rheostat through a suitable series of gear reductions and a clutch GL2. The master rheostat is of a lesser resistance than the Vernier rheostat and so wired in the variable speed drive circuit of motor 62 that a full sweep of the Vernier rheostat will control the vessel tilting rate from zero to a maximum determined by a particular setting of the master rheostat.

The master motor PM operates when NV differential motor operates. As is evident from the wiring diagram, master motor PM runs clockwise when differential motor runs clockwise and counter-clockwise when NV runs counter-clockwise.

Two distinct systems of the automatic pouring control respond to a single signal, in a different manner. The first, utilizing that portion of the circuit which comprises the withdrawal timer W and associated relay coils and contacts, causes the withdrawal mechanism 50 to start in response to the high level signal from themolten metal level indicator circuit, and operate for a length of time, determined by the setting of the withdrawal timer W. The second, utilizing the remainder of the circuit, records the time interval which elapses between the stoppingof the withdrawal mechanism and the next signal from'the molten metal level indicator circuit. If

9 this time interval is equal to a period of time determined by the setting of the dwell timer D, there is no external response from this portion of the circuit. If, however, the recorded time interval is shorter or longer than the period of time determined by the dwell timer, the control responds in such a way :as to cause the interval to again become equal to the desired dwell time. As it is the amount of metal poured into the mold during a given length of time, which determines whether the actual dwell time is shorter than, equal to or longer than the desired dwell time, it is evident that the vessel tilting rate must be adjusted in order to approach and maintain :a relatively constant rate of pour which will fill the mold to the thermocouple hot junction level during a period of time approximately equal to the desired dwell time, plus the withdrawal time. The vessel tilting rate is controlled by the positioning of the vernier and master rheostats in the drive circuit of the motor 62.

The control dampens the full elfect of a response so as to increase or decrease the rate of pour so that the desired rate of pour is approached by :a functional factor, such as by halving, thus avoiding a tendency to overshoot or hunt for the desired tilting rate.

As an aid to the operator, a visual indication of the time in seconds of each dwell period may be provided. The actual dwell time is shown by the lamp that remains glowing at the end of the dwell period. This lamp remains lighted through the succeeding withdrawal period. A one R. P. M. synchronous clock motor drives, through a magnetic clutch, the contact arm of a rotary switch having 30 contacts arranged in a semi-circle, each contact is connected to an indicating lamp. During a dwell period the clock runs and the clutch is engaged to drive the contact arm, to light a lamp for each successive second of time. At the end of the dwell period the clock stops, leaving the last lamp to indicate the length of dwell time. A few micro-seconds before the end of withdrawal a time-delay relay momentarily opens the clutch coil and a spring returns the switch arm to zero, then the clock is again started at beginning of dwell time.

In an explanation of the operation of the vessel pouring control shown on Fig. 7, it is convenient to assume a set of conditions, and to illustrate the operation of the control through a series of operating cycles. Such a sequence of operations is shown in Fig. 8. Let it be assumed that the desired rate of pour to the mold 22 is 8.6 lbs. per second. A mold of a 30.614 sq. in. cross section will be filled to the thermocouple level 55 in 18 seconds at this rate, if the withdrawal portion of the cycle lasts for 10 seconds and the average rate of casting withdrawal is 60" per minute. The remaining 8 seconds of the cycle comprise the dwell. Let it also be assumed that the geometry of the vessel is such (due for example to erosion) that the rate of pour will be close to the desired rate at the beginning, and successively tends to decrease to a value lower than the desired rate, to increase to a value higher than the desired rate, and again level off to the desired rate, over a period of 10 cycles.

Fig. 8 reveals that as long as the rate of pour is 8.6# per second, as is the case in the first cycle, the control follows a pattern of a 10 second withdrawal and an 8 second dwell. It will be noted that both LV differential motor and MV differential motor run in the same direction for the same length of time during the 8 second withdrawal so that the output of differential DA is zero, and, as NV differential motor is idle, the output of differential DB is also zero. Therefore, no change in the setting of the vernier rheostat is elfected.

As the pouring rate decreases during the next cycle from 8.6# per second to 6.45# per second, the length of the actual dwell increases from 8 seconds to 14 seconds. When the desired dwell time of 8 seconds has elapsed, M timer motor, which has been driving its nut toward the'clockwise limit switch PM (see Fig. 7), must stop when it opens this limit. Since MV differential motor is wired in parallel with M timer motor, it also stops when 8 seconds have elapsed. L timer motor, however, continues to run counter-clockwise for the full 14 seconds until the withdrawal timer starts. LV differential motor is wired in parallel with L timer motor, and continues to run in'a clockwise direction for 14 seconds. As LV differential motor continued to run in a clockwise direction for 6 seconds after MV dilferential motor had stopped, there is an output of 5 revolutions (for a given set of gear reductions) from diiferential DA in a direction which will aid in increasing the tilting rate. Since both the dwell timer D and the withdrawal timer W were idle during the last 6 seconds of the dwell, NV ditferential motor ran counter-clockwise during that period of time, for 5 revolutions adding, through differential DB, to the output of differential DA for a total of 5 revolutions (for a given set of gear reductions) from diiferential B output shaft, in a direction which will drive the vernier rheostat (through suitable gear reductions) so as to increase the tilting rate of the vessel, causing an increase in the rate of pour.

When the rate of pour begins to increase during the next cycle, the actual dwell time begins to decrease, i. e. from 14 to 12 seconds. As the actual dwell is still longer than the desired dwell, the final response of the control is still such as to increase the rate of pour. In this cycle (the 3rd) however, the dampening action, which will tend to prevent the control from overshooting the desired rate of pour, begins to take effect. Since the withdrawal timer and the dwell timer were idle during the last 4 seconds of the dwell, NV differential motor ran counter-clockwise for 3 /3 revolutions, i. e. in a direction which will drive the vernier rheostat so as to increase the rate of pour. L timer motor, however, which ran counter-clockwise for 14 seconds during the dwell of the previous cycle, must run clockwise for 14 seconds to open the clockwise limit EM. LV differential motor, in parallel with L timer motor, runs counter-clockwise for 14 seconds. Since M timer motor and MV differential motor run for only 12 seconds, i. e. for the duration of the 3rd dwell, the difference of 2 seconds result in an output of 1% revolutions from differential DA in a direction which will tend to decrease the rate of pour. The total efiect resulting from motor responses during the 3rd cycle, is therefore, an output of 1% revolutions from differential DB in a direction which will increase the rate of pour.

If the chart of Fig. 8 is followed through successive cycles in a manner similar to the foregoing description various control responses for characteristic conditions become evident.

While in accordance with the provisions of the statutes we have illustrated and described herein the best form and mode of operation of the invention now known to us, those skilled in the art will understand that changes may be made in the process without departing from the spirit of the invention covered by our claims, and that certain features of the invention may some times be used to advantage without a corresponding use of other features.

What is claimed is:

1. Continuous casting apparatus comprising an open ended substantially upright fluid cooled mold, pouring means for delivering molten metal to said mold, variable speed motor driven means for controlling the rate of discharge of molten metal from said pouring means, means for regulating the operation of said motor in response to the actual rate of molten metal delivery to said mold, and means separate from said means for controlling the rate of discharge of molten metal from said pouring means for reversing the direction of said variable speed motor means to stop the delivery of molten metal to said mold in response to a rise in the molten metal level in said mold to a selected upper position including molten metal sensing imeanslsensitiveto thepresence .of molten metal at said selected -.upper mold position' 2. Continuous .casting apparatuscomprisingan open ended substantially upright fluid .cooled mold, pouring means for delivering moltenmetalitosaid mold including atiltingtype pouringivessel and an 'adjustably positioned tun dish, motoridriven means for controlling the rate of delivery of molten metal-fromsaid pouring vessel, means for regulatingthe speed in onedirection of said motor in response to the actual-rate of molten metal delivery to said mold, heat sensitive means positioned in said mold operable to detect :the presence'lof'molten metal at an upper level in saidmold, and'separatemeans for substantially simultaneously stopping and reversingthe direction of said .variable speed motor driven means in response to a rise in the molten metal level in said mold to the positionof said heat sensitive means, so as to stop delivery of molten metaltosaid mold.

References iGited-iplthe file ,of .this patent UNITED a I s PATENTS