US3354940A - Continuous casting apparatus with improved nozzle composition - Google Patents

Description

Nov. 28,1967 F, H. WALTHER, JR., E L l3,

CONTINUOUS CASTING APPARATUS WITH IMPROVED NOZZLE COMPOSITION Filed April 16, 1965 FRANK H. WALTHER, JR. By JOSEPH R. RYAN Ill United States Patent 3,354,940 CONTINUOUS CASTING APPARATUS WITH IMPROVED NOZZLE COMPOSITION Frank H. Walther, Jr., Bethel Park, and Joseph R.

Ryan, Irwin, Pa., assignors t0 Harbison-Walker Refractories Company, Pittsburgh, Pa., a corporation of Pennsylvania Filed Apr. 16, 1965, Ser. No. 448,633 2 Claims. (Cl. 164-281) This invention relates to the continuous casting of steel. More particularly, it relates to improvements in tundish and nozzle construction for use in practicing various processes for the continuous casting of steel.

The commercial use of processes for the continous casting of steels seems destined to take an increasingly important position in contemporary steelmaking. Its many advantages in terms of cost, labor, and simplicity of practice make it very attractive to a highly-automated industry.

The concept of continuous casting of steel is not new. As early as 1865, Sir Henry Bessemer envisioned the continous casting of a steel plate or the like by continuously pouring molten metal between two water-cooled rollers to recover a strip of steel.

The recent article of Leonard V. Gallagher et al., in the December 1963 issue of Scientific American, is rec ommended for a detailed study and analysis of the evolution of processes for the continuous casting of steel from the early work of Bessemer to many contemporary process studies and practices.

Two of the more important and critical parts or steps in contemporary processes for the continuous casting of steel are considered to be:

(1) The actual pouring or directing of a carefully regulated stream of molten metal to an initial cooling or freezing stage, and

(2) The actual cooling or In this latter step, about a stream of freezing step itself.

a peripheral skin or shell is formed molten material by solidifying or freezing, as it were, of the'outer periphery or extremities of the stream passing therethrough. This skin or shell formation is extremely critical if one is to prevent later breakout of molten metal subsequent to initial cooling, and to assure the eventual recovery of a billet or the like of fairly uniform dimensions. If the stream of molten metal to the cooling or freezing stage is not constant in volume and cross sectional configuration, the formation of a uniform and sufiiciently strong skin is made difiicult, if not impossible, to accomplish.

In many of the present practices, the semi-solidified stream is subjected to a pulling action by frictional engagement with driven rollers, or by frictional engagement with upwardly and downwardly reciprocating mold walls in the initial cooling stage. Of course, gravity itself can be of substantial eifect in placing stresses on the evolving semi-solidified stream because of the substantially vertical process flow path present practices utilize.

The molten metal, usually after some manner of degassing, as, for example, vacuum degassing, is fed to what is termed a tundish. A nozzle opens through a lower portion of the tundish; for example, it opens through the bottom of the tundish. It is this nozzle which is so critical and important to controlling flow rate and stream cross section to the cooling stage mold. It must be characterized by resistance to skulling. Skulling can be defined as the localized buildup of solidified metal and slag on interior surfaces of the nozzle and about its exit orifice, for example. One manner of resisting such skulling is to provide a nozzle having very low thermal conductivity, whereby heat is not lost from the molten metal through the nozzle body, which heat loss could cause premature metal freezing and, thus, skulling.

As noted above, the cross sectional configuration of the stream passing from the nozzle exit orifice to the cooling mold is important. Skulling about the nozzle orifice obviously leads to variation irl stream volume, as well as cross sectional configuration between the nozzle orifice and the cooling mold.

The nozzle must be substantially inert to the molten metal passing through it. If it is not inert there will be erosion or corrosion of interior nozzle walls and a change in the characteristics of the nozzle orifice, thereby also disadvantageously modifying and changing evolving stream configuration and volume.

Still further, the nozzle must be chracterized by substantial freedom from expansion or shrinkage when subjected to heat, or at least such volume change must be so precisely known that one can predict the volume and configuration of the stream issuing from the nozzle after it has been put on line. And even in this case, this predictable expansion or contraction cannot be a long-term thing; for example, if the constituents of which the noozle is constructed, over a period of many hours, days, etc., slowly goes through a shrinkage or expansion, this is not considered acceptable because the system would take so long to reach a steady state.

It is, therefore, an object of the present invention to provide improvements in processes for the continuous casting of steel in which predictable stream volume and cross sectional configuration are obtained and maintained. It is a further object of the invention to provide an improvement in apparatus for the continuous casting of steel, whereby the volume and cross sectional configuration of a stream passing from the tundish nozzle to the cooling mold is more efficiently controlled.

It is yet another object of the invention to provide a nozzle particularly suited to the continuous casting of steel.

According to this invention, a conventional tundish is provided with a ceramically bonded refractory nozzle of special chemistry. According to a preferred embodiment, the nozzle is mineralogically characterized substantially entirely by the phase MgO-Cr O Such a nozzle is made by a process as follows: Substantially stoichiometric quantities of caustic calcined (i.e., chemically active) magnesia substantially free of free water and water of crystallization and of very high purity, and very high purity chrome sesquioxide, are combined. Both the chrome sesquioxide (Cr O and the caustic magnesia are very finely divided, i.e., substantially all (preferably over by weight) of the materials are 200 mesh Tyler. Preferably, the material is entirely 325 mesh Tyler (44 microns and less). We have said that these powder materials must be very pure. By this, we mean they must be of such purity that the total quantity of ingredients or constituents other than MgO or Cr O does not exceed about 5% of the total mixture and, preferably, such adulterating constituents amount to less than 2%, by weight.

The preferred weight combination of the two ingredients is 21% caustic magnesia and 79% Cr2O3. There can be some variance from this precise stoichiometric ratio. However, the variance must be such that in the final nozzle there is at least 80% of the phase MgO C130 The other 20%, in the final nozzle, can be a combination of unreacted caustic magnesia and Cr O can consist entirely of either of these materials, and, of course, can include varying proportions of these materials between 0 and parts, by weight, of the excess 20%.

In the preferred embodiment, the very finely divided powders are intimately admixed and are subjected to an initial reaction treatment. This reaction treatment con- Patented Nov. 28, 1967 i sists of placing the powders, for example, in a ceramic crucible or the like, and then subjecting the powder mixture to a temperature of about 2000 F. for on the order of 8 hours. This temperature-time relationship, however, must be of such character as to obtain a product which, by X-ray diffraction techniques or the like, shows at least 80%, by weight, of the phase MgO-Cr O Still further, the temperature must be sufliciently low as to substantially exclude ceramic bonding or sintering, or agglomeration of the constitutent powders and the reaction product thereof. Using our preferred heat treatment parameters of 2000 F. for 3 hours, with all the caustic magnesia and Cr O being 200 mesh, we recovered a reaction product of substantially the same sizing as the powders which had been initially subjected to this treatment, i.e., all was 200 mesh, and there was no discernible sintering or agglomeration-at least none that could be seen with the naked eye.

The powder reaction product is recovered and subjected to a briquetting treatment as, for example, passing through the well known Komarek-Greaves briquetting machine under a pressure of about 5000 p.s.i. Considerably higher pressure can be used; for example, about 20,- 000 p.s.i., if so desired. We prefer to press at about 20,000 p.s.i. Higher pressures do not seem particularly necessary The briquettes are subjected to a hard fire; for example, at least 3000" F. and, preferably, at least 3500 F., for about hours. This time-temperature relationship is also variable. However, the hard-fired product must be characterized by ceramic bonding throughout, to such extent as to provide a source of refractory grain.

The hard-fired briquettes are crushed to recover a grain material. The grain material is fabricated into a batch having approximately the following size gradation: 40-60% +65 mesh, the remainder 65 mesh with from 40-60% of the 65 mesh material being retained on a 325 mesh screen. Since we prefer to form shapes by casting techniques, we suggest that the coarsest grain pass a 4 mesh screen. Of course, somewhat larger particles, for example /2 inch, can be included, although We do not suggest this for best results.

In any event, the briquettes are ground to provide a size graded batch of grain material which is then tempered with a suitable agent; for example, water. The water content is variable. We suggest between 6 and 10%, by weight, based on the weight of the grain. Larger and lesser amounts of water can be used, if desired. However, the more water one uses, the longer it takes to cure and dry the resulting nozzle shape before firing; and with lesser amounts of water, one is very apt to encounter difficulty in casting due to insufficient lubrication or fluidity. Airramming techniques could be used for forming, as could power pressing techniques on a modified brick press. Casting, though, is our preferred suggestion. Still further, we can use other than water as the tempering agent. Such agents are well known in the art.

The batch, as above described, is cast into the shape of a nozzle. The cast nozzle preferably is fired according to a conventional refractory brick-burning schedule; for example, 2900 F. for on the order of 10 hours. However, the selected temperature for this firing is variable. It is required, however, that the selected temperature-time relationship be such as to obtain ceramic bonding between the grain particles which constitute the nozzle.

It is essential that the raw materials which are used to form the nozzle just discussed and, of course, the resulting nozzle itself, must be free of vitrifiable materials, i.e., glass formers. Such terminology is conventional in the refractories art, although it can be misleading. Vitrifiable or glass-forming materials are those which melt at a relatively low temperature to form what can be termed an amorphous phase. For the purpose of this invention, when we say that the material is free of vitrifiable materials or glass formers, we mean to describe freedom from constituents which will melt and form liquids when subjected to a temperature of about 2000 F. When we say substantially free, we mean that they can constitute no more than 5%, by weight, of the refractory grain and, as noted above, preferably they constitute no more than 2%, by weight, of the grain.

One reason for the criticality of this requirement to exclude glass formers is that the nozzle, when placed in the continuous casting process, usually is subjected to temperatures in excess of 2000 F., and glassy melt formation can lead to changes in nozzle orifice configuration and, thus, volume in cross sectional configuration of an issuing stream as has been discussed in some detail above. Not only that, if such ingredients form in relatively continuous zones or films or lamina across and through the wall of the nozzle, thermal conductivity is undesirably increased, thus leading to skulling with its attendant disadvantages, as also has been discussed above.

A nozzle made according to our invention, as above discussed, is characterized by low thermal expansion and thermal conductivity as compared to more conventional magnesia-chrome and high alumina systems, and is inert to most all conventional steels and their slags of which we are aware.

If less than about of the nozzle is other than the phase MgO-Cr O there is undesirable reaction between constituents; for example, MgO and Cr O in nozzle service, which reaction is an expanding one which, of course, can lead to undesirable change in dimensions.

The preferred embodiment and the best mode now known to us for the practice of our invention is as follows: 79%, by weight, of 325 mesh chrome sesquioxide as, for example, a chemically precipitated chromic oxide, is mixed with 21 parts, by weight, of caustic magnesia of like sizing. The Cr O or chrome sesquioxide is at least 99% Sr O by weight, on an oxide basis. The magnesia has approximately the following analysis, by weight on an oxide basis:

Percent Silica (SiO 0.7 Alumina (A1 0 0.3 Iron Oxide (Fe O 0.3 Lime (CaO) 0.9 Magnesia (MgO) 97.5 Loss on ignition 0.1

The magnesia has been calcined at a temperature of about 1700 F. for a time period sufficient to remove all crystalline water and substantially all free water and carbon dioxide. When we say substantially all of these latter materials have been removed, we mean there is an ignition loss of no more than 1% for the material.

The ingredients, as just mentioned, are intimately admixed, placed in a ceramic crucible, and fired at 2000 F. for 8 hours. The reaction product, which is recovered, is pelletized or briquetted on a Komarek-Greaves briquetting machine at about 20,000 p.s.i. The resulting briquettes are hard-fired at a temperature of 3500 F. The hardfired briquettes are crushed to form a size graded batch of substantially the following overall screen analysis (by weight): -4+10 meshabout 10%, 10+65 meshabout 30%, 65 mesh-about 60%. 55% of the 65 mesh material passes a 325 mesh screen.

The size graded refractory material is tempered with about 7 parts, by weight, of water, based on the weight of the refractory. The tempered batch is cast into mois-' ture-absorbent plaster molds. After setting, nozzle shapes are recovered from the molds and dried for 12 hours at about 240 F. The dried shapes are fired at 2900" F. for 10* hours.

The schematic diagram which forms the single sheet of drawings in this case better illustrates the apparatus concepts involved in our invention. A ladle of steel 10 is placed above a tundish 11 in such position as to continuously discharge a stream of molten metal 12 to the tundish, and at such a rate as to maintain a substantially even bath depth in the tundish. A nozzle 13, according to this invention, opens from the bottom of the tundish, and is so positioned that its bottom orifice discharges a stream 14 of substantially constant volume and cross sectional dimension into the mold cavity 15 wherein a shaping and skin formation is caused to form the steel strip 16.

The strip 16 passes downwardly between rollers 17 until it eventually is entirely solidified, at which point it is cut into lengths, as at 18, whereat we have schematically indicated an oxyacetylene burner 19 cutting off lengths of steel 20 in the container 21. The container 21 is pivoted about pin 22. As soon as a length of steel 20 has been cut, the container rotates-as shown in dotted lines-to a substantially horizontal position and the cut billet or length of steel 20' is discharged onto, for example, a conveyor system for movement to additional treatment stages, storage, etc. (not shown.)

The tundish 11 very satisfactorily can be fabricated as by casting of a composition identical to that from which the nozzle 13 is made. However, it could be made from other constituents which are chemically compatible with the nozzle; for example, dead burned periclase. It must also have compatible thermal expansion properties.

Of course, dimensions for the tundish and nozzle vary, depending upon the installation in which they are used. The shape of the nozzle is also variable, depending upon the desires of a user. It can be generally inverted bellshaped with an orifice opening through the smaller end of the bell. It can be a truncated, conical shape with the exit orifice formed through the smaller end of the cone.

Having thus described the invention in detail and with sufficient particularity as to enable those skilled in the art to practice it, what is desired to have protected by Letters Patent is set forth in the following claims.

We claim:

1. The combination with apparatus for the continuous casting of steel, including a container of molten metal positioned above a tundish, which tundish includes a nozzle opening from a lower portion thereof with its exit orifice positioned above a cooling mold, the improvement comprising said nozzle being ceramically bonded and characterized in that at least by weight, of its structure is the phase MgO-Cr O and being substantially free of materials which form glass at temperatures of about 2000 F.

2. A nozzle for use in the continuous casting of steel consisting essentially of a tubular nozzle having a relatively large open top and converging to an exit orifice at its bottom, said nozzle being made from a batch of grain characterized in that at least 80% thereof is the phase MgO-C1 O and is substantially free from materials which form glass at about 2000 F.

References Cited UNITED STATES PATENTS 3,094,424 6/ 196-3 Ratclilfe 2285 3,116,156 12/1963 Charvat 10659 3,132,954 5/1964 Alper et a1. l0659 3,284,217 11/1966 Walther 106-59 FOREIGN PATENTS 724,980 2/ 1955 Great Britain.

J. SPENCER OVERHOLSER, Primary Examiner. R. D. BALDWIN, Assistant Examiner.

Claims (1)

1. THE COMBINATION WITH APPARATUS FOR THE CONTINOUS CASTING OF STEEL, INCLUDING A CONTAINER OF MOLTEN METAL POSITIONED ABOVE A TUNDISH, WHICH TUNDISH INCLUDES A NOZZLE OPENING FROM A LOWER PORTION THEROF WITH ITS EXIT ORIFICE POSITIONED ABOVE A COOLING MOLD, THE IMPROVEMENT COMPRISING SAID NOZZLE BEING CERAMICALLY BONDED AND CHARACTERIZED IN THAT AT LEAST 80%, BY WEIGHT, OF ITS STRUCTURE IS THE PHASE MGO$CR2O3 AND BEING SUBSTANTIALLY FREE OF MATERIALS WHICH FORM GLASS AT TEMPERATURES OF ABOUT 2000*F.

Images